From the Streets to the Chamber: Social Movements and the Mining Ban in El Salvador

Following an extended anti-mining campaign, El Salvador became the first country to adopt a legal ban on all forms of metallic mining. This article uses process tracing to map direct, indirect and mediated linkages between the anti-mining mobilization and the formal adoption of a mining prohibition by the national legislature in 2017. It draws on 78 interviews with campaign activists, legislators, government officials, business leaders and legal teams, and combines this information with legislative documents and reports, public opinion data, legal documents from an investment dispute filed against the Salvadoran government, and blogs and website of the Mesa Nacional Frente a la Minería Metálica. This analysis gives particular attention to the spatial reach and breadth of the anti-mining networks; fissures within and situational realignment of the political elite; and the strategic use of diverse institutional openings (docking points), some of which were adapted to new purposes by movement entrepreneurs. Although major obstacles to sustainable development and environmental protection remain in El Salvador, this article identifies a set of iterative interactions between activist alliances and institutional actors that can successfully contribute to policy change.


Introduction
In 2017, El Salvador became the first country to adopt a national ban on all forms of metal mining.This remarkable policy development came after twelve years of antimining mobilization.This study highlights the distinctive features of that campaign and analyses the ways in which social mobilization connected with formal institutions and official policymaking.As such, it brings together two fields of study that have often been separated along disciplinary lines -the study of informal, non-institutional, "street-level" politics of social movements, and the study of formal, institutional politics and policymaking by government officials.
A large body of research has emerged to analyse social movement origins, alliance structures and tactics.The connections between movements and their policy consequences (or lack thereof), however, are less well documented (Bosi, Giugni & Uba, 2016;Silva, 2017).Social movement research generally finds that mobilizations have identifiable political goals, and that they are often at least partially successful in achieving their objectives (Amenta, Caren, Chiarello & Su, 2010;Biggs & Andrews, 2015;McAdam & Boudet, 2012).How and why they have an impact on policy remains a matter of dispute.Given the significant role of social movements as an expression of collective demands, and the frequency with which people employ protests in search of substantive change, the circumstances under which popular mobilization becomes politically consequential should be better understood.Movement theorists and public policy specialists are now attempting to model the sequences and interactive effects that animate the movement-policy nexus, and more accurately map possible causal mechanisms.
Documenting the connections between movements and policy is, however, an inherently difficult task.The challenges are in part due to movement volatility and the wide array of processes and timelines involved (Bosi, Giugni & Uba, 2016;McAdam & Boudet, 2012).As Silva (2017, 6) notes, "the very characteristics that draw us to social movements -their passion, fluidity, malleability, precariousness, creativity, and contingent nature -make it hard to establish their connection to observed changes with high degrees of confidence."Solid research on the movement-policy link requires careful process tracing, a labour-intensive methodology that tends to limit the scope of the analysis to a small number of cases, which can blunt the contribution to theory building and rigorous testing.This research also requires a commitment to studying movement dynamics and policy processes over time, so that interactions can be monitored across various phases and any long-term consequences can be identified.Analysis across time is particularly critical in megaproject studies, where economic incentives in favour of project development are persistent and achievements in one campaign may give way to defeat in another.
Social movement analysts and public policy specialists have begun to develop nuanced conceptual tools and frameworks, introducing concepts like direct, indirect, mediated and joint effects, as they track the ways that movements interact with other political processes (Giugni, 2004;Amenta, Caren, Chiarello & Su, 2010).In his ambitious study of ecology, antinuclear and peace movements in the United States, Switzerland and Italy between 1975and 1999, Marco Giugni (2004) found that protest alone had "at best a marginal to moderate effect" on policy, at least as defined in terms of adoption of legislation and increases/decreases in spending (p.70).Instead of "direct effects," in which policy outcomes are determined exclusively by the demands of activists and protesters, change was more likely to depend on a "mediated" process that operated through a "joint effects" model.In Giugni's formulation, joint effects referred to the combined impact of movement mobilization (typically in the form of increased protest), alliances with strategic political elites in the institutional arena, and shifts in public opinion demonstrating increased support for movement goals.
Other analysts have also called attention to this interactive effect.Reviewing the literature on the political consequences of movements, Amenta, Caren, Chiarello & Su (2010) found movement success aligned with organizational size, multisectoralism, and diverse and complex leadership structures; with the construction of "resonant prognostic and diagnostic frames" which identified problems clearly and presented viable solutions; and with the use of political and electoral strategies to gain penetration and leverage, including activist insertion into formal institutional roles (pp. 296-97).To trace the connection between movement activities and policy results, these analysts employed a "political mediation model," which highlighted the importance of "engaging in collective action that changes the calculations of institutional political actors, such as elected officials and state bureaucrats, and adopting organizational forms and strategic action that fit political circumstances" (p.298).
Although research on the interconnections between movements and policy is growing, much of it has produced only tentative results.A small number of classic cases in U.S. and Europe have drawn disproportionate attention, making it difficult to know how well the findings apply to less celebrated cases or to dynamics in other regions, including Latin America.Research Group Mega (RGM) has taken on the challenge of exploring the movement-policy nexus in Latin America "to contextualize social movement pressure on polices and institutions in a broader political economy framework, taking into account how actors with various interests and power resources form contending coalitions around policies and institutions relating to mega-development projects" (Silva, Akchurin & Bebbington, 2018).Megaprojects in this region typically advance with state support under regulations designed with limited local consultation and with inadequate attention to environmental, social and human rights consequences (De Castro, Hogenboom & Baud, 2016;Bebbington & Bury, 2013).Social movements sparked by megaprojects in Latin America struggle to break through these constraints and gain access to the policymaking process.
The present study provides a longitudinal analysis of megaproject mobilization and policy change by tracing the arc of transition in El Salvador's metal mining policy.It plots the path from policies that opened up mining investment to legislation that formally closed the sector down.In line with other research in this field, my analysis gives close attention to processes of political mediation that connect movement activists with state actors and formal institutions.I use the concept of punctuated equilibrium to describe rhythm of policy pulsation, in which the "long periods of stability that students of public policy have noted to be characteristic of most public policies most of the time" combine with an "occasional burst of policy innovation" (Baumgartner 2006, 25-26).
This work draws on responses provided in 78 semi-structured interviews conducted in El Salvador in 2010, 2012 and 2017.Interviewees included anti-mining activists and movement organizers; legislators, party leaders and administrative officials; business leaders and legal representatives; and research analysts involved with the Salvadoran mining conflict. 1This study also incorporates documents from websites and blogs of the Mesa Nacional Frente a la Minería Metálica and its affiliates; recorded testimony and legal briefs from the investment dispute between Pac Rim Cayman LLC and the Republic of El Salvador at the World Bank's International Centre for the Settlement of Investment Disputes (ICSID); public opinion data from the Instituto Universitario de Opinión Pública (IUDOP) at the Universidad Centroamericana (UCA) in El Salvador and from CID-Gallup; and reports on legislative processes posted on the website for El Salvador's Asamblea Legislativa. 2 This study points to five observations: 1. Analysis of the movement-policy nexus requires a long timeline, extending beyond a particular campaign or legislative session to capture pauses, advances, reversals and strategic adaptations that emerge over time.Such slow movement studies allow scholars to trace the ways in which movements both succeed and fail in their attempt to affect political processes, and to better understand the circumstances under which movements have consequences in terms of policy change and institutional adaptation.2. The ability of social movements to influence public policy may depend on their spatial reach (ability to connect across local and national scales), and breadth (ability to link across sectors).Movement networks can disseminate information through interconnected clusters built on common identities and shared experiences (bonding).They can also extend connective tissue by linking sectors across social and political divides (bridging).Bridging processes may be particularly important if durable policy change requires the construction of alliances that transcend partisan polarization.3. Movements are more likely have a direct effect on the policy process in the agenda-setting phase of policy development.Movement actors can articulate grievances, raise awareness, and frame possible solutions, thereby opening the policy arena to new ideas and debates.4. Movement influence on advanced phases of the policy process, including policy formulation, adoption and implementation, may be indirect, serving to spark official deliberations that then proceed without further movement involvement, or mediated, involving iterative interactions between activist alliances and institutional actors who preside over formal decision-making.5. Crossing the boundary between mobilization and policy networking requires the identification of docking points, or spaces where activists can insert public demands into the formal mechanisms of policy revision.In formally democratic systems, institutional engagement may be facilitated by mobilization of supportive public opinion.
The paper begins with an overview of mining policy in post-war El Salvador and describes the early advance of the industry.It then turns to the emergence of an opposition movement, and examines the bridging and bonding linkages that shaped movement expansion.The unfolding dynamics of the policy debate are traced by examining mining bills introduced in the Salvadoran legislature.The processes that culminated in the official mining ban are described.Finally, the criteria suggested by Amenta (2006) to assess the policy consequences of social movements are reviewed, drawing on the mediated effects model to summarize the findings.

Political transitions and market openings
After a decade of civil war, El Salvador's post-war political settlement left the newly created Alianza Republicana Nacionalista (ARENA) party in power. 3The 1992 peace agreement gave the demobilized Frente Farabundo Martí para la Liberación Nacional (FMLN) legal status and the opportunity to compete for power, but ARENA consistently won the presidency and, with allied parties, controlled the main institutions of government for two decades.Its policies favoured the interests of a transnationalized sector of the economic elite, which had acquired a strong position in banking and finance during the post-war transition (Segovia, 2002).Policymaking relied heavily on the planning proposals of FUSADES, a USAID-funded think tank that helped to staff the technical ministries (Spalding, 2014).
Achieving four consecutive presidential terms, ARENA leaders had the opportunity to oversee the neoliberal transformation of El Salvador's economic model.Reforms included privatization of the banking and telecommunications systems, trade liberalization, expansion of the export processing zones, deregulation and, in 2001, a controversial policy of dollarization.To promote foreign investment, ARENA presidents signed investment laws, bilateral investment agreements, and free trade agreements that allowed foreign investors to bring cases to international tribunals in the event of an investment dispute.
As elsewhere in the region, El Salvador revised its mining law during this period of neoliberal reform.New legislation, adopted in December 1995under President Armando Calderón Sol (1994-1999), incentivized investment in this largely undeveloped sector by providing additional legal protections and tax exemptions.Exploration by junior mining companies progressed in the early 2000s under Francisco Flores (1999)(2000)(2001)(2002)(2003)(2004), when the government adopted a yet more investor-friendly amendment to the mining law, reducing the royalties from 4 per cent to 2 per cent and increasing the size limit on concessions (Henríquez 2008, 28;TAU 2011, 33).The new rules attracted investor attention; twentynine companies held exploration licenses in 2006 (TAU 2011, 39).
The most advanced project in 2004 was El Dorado mine, where the Canadian mining company, Pacific Rim, had acquired an existing concession through merger in 2002.The clock was ticking for the expiration of the concession when Pacific Rim acquired it, and the new management moved quickly to seek an extraction permit in December 2004.Pac Rim proposed a subterranean mine in Cabañas, one of the country's poorest departments.This project envisioned a 10year production process, with two years to start up, six years of operation, and two years to close down.The projected value of gold exports from Cabañas mines topped US$1 billion, with a formal sector workforce of 2,818 in six proposed mines (TAU 2011, 41-42, 45), making this envisioned mining complex one of the largest (albeit short lived) enterprises in the country.According to an economic projection commissioned by Pacific Rim with former Finance Minister Manuel Hinds (2007), the El Dorado mine would produce an 8.4 per cent increase in the GDP of Cabañas, leading to a 23 per cent drop in the department's extreme poverty levels.

Constructing a national opposition movement
Launched formally in 2005, El Salvador's anti-mining movement forged crossspatial linkages between local and national-level associations and built crosssectoral connections of both "bonding" and "bridging" varieties (Putnam, 2000).This array of ties helped the movement to expand its reach and diversify its strategies, and provided the flexibility needed to adapt to shifting opportunities.Community-level allies brought local knowledge and urgency to the process, as residents mobilized to prevent imminent mining encroachment.National-level allies, although further removed from the mining zones, provided physical proximity to decision-making centres and greater familiarity with policymaking processes, which allowed resistance activists to make strategic choices about alliances, timing, and funding.Bonding networks drew on relationships among similar actors, and were infused with mutual understanding and trust.Bridging, by contrast, connected groups across wider cultural and power divides, which permitted valuable resource transfers that were sometimes accompanied by tension and mistrust.
These networking strategies parallel the distinction between "close" and "weak" ties that has generated substantial academic discussion in the study of social capital (Granovetter, 1982;Marsden & Campbell, 2012;Biggs & Andrews, 2015).Networks built on close ties, in which participants have frequent, routine, and often organic (based on family, friendship and neighbourhood) connections, are imbued with familiarity and permit a ready flow of interaction and communication.These mechanisms produce closed-circuit patterns of information diffusion that tend to reinforce message content through repetition and trust-based validation.Weak ties, in contrast, link actors beyond their primary groups and provide more distinctive and less predictable flows of information.As the literature on the strength of weak ties indicates, these connections allow networks to extend beyond the clusters of redundant communication that characterize close-tie alliances, and kindle new linkages that can both broaden and accelerate the diffusion of mobilizing information.The cultivation of weak ties can also facilitate multisectoralism and permit greater functional specialization in keeping with the variation in resources that broadened networks can mobilize.These differences in capacity encourage repertoire diversification by connecting those, for example, who have different kinds of expertise with regard to lobbying, media access, legal strategies, and political party ties.
In terms of social movement impacts on public policy, close ties built on bonding techniques can stoke attitudinal uniformity and inspire recurring mobilization, including passion-driven protest and electoral engagement, by providing emotional support and validation.Weak ties built on bridging principles, in contrast, can extend the network beyond its historical boundaries and encourage broader awareness of issues, promoting fuller shifts in public opinion.Although the latter increase the risk of internal tensions and network fragmentation, they may also provide the leverage needed to affect a broader swath of popular views.This strategy, however, requires brokers who can link previously disconnected movement segments and build bridges that span the gaps.

Local mobilization and bonding ties
"Vulnerability theory" contends that people who are most at risk due to a proximate threat are most likely to find it troubling (Eisenstadt & West, 2017).In a recent study of the intersection between vulnerability, extraction, and public opinion, Eisenstadt and West drew on this theory to predict which sectors in Ecuadorian society would express the greatest level of concern about environmental conditions.Their survey results and geocoding of respondent location confirmed that concern was greatest among those facing objective vulnerability to harm (experiencing water scarcity and working in ecotourism, for example), and those residing within 30 kilometres of an active mine.This link between physical proximity to mining, the perception of threat, and environmental concern suggests that residents of communities where mining projects are planned may develop heightened concern about detrimental consequences.
Grievances alone, however, are not normally sufficient to trigger protest.Political process theory highlights the impact of opportunities, emphasizing the significance of a supportive political environment, which permits open organizing and attenuates risk (Tarrow, 2011).Prior experience with mobilization has been associated with protest recurrence through the creation of "strategic capital," which Almeida (2014, 25) defines as "repositories of collective knowledge about which strategic and organizational elements in their localities are best employed to generate joint action."Eckstein (2001, 33-37) has identified "popular cultures of resistance" as a factor contributing to bouts of defiance and protest in Latin America.Previous mobilizing can generate both enduring organizational forms and "abeyance structures" (Taylor, 1989) that fade during periods of inactivity but spring quickly to life when new opportunities arise.
In the Salvadoran case, the history of community mobilization in the prospective mining region during the civil war and post-war transition left an organizational trace on which new rounds of resistance could build.The Salvadoran "gold belt" included areas that had experienced active combat during the 1980s civil war.Several communities in concessioned territory had endured expulsion and flight following massacres and military campaigns only a little more than a decade before.Community residents underwent lengthy relocation in refugee camps, particularly in nearby Honduras, where Mesa Grande served as a temporary resettlement camp (Smith-Nonini, 2010).The experience of desperate flight followed by life in the camps left the trace of collective trauma, as victims attempted to cope with the magnitude of their losses.It also left an organizational foundation, as this population constructed various forms of self-governance while displaced.Although formally separated from the military conflict, some refugees had family or community ties to the FMLN and maintained contact with combatants, who had an unofficial presence in the camps (Viterna, 2015, 94).
Following years of displacement in the 1980s, many Salvadoran refugees returned to the villages in the northern region of El Salvador from which they had fled (ES2417; Smith-Nonini, 2010).They formed local and regional organizations like the Coordinadora de Comunidades de Refugiados y Repobladores (CCR, later Asociación de Comunidades para el Desarrollo de Chalatenango) and the Asociación de Desarrollo Económico (ADES) in the repopulated community Santa Marta.These organizations in turn combined forces under the national umbrella of the Asociación de Comunidades Rurales para el Desarrollo de El Salvador (CRIPDES), which claimed to represent 100,000 people in 398 communities. 4Many in the repopulated communities maintained strong ties to the FMLN, which, following its post-war transition into a political party, tended to win local elections there by a large margin.
This was the setting into which the mining companies advanced.Although some people in the gold belt were won over by the prospect of mining jobs or the general economic growth promised by company representatives, others viewed mining investors as intruders who threatened their communities with renewed displacement and political annihilation.Chalatenango activists tell of residents in the municipios of Las Flores, Guarjila, Arcatao and Nueva Trinidad who repeatedly disrupted the activities of the company advance teams, closely monitoring their comings and goings, and surreptitiously removing survey markers planted in their fields.As exploration advanced, activists took more forceful steps, blocking company vehicles that attempted to enter their zone, tracking down drilling sites, and threatening to burn equipment if it was not hauled away. 5 In communities where the mining opposition was weaker, the movement focused on identifying allies and organizing informational sessions.Local leaders learned, according to one organizer, to "hablar sin hablar," that is, communicate without making explicit statements that would invite rejection and retaliation.Their protest style emphasized artistic and cultural activities, including a "Caminata Verde en Defensa del Agua," for example, instead of a marcha or protesta, terms which local people traditionally identified with leftist organizations (ES0417).

Spatial alliances and local-national connections
As community opposition deepened, wartime collaboration and post-war mobilization left organizational tracks on which to transit across the local-national divide.Wartime collaboration established links between local and national level activists and left trust bonds forged through shared experience.Post-war mobilization in opposition to ARENA-backed market reforms, including health care and water privatization (Almeida, 2014, 70-78), also connected local and national leaders.CRIPDES affiliates from the former war zone repeatedly journeyed to the capital to join in national-level protests.Prior experiences with organizing permitted the rapid identification of potential allies whose usefulness and credibility as partners could be reliably assessed.
As the prospect of mine development became imminent, community leaders in the emerging antimining front reached out to "trustworthy" organizations at the national and international levels for assistance in interpreting and framing the evolving mining debate. 6In 2005, ten organizations came together to form the Mesa Nacional Frente a la Minería Metálica. 7This coalition, which combined grassroots community organizations from the mining zone with nationallevel NGO allies, became the backbone of El Salvador's anti-mining mobilization.In addition to CRIPDES, national network partners included the Unidad Ecológica Salvadoreña (UNES), a national environmental organization; Fundación de Estudios para la Aplicación del Derecho (FESPAD), a human rights organization; and the Centro de Investigación sobre Inversión y Comercio (CEICOM), an activist-oriented research centre, among others.(See figure 1).Between the community organizations operating at the local level in territory targeted for mining concessions and their carefully curated national allies with specializations in environmental rights and legal protections, El Salvador's antimining front achieved notable spatial and multisectoral reach.
Mesa activists also collaborated with an expanding network of transnational allies, who provided additional resources and support (Democracy Centre, 2017).These allies included similarly situated organizations in neighbouring Central American countries, where mining development was already underway.Other allies hailed from U.S. solidarity organizations, such as the U.S.-El Salvador Sister Cities network, whose work in El Salvador dated back to the civil war.Their long-term presence and community-level work permitted movement "domestication" and promoted horizontal networking structures (Spalding, 2013).Assistance also came from international non-governmental organizations (IN-GOs) that operated in global power centres and provided thematic specialization and technical expertise.These allies included networks like MiningWatch Canada, which documented the growing number of mining conflicts around the world, and legal rights organizations like the Centre for International Environmental Law, that provided strategic support for communities engaged in international investment disputes. 8This international network took formal shape as the "International Allies" in 2010, and provided funding that allowed the Mesa to hire a full time staff.Organizational meetings of Mesa members were scheduled every two weeks (ES2017), and affiliates cross-posted messages on social media sites, where these organizations had a robust presence.
Mesa leaders report working hard to frame a consistent story about who they were and what they wanted.They adopted a razor sharp focus on a mining ban as their principal policy objective, a position they adhered to with notable tenacity.They orchestrated a long and growing list of actions that included week-long events carried out simultaneously in several cities; hundreds of people marching from the communities to demonstrations in front of central government buildings in San Salvador; delegations to visit mining communities in neighbouring countries; petition and letter-writing drives; mock internment of mining companies; and local festivals in celebration of life and water.Their action frame crystallized around water rights, offering a resonant message that connected communities across the country.This message focused on the Río Lempa, which was "the most important freshwater source in El Salvador, providing 50 per cent of the drinking water, 50 per cent of hydroelectric energy, and irrigation for agriculture in more than half of the country" (McKinley, 2018).Concern about the Río Lempa, repeatedly described as the "aortic artery of the country," seeded both local and national alarm.The impact of mining on water access and quality in multiple regions of the country connected people across the local and national scales, expanding network spatial reach.

Bonding and bridging linkages to Catholic Church allies
Diverse church actors became recurring and critical partners, with Catholic leaders and activists operating within the Mesa and beyond it.These linkages contributed to both spatial reach, drawing on connections between the national Church hierarchy and local parishes, and breadth, as new sectors were incorporated into the network.These connections built on both the bonding ties of familiar long-term collaborators and bridging ties between autonomous actors with infrequent and tenuous links.
Close ties with the Mesa emerged initially through the Catholic Church's social action wing, Cáritas El Salvador, which was a founding member of the Mesa.Cáritas provided a physical office space for the Mesa's coordination activities during the first two years of its operation (ES0817).Justicia, Paz e Integridad de la Creación (JPIC), a Franciscan-sponsored organization, joined the Mesa soon after its founding, and its representatives became featured contributors and emissaries at national and international events (ES0112).These activist church agents had previously-established trust bonds with Mesa activists based on prior collaboration, and they played a critical role as brokers, linking the antimining network with more establishment-oriented church leaders in a classical bridging alliance (See figure 1).
The movement achieved greater breadth when Catholic ecclesiastical leaders aligned with the campaign, following their own institutional discourse and processes.Although less consistently engaged with the mining issue and more politically and socially removed from the Mesa core, the upper hierarchy of the Salvadoran Catholic Church entered the fray publicly in 2007 as the Latin American Catholic Church leaders more generally turned to embrace an environmental message.At the urging of the Archbishop of San Salvador, Fernando Sáenz Lacalle, 9 the Conferencia Episcopal de El Salvador (CEDES), composed of the country's eleven bishops, issued a joint statement calling on the government to deny permits for metal mining (CEDES, 2007).Declaring "this class of mining causes irreversible damage to the environment and surrounding communities," the bishops concluded that metal mining in El Salvador "should not be permitted." As mining policy reform continued to stall, CEDES repeated this call three years later under the incoming Archbishop of San Salvador, José Luis Escobar Alas (CEDES, 2010).Public pronouncements by these widely respected elite actors helped to energize and expand the movement beyond well-worked circuits on the left. 10Although some individual parish priests still preached in support of mine development (Cartagena, 2009, 244;ES1517), national church leaders maintained an oppositional stance, facing down weekly pro-mining protests outside the Cathedral where the Archbishop offered the Sunday mass (ES1710; ES3510).

Conflict and the electoral arena
As the anti-mining network expanded, activists identified docking points through which they could connect with policymakers and press them to include this issue on the policy agenda.Through iterative engagement with institutional actors, they helped to open space for fuller discussion in which new questions could emerge.These mediated effects ultimately contributed to the punctuated equilibrium dynamic, in which a durable policy suddenly experiences sharp and sweeping rejection and is replaced by an alternative approach that garners broad support.
As protests sponsored by the Mesa and/or the FMLN raised the profile of this conflict, and additional organizations aligned with and expanded this network, the electoral risks associated with continued mine promotion began to increase.After almost a decade of collaboration with would-be gold mining investors, the ARENA government slowly backed away from mine promotion.In the lead up to the 2006 legislative election, Pac Rim officials who once reported an array of accommodations now found permits delayed and paper work returned for clarification.Gaps concerning the failure to acquire surface property rights were found in their extraction application, and corporate assurances about cyanide remediation and water safety no longer bought official acquiescence.In the space that opened with application deceleration, officials acquired additional information about mining risks.After Minister of Environment and Natural Resources (MARN), Hugo Barrera (2006), expressed personal concerns about mining, permitting processes ground to a halt.
The development of a well-organized anti-mining movement, combined with a growing electoral threat to ARENA dominance, encouraged a re-calculation of the political costs of mining advocacy.The business subsector that had initially supported mine development failed to lock down broad elite backing.With no strong business lobby anchoring the pro-mining side, the fourth ARENA president, "Tony" Saca shifted toward the opposition.Facing the country's first truly competitive post-war presidential election, with popular journalist Mauricio Funes now running as the FMLN's standard-bearer, Saca moved official opposition out into the open in March 2008, proclaiming that "in principle, I do not agree with granting these permits" (as quoted in Pac Rim Cayman, 2009, 30).The following year, only weeks before the March 2009 election, he called in to a live Catholic radio station programme on mining, and pledged that he would "not grant a single permit" (López Piche, 2009, 2).In defiance of Pacific Rim's threat to file an investment dispute claim, Saca continued, "They are about to file an international claim and I want to make this clear: I would prefer to pay the $90 million than to give them a permit" (p.2).
FMLN candidate Funes, along with his vice presidential candidate, FMLN leader and former guerrilla Salvador Sánchez Cerén, campaigned on a platform critical of metal mining, while their ARENA rival, Rodrigo Ávila, remained silent.Launching their campaign in November 2007, the Funes team spent 17 months bringing their "Caravan of Hope" to municipalities across the country (Almeida, 2014, 82-86).During a stop in Cabañas, Funes declared, "As long as [the mining companies] fail to demonstrate that these projects do not contaminate the environment and [do not damage] the health of our population, we are not going to permit metal mining" (Redacción Diario Co Latino, 2008).Building outward from the FMLN's formal membership of 80,000, his campaign networked with an array of social movement activists to get out the vote.Resistance organizations that had mobilized against privatization of healthcare and water in previous years were remobilized to support Funes's election (Almeida, 2009).Although the race tightened in the final lap, Funes emerged victorious in a 51.3 per cent vs. 48.7 per cent split.
A few weeks after Funes was elected, Pac Rim filed its claim against El Salvador at ICSID.The company's initial $90 million claim rose to $314 million (Pac Rim Cayman, 2013) before settling at $250 million as the case progressed.As the Funes administration officials moved into government offices and FMLN leaders renewed their commitment to a mining ban, the assassination of antimining leaders in Cabañas began.By the end of the year, three activists in Cabañas had been murdered.Intimidation practices reminiscent of the civil war era reappeared, taking an additional toll in the months that followed.

From the streets to the chamber
Unlike much of Latin America, where mechanisms of direct democracy became part of the institutional landscape during the period of electoral democratization (Breuer, 2007), direct forms of citizen participation remained limited in El Salvador.The country lacked a grassroots recall process, for example, and citizen groups could not introduce bills directly (Artiga González, 2007, 335-37;ES0117).Civil society groups with an interest in changing the law had to recruit cabinet officials or legislators to serve as sponsors of their bills.
Interest in mining remained high, however, and over the twelve years in which the metal mining conflict persisted, three kinds of proposals were presented to the legislature.The first two proposals involved top down initiatives devised by corporate lobbyists and administrative technocrats, respectively.The third bill, in contrast, reflected the iterative interactions between social movement and institutional actors that are commonly found in the mediated model of policymaking.
The first bill, introduced by Partido de Conciliación Nacional (PCN) legislators in November 2007, contained reforms that would clear the way for approval of extraction licenses.As noted in the ICSID testimony of Pac Rim managers, this bill was elaborated in close consultation with company lawyers (Pac Rim Cayman, 2016, vi-36).Corporate lobbyists and consultants were hired to mobilize support for a change in existing law that would remove provisions requiring applicants for extraction licenses to acquire surface property rights for land above subterranean tunnels. 11Lawmakers from the PCN, an old-line conservative party and frequent ARENA partner in the legislature, agreed to introduce a bill to this effect so that Pac Rim could resolve the deficiencies in its extraction application (Pac Rim Cayman, 2016, vi-36).This bill languished in 2008 as the Saca administration backed away from mine development.
The second bill emerged from technical discussions involving top state administrative officials and external experts.Based on a growing recognition of the kind of state capacity that is required for successful mine regulation, Salvadoran officials opened an internal discussion about a possible permitting freeze.In 2010, the Funes administration commissioned a full-scale external report to assess mining viability in El Salvador.Carried out by the Spanish consulting company, TAU Consultora Ambiental, and reviewed by a blue ribbon oversight committee to ensure its quality, 12 the report identified a long list of environmental vulnerabilities and described in detail the deficiencies in the government's regulatory capacity (TAU, 2011).According to this assessment, major policy change and institutional development would be needed in 17 areas, with 63 specific actions identified for change (TAU, 2011, Table 19, 81-84).Armed with this document, the Ministers of the Environment and the Economy jointly presented a proposal to the legislature in August 2012, calling for a formal moratorium.Noting that an outright ban could trigger ICSID demands by any of the 26 companies that currently had exploration rights, Environment Minister Hernán Rosa Chávez (2012) concluded, "In reality, [the moratorium] is the only option." Mesa representatives rejected this initiative, arguing that its moratorium reversal provisions could allow a future administration to declare the conditions ripe for mining, in spite of any community and environmental damage that would be unleashed (ES1012).Instead, the Mesa-backed bill called for an outright ban on metal mining.First introduced in December 2006, this bill was presented repeatedly on behalf of the Mesa by a dedicated group of FMLN legislators.According to the Asamblea Legislativa's online database, this bill or a lightly adapted equivalent was reintroduced between one and four times a year over the next eight years, for a total of 15 submissions between 2006 and 2014.This collaboration between Mesa activists and FMLN bill sponsors illustrates how movements may have a direct effect on the policymaking process in the agenda-setting phase.
A growing body of literature analyses movement-based political parties in Latin America, particularly in Bolivia (MAS) and Ecuador (Alianza País) (Anria, 2013;Van Cott, 2005).This kind of linkage emerges when a political party assumes the discourse, frames and mobilization strategies deployed by social movements; when movement actors support a political party's electoral work, encouraging an identification with a party, sharing its messages, and getting out the vote in favour of the party's candidates; and when, if victorious, party leaders name social movement actors to positions in government.These connections are often regarded as inherently unstable because of the tension between the institutionalization required for political party development and the non-institutional and transitory characteristics of social movements.The Salvadoran case represents a relatively durable version of this blend, perhaps because of the high degree of ideological polarization that persisted there in the post-war years.
The close link between the FMLN and Salvadoran social movements, described by Paul Almeida ( 2006) as "social movement partyism," has drawn repeated attention in the academic literature.Analysts of the large-scale mobilization against health care privatization in [2000][2001][2002][2003], where up to 200,000 participants turned out in waves and dressed in white to express solidarity with hospital workers, note the strong support provided by FMLN leaders (Almeida, 2014, 70-78;Smith-Nonini, 2010).Guillermo Mata, the president of the national association of physicians and the high-profile leader of this protest, went on to become the FMLN's Vice Presidential candidate in 2004 and, following that defeat, a three-term FMLN deputy.Mesa activists secured statements of opposition to mining from the FMLN's presidential candidates, and celebrated the appointment of Mesa allies to administrative positions when they won.

Adoption of a formal legislative ban
The antimining movement had succeeded in getting the mining debate onto the policy agenda in El Salvador by 2006, but this coalition had been unable to secure significant policy change.Counter pressure, in the form of corporate penetration of the domestic political process and use of international mechanisms through the ICSID process, stymied the Mesa's efforts to get mining banned, as did the overall post-war political settlement, which had long favoured market openness and economic deregulation.In 2017, however, after a twelve-year campaign, the situation suddenly changed.Within weeks, a new mining law, which prohibited all forms of metallic mining, won unanimous support in the national legislature.This dramatic policy shift leaves us with three questions: 1. How did this measure finally win legislative approval?2. How can we explain the absence of opposition votes?3. What was the role of the antimining movement in securing this outcome?
The present study points to three processes that ultimately contributed to the adoption of El Salvador's mining ban.First, the Mesa network continued to mobilize in opposition to mining, with demonstrable consequences in terms of public opinion.Second, external pressure was successfully deleveraged when the ICSID case was resolved, while leaving behind a sense of national vulnerability that added urgency to the deliberations.Third, the antimining movement recalibrated its leadership structure, building on bridging mechanisms to extend its reach beyond partisan alliances and stitch together a broadened coalition.This three-part explanation combines continuous elements (i.e., ongoing Mesa organizing), which fuelled a slow but persistent erosion of public support for mining, with catalysts (i.e., the resolution of the ICSID case and shifts in the anti-mining leadership structure), where sudden changes in the alignment of players opened new territory for policy debate.
Securing both domestic and international assistance, the Mesa maintained a durable coordination capacity and a growing base of public support.The movement had particular success in communities targeted for mine development.A public opinion survey in municipios where exploration concessions had been granted revealed that opposition to mining was already the dominant view in 2007.When asked, "Do you consider El Salvador an appropriate country for mining operations?" 62.5 per cent of respondents in mining communities said "no" (IUDOP, 2008, 1).A repeat survey eight years later found that opposition had become yet more comprehensive, with the per cent saying "no" increasing to 79.5 per cent (IUDOP, 2015, 58).Almost four out of every five respondents in this zone of heightened vulnerability concluded that mining was not an appropriate activity for El Salvador in 2015.
Further evidence of local opposition appeared when opponents adopted a variant of the consulta process employed by anti-mining movements elsewhere in Latin America (Walter & Urkidi, 2016).Using the citizen consultation provisions included in El Salvador's 2005Municipal Code (Código Municipal, 2010, 21), CRIPDES organized five municipal-level consultations on the mining question in the country's gold belt between 2014 and 2017 (ES1617; ES1517). 13The towns selected for these consultations were strongholds of the FMLN and politically atypical, even for their departments. 14The NO vote was predictably overwhelming in these towns, ranging from 98 per cent to 100 per cent.Strategically placing the consultas in propitious territory; dispersing the operational costs across the seasons and years; inviting domestic and international observers to witness the process; and promoting media coverage of the results, Mesa affiliates sustained the message of community opposition to mining and kept the debate in the public eye.Pacific Rim's ICSID demand raised the stakes on this issue, even as it impeded legislative action.After almost eight years of legal pressure, the company lost its ICSID case in October 2016. 15Once the threat of repeated pounding by disgruntled junior mining companies retreated, the legislative deliberations entered a few phase.The ICSID experience had raised alarm in the legislative chambers, and it contributed now to the forward movement.As one legislative observer (ES0117) noted, "With the claim filed in ICSID, we could see that we were unprotected, that any other firm could file a demand against us and we would be vulnerable.We needed to act to prohibit [metal mining] to prevent that from happening again.That [ICSID] decision could have gone the other way."Salvadorans had faced the threat of paying out $250 million to a company that had never fully established its right to mine, or of being required, through external pressure, to authorize the initiation of an industry with well-documented environmental risks, which the country was, on the basis of the best evidence, unprepared to manage.The possibility of a ripple effect, as other similarly situated mining companies followed suit, perhaps with greater success, brought the potential costs of inaction into sharp relief.
Ironically, an institution and process that had been designed to promote international investment by increasing protections available for investors may have served, at this stage, to accelerate investment restriction.As anti-mining mobilization continued and evidence that the country was unprepared to control the industry accumulated, political leaders increasingly recognized the need to take definitive action that would contain the threat of new legal demands.When the policy freeze imposed by active litigation was lifted, a ban suddenly became a feasible alternative.In the words of an ARENA legislator (ES0317), "This year we found there was a window of opportunity." During this juncture, a new round of high profile intervention by Catholic Church leaders drove the process forward.In this final lap, collaboration between the Mesa and the Church hierarchy shifted to accent the role of the Church.This recalibration of the antimining alliance triggered some tensions and indications of mistrust, which are not uncommon in bridging relationships.Although this dissonance came at a cost, the emphasis on Church leadership also succeeded in easing the partisan identity of the cause and in giving the process a fresh start.As one Asamblea respondent (ES0117) observed, One thing that helped the Committee and the Assembly to move forward in 2017 was that the Church presented a new proposal.The Committee wasn't required to go back to the old proposals that had already been debated and had already generated polarized responses.Instead, it could start fresh with a new proposal that was clear and straightforward.With the earlier ones, sometimes the bills were internally contradictory, or they missed an important issue.With this new one from the Church, those problems did not arise.This step involved collaboration among a variety of Church-aligned institutions and policy strategists, several of which were identified as bridging allies in figure 1.The Jesuit-run Universidad Centroamericana (UCA) took the lead on developing a new mining prohibition bill (ES1017; ES1717; McKinley, 2017).With assistance from the university's legal team and faculty members in the department of legal studies, the UCA's mission-driven "Social Projection" office oversaw a technical analysis of the bills under legislative review and, finding deficiencies, moved to draft a new version in 2016.The new bill prohibited metallic mining both above and below ground, and it set a two-year deadline for the elimination of artisanal gold mining operations.
UCA administrators presented the new proposal to Archbishop Escobar Alas, who embraced it as his own.Under his leadership, a delegation of Catholic authorities formally presented the draft bill to the Asamblea on February 6, 2017 with pre-arranged sponsorship from key legislative leaders. 16Three weeks later, the Archbishop organized a major signature drive in support of the ban, with priests circulating petitions before and after Sunday mass.More than 37,000 signatures were collected in a little over a week, and Church leaders gathered several thousand followers to accompany them as they carted boxes of petitions to the Asamblea.They were met at the entrance by a contingent of Assembly members, including Guillermo Gallegos, Assembly president and GANA representative, who promised rapid action.Claiming that mining was a "life or death threat", the Archbishop called on the legislators to step away from partisanship and deliver unanimous support (Gómez & Orellana, 2017).
Push back against heightened Church activism followed from well-positioned mainstream media and conservative leaders.An unnamed ARENA financier reportedly pressured Auxiliary bishop Gregorio Rosa Chávez (soon-to-be named as El Salvador's first Cardinal) to abandon his opposition (Jimenez, 2017).The Executive Director of El Diario de Hoy, the country's major conservative daily newspaper, tweeted "The return of the Church to politics fills us with sadness" (Redacción UH, 2017).But ANEP, El Salvador's business peak association, took no formal organizational position on the issue (ES1917).Even ARENA voices that had once favoured mining were now subdued (ES2617)."After the Pacific Rim case," one observer noted, "no one wanted to defend mining" (ES1017). 17 The bill was approved quickly in a 69-0 vote on March 29, 2017 (Asamblea Legislativa, 2017).After a decade of legislative delay, debate, and dismissal, legislators from all five parties lined up to endorse this new policy.Support was a given with FMLN lawmakers, who festooned the chamber with banners reading "No a la minería, sí a la vida."GANA, which had been formed in 2009 following Saca's expulsion from ARENA and served as the FMLN's frequent legislative partner, also endorsed the measure.The solitary PDC vote had been previously pledged in support of the ban, and provided mining opponents a slim margin of victory.The only question, then, concerned the legislators from ARENA and the PCN, who controlled the remaining 49 per cent of the seats (see appendix 1).
Deputies in El Salvador traditionally owed their candidacy to their party, not directly to the voting public, and party bloc voting in the Asamblea was the norm (FUSADES, 2017).Changing the voting position of a legislator normally required persuasion at the level of party leaders rather than recruitment of individual members.New election rules governing preferential voting for individual candidates and the option of split ticket balloting, however, hinted at a possible reconfiguration of legislative voting practices down the road.Change was developing slowly, although disciplined party voting was still the norm.
Reflecting on the unanimity of the prohibition vote, one long-time legislative observer noted that the 2015 legislative election brought into office a number of new representatives, and that ARENA in particular "had diversified" (ES0117).One member of this cohort was John Wright Sol, a US educated former firefighter in Fairfax County, Virginia, who hailed from a prominent ARENAfounding family.As a newly minted legislator, Wright Sol had volunteered for the Environment and Climate Change Committee and registered growing concern about the country's environmental vulnerability.His decision to throw his weight behind the 2017 mining ban helped move his party toward the emerging legislative consensus.
The biggest surprise was the favourable vote by the PCN deputies, who had a history of support for the mining industry, as discussed above.The PCN had been ARENA's most durable legislative coalition partner and was crucial to its ability to secure legislative majorities (FUSADES, 2010, 126).With ARENA's shift toward mining opposition, the PCN legislators were now abandoned.One legislative observer (ES0117) interpreted the PCN vote as an effort to avoid political isolation, which was destined, in any case, to be futile ("Their thinking was 'we'll be alone in our opposition and lose anyway'").Another (ES2517) viewed it as an attempt to avoid public condemnation, not just by Salvadorans but by the numerous international organizations that were now focused on the case.
This voting pattern suggests that strong "social cascade effects" (Buamgartner, 2006, 43) contributed to the legislative sweep as political parties that had once endorsed pro-mining legislation now moved into the opposition.As mining risks grew more visible and prohibition support accumulated, the ban coalition achieved a tipping point, and formerly oppositional parliamentary groups tumbled into the swelling majority, party by party.For those legislators who were not in agreement, the options were to adhere to party decisions to support the ban, in spite of personal reservations, or to not vote, either through absence or by non-participation.
Analysis of national public opinion data collected in the aftermath suggests that the ban enjoyed broad national support, and legislators may have been aligning themselves with shifting popular preferences.In national polls carried out in eight Latin American countries by CID Gallup in January 2018, respondents were asked, Do you believe that mining company operations should be permitted, or do you believe that they be prohibited?Fifty-eight percent of Salvadoran respondents said that mining company operations should be prohibited, second only to the 59 per cent registered in Costa Rica, where an open pit mining ban had been adopted in 2010 (CID Gallup, 2018).Fewer than a quarter (24 per cent) of Salvadoran respondents stated that mining company operations should be permitted, the lowest per cent in any of the eight countries. 18When those calling for prohibition were asked why they opposed the operation of mining companies, 68 per cent of those Salvadoran respondents identified environmental damage as their main concern.

Discussion and conclusions
As with several other contributions to this volume, the present study analyses the role of social movements in the policymaking process over an extended period of time.This kind of slow movement analysis permits long-form process tracing that identifies movement pulsation through surges and eddies.This extended analysis points to the usefulness of a mediated effects model, which follows activists and institutional actors as they engage in iterative interactions and identifies institutional docking points where proposals advanced toward policy change.By tracing the dynamics of contention, we see how movements may fail to gain traction when confronting agreements that are firmly embedded in elite political settlements or when venue shifts transplant contention from the national to the international scale.We also see how movements may ultimately succeed in building legislative support even in formerly oppositional quarters, and how policy equilibrium may be punctured, followed by the rapid construction of a new consensus.
Determining the extent of social movement impact on policy change is an inherently difficult task, however, given the number of factors that can contribute to this outcome.To address these conceptual challenges, Amenta (2006) used the tools of historical institutionalism to carefully excavate and sequence primary source material as he examined the impact of the Townsend movement on social security policy development in the United States.To determine "when movements matter," Amenta (2006, 31) reviewed the archival record to assess the following: whether a challenger altered the plans and agendas of political leaders; influenced the content of the proposals as devised by executives, legislators, or administrators; influenced the votes of representatives key to the passage of proposed legislation; or influenced the speed or nature of the implementation of legislation.
Using Amenta's guidelines as the standard against which to judge the impact of the anti-mining movement in El Salvador, the evidence presented in this study supports a positive assessment.The alliance formed by the Mesa Nacional and the leadership of the Catholic Church repeatedly pressed for a mining prohibition, encasing deliberations in an assessment of the environmental and human rights costs associated with mining development.This coalition drew on bonding and bridging ties to expand the network's spatial reach and multisectoral breadth, combining the Mesa's capacity to connect grassroots activists at the local and national levels, with the Church leaders' access to both elite and mass publics.Thick movement-party connections that linked Mesa activists and the FMLN allowed anti-mining leaders to gain access to party candidates who could be persuaded to pledge public support, including at the presidential level.These alliances also allowed Mesa leaders to locate FMLN legislators who would sponsor anti-mining bills on their behalf, thus shaping the content of proposals that would place this issue on the legislative agenda.
Challengers found it difficult to advance when movement targets shifted the venue to Washington, D.C., and the ICSID process constricted domestic debate.Transferring a hefty piece of the deliberative process to the international scale and encasing it in the strictures of administrative law disrupted the processes of democratic contention that had permitted social movement actors to bring this issue into the national policy agenda.Activists secured a rapid legislative finale, however, once the ICSID obstacle was removed.Church leaders played a direct role in drafting the bill that was ultimately approved and in lining up agents to spearhead the final push.By shifting the framing of the debate away from partisan identification and toward national environmental and legal risks, they broadened the consensus on mining and brought along the votes of key representatives who had once been on the opposing side.
It is possible, of course, that the prohibition would have been approved even without the sustained effort of mining opponents.Precise estimation of movement impact requires consideration of such counterfactuals (Amenta, 2006, 30-31).El Salvador's small size, heavy dependence on a single river system and well-documented vulnerability to climate change could have prompted a policy shift away from metallic mining based on technocratic criteria, even in the absence of a social movement advocating a ban.But it seems unlikely that the government would have taken this step without pressure from and accompaniment by movement actors.Bold technocratic leadership by elected or bureaucratic officials in El Salvador has not been widely observed, nor is it commonly seen in mining policy disputes elsewhere in the region.Only Costa Rica has adopted similar legislation, and in that case as well, the ban followed years of forceful organizing by environmental and community activists.
Mobilization by itself, of course, is "often insufficient to effect changes in public policy" (Amenta 2006, 24).Giugni (2004, 161) found "no consistent pattern of co-variation" between changes in organizational resources, such as increased movement membership and financial resources, and policy shifts that occurred over time.A spike in protests does not translate easily or automatically into an outpouring of new policies and programmes.To deal with the divide between the "streets" and the "chamber," social movement activists need to identify office-holding collaborators and commit to enduring forms of engagement with institutional actors, taking on the risks of formalization and centralization.
Achievement of a mining ban did not mean a radical redefinition of El Salvador's development model.Nor did it indicate that this policy change would be irreversible.Much of the political settlement devised by Salvadoran political and economic elites in the post-war era remained intact, and rollback of environmental reforms and implementation failures have been common problems in the region.However, the breadth of mining opposition, as registered in public opinion polling and the unanimity of the legislative vote, suggest that a push for renewal of metallic mining would face significant opposition.
We are left with the question of generalizability, the Achilles heel of every case study.Although El Salvador's national ban on metallic mining is highly distinctive, this study offers insight into the ways antimining coalitions may emerge, gain momentum and secure policy approval.The adoption of a ban on open pit mining in Costa Rica, the existence of mining prohibitions at the province level in Argentina, and local level mine suspensions in Guatemala and Peru, among others, suggest that there may be various circumstances under which antimining movements have consequences.Fuller specification of those conditions awaits future research.
underground mine construction from the owners of surface property, as required by Salvadoran law (Pac Rim Cayman, 2016).On the basis of this finding, the company's case was dismissed.16.The bill was presented by the Archbishop of San Salvador, José Luis Escobar Alas; auxiliary bishop Gregorio Rosa Chávez, the president of Caritas; Andreu Oliva, the UCA Rector; and José María Tojeira, the director of the Instituto de Derechos Humanos of the UCA.Fray Domingo Solís, in representation of the Franciscan JPIC, accompanied the second delegation the following month.17.See Broad and Cavanagh (2015) on the absence of strong domestic business support for mining in El Salvador and Costa Rica; see also Bebbington, Bury, Cuba and Rogan (2015, 198).18.Since the 2018 poll was carried out nine months after the Salvadoran ban was approved, reported opposition to mining may have been enhanced by the rallying effect of policy victory.

Figure 1 :
Figure 1: Overview of Salvadoran Antimining Movement Alliance and Mechanisms