Research, part of a special feature on State-Reinforced Transformative and Adaptive Governance of Sustainable Social-Ecological Systems

INTRODUCTION

Transformative governance has emerged as a new research area at the intersection of scholarship on governance, sustainability, and societal transitions. In particular, transformative governance research combines work on adaptive governance and sustainability transitions to investigate when and how environmental governance can shift a social-ecological system from an unsustainable, undesirable regime to a more sustainable, desirable one (Chaffin et al. 2016). Transformations emerge through interlinked shifts in resource characteristics, power relations, knowledge networks, and forms of social mobilization, and generally occur in phases. In the preparatory phase, actors recognize resource degradation and build relevant knowledge and networks. In the transition phase, often linked to a window of opportunity such as resource collapse or political change, agents bridge different levels and scales to mobilize participation, reconfigure power relations, and build new governance institutions. In the resilience-building phase, actors strengthen the adaptability and legitimacy of the new system through continued investments, deliberation, and experimentation (Gelcich et al. 2010, Resilience Alliance 2010, Olsson et al. 2014). Importantly, although adaptive governance supports the resilience of a particular social-ecological regime, and “typically emerges organically among multiple centers of agency and authority in society as a relatively self-organized or autonomous process” (DeCaro et al. 2017:1), regime transformation is expected to require combining self-organization with “more deliberate and structured intervention,” including structured coordination between stakeholders across multiple levels and substantial investments of institutionalized political, social, and financial capacity (Chaffin et al. 2016:409).

Accordingly, understanding transformative governance requires attention to what DeCaro et al. (2017) termed “state-reinforced self-governance” (SRSG), in which the state supports self-governance by local stakeholders (see also Sarker 2013). State support might materialize through various legal and institutional pathways, and DeCaro et al. (2017) proposed potential SRSG legal design principles to complement the institutional design principles established in the commons governance literature (Dietz et al. 2003, Cox et al. 2010). These proposed principles include reflexive and iterative mechanisms for adapting legal frameworks; allocation of legally binding authority and responsibility for social-ecological system governance; and tangible support from government to fulfill governance requirements. This framework provides a valuable starting point for empirical investigations of the state’s role in adaptive or transformative self-governance. Chaffin et al. highlighted that “Our understanding of transformative capacity in governance systems is in its infancy,” and they called for “Researchers and practitioners... to develop a set of global case studies investigating transformative governance to test the limits of constraints and opportunities...” (2016:416, 418). Crucially, the SRSG framework was formulated primarily with reference to Global North cases, and DeCaro et al. recognized that “work is needed to investigate if and how these ideas translate to more diverse contexts” (2017:14). We expand empirical application of the SRSG framework through a case study of Amazonian fisheries.

Specifically, we examined fisheries agreements in the Brazilian Amazon to help test and refine our understanding of SRSG. Fisheries agreements (acordos de pesca) are community resource governance institutions that arose as a grassroots response to conflict and resource degradation in Amazonian fisheries. In the late 1990s, the Brazilian government codified legal frameworks for state recognition and support for fisheries agreements, making them an important SRSG case from the Global South, where communities are especially vulnerable to climate change and extractivist resource relations. Amazonian fisheries agreements thus offer an opportunity to evaluate the SRSG framework in a new context, in which governance targets regime transformation and community empowerment. Our analysis shows that the SRSG framework effectively captures many governance dynamics in Amazonian fisheries, but the Amazonian case indicates a need for greater attention to questions of scale and social capital in SRSG.

Fisheries agreements in the Brazilian Amazon

At its core, SRSG explores the interplay between state rules for environmental governance and self-organization of resource users. The framework links principles for common-pool resource governance, grounded in often-informal community management regimes (Ostrom 1990, Cox et al. 2010), with a focus on formal state laws and regulations that can reinforce local institutions to achieve adaptive, sustainable governance (DeCaro et al. 2017). Fisheries are a paradigmatic common-pool resource system (Acheson 1988, Ostrom 1990), and examining fisheries governance in the Amazon applies the SRSG framework in the world’s largest fluvial ecosystem.

Artisanal fishing is central to the identities, heritage, and livelihoods of traditional riverine (Ribeirinho) communities in the Brazilian Amazon (Furtado et al. 1993, Furtado 2006). These fisheries are also repositories of social values and ecological and cultural knowledges. Fishing practices are embedded in these traditional knowledges, which generate informal rules and norms that help fishers conserve fish stocks (Smith 1985, Jimenez et al. 2019).

Formal state institutions, meanwhile, supported large-scale commercial fishing in the Brazilian Amazon beginning in the 1960s through federal tax breaks supported by the Superintendency for the Development of the Amazon (SUDAM). The industrial fishing complex that emerged was centered initially in Pará State and directed toward international markets (Smith 1985, Almeida 2006a). Facilitated by the legal doctrine of “res nullius” (regarding fish stocks as an unowned resource belonging to the first claimant), industrial fisheries development drove overfishing, both of industrially harvested species for national and international markets and species more commonly consumed in regional markets. Fish stock degradation is apparent even in remote areas where fisheries have historically been less exploited (Mello 1985, McGrath et al. 1993), and competition and resource degradation led to persistent conflicts over fishing rights (Benatti et al. 2003, Cerdeira and Camargo 2007, Costa 2010, Santos 2014) and an urgent need for fisheries transformation at the regional level.

In the initial preparatory phase of governance transformation, from the mid-1960s to early 1970s in Pará and Amazonas states, Ribeirinhos, organized through the Catholic Church’s Christian base communities, developed new fisheries management strategies to codify artisanal fishing rules and to oppose fishers “from away.” These external actors were often industrial fishing vessels known as geleiras (freezer ships) using indiscriminate fishing methods like trawl nets (Costa 2010). The first community agreements were not recognized by the state, however (de Castro and McGrath 2001, D’Almeida 2006, Cerdeira and Camargo 2007). Closer integration of community fisheries agreements with state institutions began in the 1990s, when the transfer of federal fisheries management from the Superintendency for Fisheries Development (SUDEPE) to the federal environmental agency (IBAMA) created a window of opportunity for a transition. Strengthened connections between Ribeirinho organizations, non-governmental organizations (NGOs), fishers’ syndicates, and the Catholic Church were reinforced by tangible support from international projects promoting participatory Amazonian fisheries management. Called colônias de pescadores (fishers’ colonies), fishers’ syndicates are today analogous to Brazil’s rural unions as a form of corporatist organization advocating for members’ interests and guaranteeing members’ access to state benefits.

Foremost among these international initiatives was the Iara Project (1991–1998), administered by IBAMA with German bilateral cooperation and funding from the World Bank, followed by the Floodplain Natural Resources Management Project - ProVárzea (2000–2009) implemented under the internationally funded Pilot Program to Conserve the Brazilian Rainforest. Through these processes, the state began to legally recognize and formalize local fisheries agreements through individual ordinances (Cerdeira and Camargo 2007, Oviedo et al. 2015), and Ribeirinho communities intensified their mobilization at the nexus between their traditional management practices, market forces, and state authority. Notably, these early fisheries agreements included local rules and norms within management programs, bridging local (previously informal) institutions with the formal legislative processes operating at larger scales (Begossi et al. 2011), and the agreements involved decisions by local residents to limit their own resource access to maintain resource productivity and guarantee local control over fisheries (de Castro 2002).

This transition phase of formal co-management experiences and growing Ribeirinho mobilization culminated in a 1998 federal provisional measure, later converted into a Regulatory Order (IN IBAMA 29/2002) still in effect, which established the formal legal framework for community fisheries agreements (Almeida 2006b). These state-recognized agreements comprise rules established through a participatory process by riverine communities that are direct users of a particular fishing territory, with the principal objective of stabilizing or reducing pressure on local fish stocks. According to the order, agreement regulation is the legal responsibility of the relevant government environmental agency, in accordance with the legal status of the area in question. In Pará, for example, an agreement in a state river or lake would be regulated by the Pará state environmental secretariat (SEMAS); if in a federal body of water, regulation falls to the federal environmental agency (IBAMA); if in a federal conservation area, regulation is by the federal conservation authority (ICMBio); and if in a state conservation area, regulation is by the Pará state conservation authority (Ideflor-Bio). The order thus gives autonomy to Ribeirinho populations to develop their own rules but simultaneously affirms the state as the “managing body” for the accords and the only actor capable of applying sanctions. Moreover, the creation of the community fisheries agreement framework does not exclude access to resources by outside users, as the 2009 Fisheries Law (Law 11.959) and general lack of state presence have reinforced res nullius for Brazilian fisheries. Nonetheless, this governance institution has attracted government and civil society attention as a potential model for promoting sustainable resource management by Ribeirinho populations (McGrath et al. 1998, Vidal 2010).

Existing literature on Amazonian fisheries agreements comes mainly from floodplain (várzea) lake systems in Amazonas State and in the lower Amazon near Santarém in Pará State. Many of these agreements focus on managing commercially valuable species, especially pirarucu (Arapaima spp.). Studies report positive outcomes for fish stocks and fishing communities from pirarucu fisheries agreements in Amazonas, which are based on community management methods developed at the Mamirauá Sustainable Development Reserve (Castello et al. 2011, McGrath et al. 2015, Campos-Silva and Peres 2016). Factors supporting the success of these agreements include community social capital built through prior Catholic Church initiatives (Gillingham 2001, Pinho et al. 2012); support from government and civil society organizations, especially through donor-funded projects (McGrath et al. 2015, Oviedo et al. 2015); and specific political-ecological characteristics of the pirarucu fishery, namely, pirarucu are a valuable source of income for local communities, and the fish are non-migratory, living predominantly in floodplain lakes where their numbers can be effectively monitored and communities are able to exclude outsiders and assert access and use controls (Gurdak et al. 2019). Pirarucu management in the Santarém region has been less successful, however, and relative to Amazonas, Pará State policy was less influenced by grassroots movements and communities lacked government support, especially for defining boundaries and enforcing agreements (McGrath et al. 2015, Gurdak et al. 2019, Arantes et al. 2022).

Beyond managing high-value species, some fisheries agreements aim more generally at protecting subsistence fisheries from commercial exploitation. Many fisheries agreements near Santarém were negotiated for this purpose, with agreements banning certain gear and practices to effectively exclude commercial fishing from floodplain lakes (Almeida et al. 2009). Similar conflicts motivated fisheries agreements in lake systems around Silves, Amazonas (Pinho et al. 2012). Studies report productivity and conservation benefits from these agreements, such as increased fish abundance and increased catch per unit effort. They also highlight the importance of local knowledge for defining lake system boundaries and monitoring by community members for excluding predatory fishing boats (Almeida et al. 2009, Pinho et al. 2012).

In short, research indicates that community fisheries agreements in floodplain lake systems aimed at managing high-value species or protecting subsistence fishing from commercial interests can reduce conflicts and increase fishing productivity. Monitoring, enforcement, and weak government support are persistent challenges, however, even for relatively successful agreements (Oviedo et al. 2015, Gurdak et al. 2019, Arantes et al. 2022). Research on fisheries agreements beyond the social-ecological contexts of pirarucu management and lake system fisheries, however, is sparse.

We examined a fisheries agreement in the Canaticu River of the Amazon Estuary. Unlike lake systems upriver, the Canaticu is a tidal river system, and the Canaticu Fisheries Agreement, which covers numerous communities and a complex array of fisheries resources, emerged not in response to external threats or commercial interest, but rather to address resource degradation internal to subsistence fisheries. The Canaticu case thus offers a distinct context for evaluating how SRSG and transformative governance operate within artisanal subsistence fisheries. We use the SRSG framework to analyze the Canaticu Fisheries Agreement, exploring how legal and institutional design principles have or have not been met in this case, and we draw lessons to improve our understanding of Amazonian fisheries agreements and the conditions for effective state-reinforced self-governance of common-pool resources.

Study area

Marajó Island (the world’s largest fluvio-maritime archipelago), located in the Amazon Estuary, is a center of Amazonian artisanal fishing, practiced by diverse Ribeirinho communities. The black-water Canaticu River flows from the forested interior of western Marajó southward into the white-water Pará River (Fig. 1). The Pará River connects Amazon River discharge, arriving through the Breves Straits, with the mouth of the Tapajós River and Marajó Bay. The estuary has a complex hydrology influenced by seasonal variation in river discharge as well as semi-diurnal tides (Almeida et al. 2004, Fassoni-Andrade et al. 2023). This dynamic environment supports diverse aquatic species that vary across the shifting waterscape. Tidal floodplain forests of this Marajó várzea ecoregion are generally characterized by limited tree diversity relative to upland (terra firme) forests and are dominated by a small number of species, especially palms such as açaí (Euterpe oleracea; Montagnini and Muñiz-Miret 1999, Almeida et al. 2004).

The Canaticu is the largest and most populous of the four major rivers of Curralinho municipality. From the 19th-century rubber boom through the 1950s, Ribeirinho settlement in the Canaticu was characterized by geographically dispersed households linked to patrons through the debt-based aviamento trading system that facilitated timber and rubber extraction from Amazonian forests. Notions of property centered around resources and viewed terrestrial and aquatic resources in a unified frame. Rubber trees, creeks, or sections of river were considered property of particular families or individuals who controlled who was allowed to fish or harvest forest resources on their property (cf. de Oliveira 2018).

Concentration of Ribeirinho families in larger settlements began in the 1960s when Catholic priests encouraged the formation of ecclesial base communities, which catalyzed various forms of collective organization, including community schools and rural workers’ syndicates (de Oliveira 2018). Today, the term “community” remains identified locally with Catholic base communities, while evangelical Christian settlements are called “congregations.” Accordingly, the roughly 1500 families (around 6000 people) inhabiting the Canaticu watershed are organized into 51 communities and congregations (Potiguar and Costa 2015), though we use “community” hereafter to refer generically to self-defined local social groups. Establishment of communities was associated with economic shifts as the timber industry declined and the aviamento system withered. Ribeirinho livelihoods became increasingly dependent on açaí palm: palm hearts (palmito) grew as a commodity while açaí fruit remained a foundation of Ribeirinho subsistence, comprising the second most important caloric source in traditional diets (after manioc flour) (Brondízio and Siqueira 1997, de Oliveira 2018).

During the 2000s, growing trade in açaí fruit and government land-tenure initiatives drove new organizational processes in the Canaticu. Building on social capital of the ecclesial base communities, residents began to form community associations to improve their negotiating power vis-à-vis açaí middlemen (de Oliveira 2018). Association formation gained further impetus in 2006 when the federal agrarian reform agency (INCRA) began to designate agroextractivist agrarian reform settlements (Projetos de Assentamento Agroextrativista - PAEs) in Amazonian floodplains to provide legal recognition and tenure security for Ribeirinho communities. Thirteen PAEs were established in the Canaticu watershed, and additional community associations formed to participate in PAE co-management. That same year the federal government created the Terra Grande-Pracuúba Extractive Reserve (a sustainable-use conservation area for traditional populations) encompassing the Canaticu headwaters. The PAEs and Extractive Reserve (RESEX) recognized the collective use rights of communities within their boundaries, but the territorial logic of these state categories, drawn as contiguous polygons of land, focused on terrestrial resource management and was not necessarily congruous with local practices. For example, residents of Boa Esperança community live on both banks of a Canaticu tributary, but PAE boundaries were drawn at the riverbanks, so residents of one side of the river are in a different PAE from residents on the other side, contradicting the social-ecological realities of Ribeirinho livelihoods.

In 2007, the Union of Rio Canaticu Community Associations (Central de Associações do Rio Canaticu) formed to unite local associations with a primary goal of collectively marketing açaí production. Residents also developed three “açaí ports” where families collectively commercialize production, and in 2014 they established a cooperative to take over collective marketing functions from the Community Associations Union (de Oliveira 2018). In short, the açaí economy, which provides roughly 50–75 percent of local incomes (Instituto Peabiru 2016), has been a catalyst for collective mobilization in the past two decades. Unlike in past resource booms, Ribeirinho families today are organized in communities and empowered by PAEs to govern natural resource trade. In the early 2010s, community leaders sought to build on this collective action experience and organizational capacity to confront the increasing scarcity of fish stocks in the Canaticu.

Unlike açaí, fisheries are not a major source of income for Canaticu communities, yet fishing is practiced by all households and is the primary source of animal protein in traditional diets (Murrieta et al. 1999). Although traditional Ribeirinho notions of property recognized individual or family rights over particular fishing areas, in recent decades fishers have increasingly operated from the perspective of Brazilian fisheries legislation, which considers fish as res nullius and allows registered fishers to fish wherever they wish so long as they obey regulations. Although many fishers adopted this open access principle of federal legislation (leading sometimes to conflicts with fishers asserting traditional claims), many also failed to obey federal regulations on fishing gear, practices, and seasonal closures. With virtually no state enforcement of fisheries regulations and little respect for traditional boundaries, Canaticu fisheries became effectively unregulated.

Moreover, fishers increasingly adopted predatory practices, such as bank-to-bank gillnetting (tapagem), poison fishing (bateção de timbó), and use of nets with mesh sizes smaller than the legal minimum. Tapagem (river-blocking) involves placing a net from one riverbank to another, blocking fish passage. Accounts from Canaticu describe nylon nets up to 100 meters long, with mesh smaller than 70 mm, blocking river headwaters less than 10 meters wide (i.e., blocking the river 10 times from bank to bank) during fish runs, interrupting the reproductive cycle of these species. Timbó (Ateleia glazioveana) is a plant used for illegal poison fishing in which macerated branches are used to “beat” river headwaters. Nearly all fish in the area where timbó is used die from asphyxiation due to lack of oxygen in the water. Regarding fishnets, the legal minimum mesh size is 70 mm (Instrução Normativa Interministerial MPA/MMA no. 12/2012), but nets with smaller mesh sizes are commonly used in the Canaticu, catching juvenile fish before they reach reproductive maturity. A growing local population and increasing use of predatory fishing practices pushed Canaticu fisheries into a classic tragedy of the commons, and local fish stocks began to decline (Oliveira 2012).

Community leaders observed deteriorating Canaticu fisheries and knew of fisheries collapses in nearby regions. Several leaders participated in the Marajó Territorial Development Forum (CODETEM) and knew that the federal government’s Sustainable Territorial Development Plan for the Marajó Archipelago, launched in 2007, identified fisheries agreements as a priority environmental management policy. Local NGO Instituto Peabiru had worked with Canaticu communities since 2008 to develop the açaí economy. When in 2012, Petrobras, the Brazilian parastatal oil company, announced its new environmental funding program, Petrobras Socioambiental, community leaders partnered with Peabiru and seized this window of opportunity to propose a project supporting a Canaticu community fisheries agreement. This Marajó Viva Pesca project launched in 2014 with Petrobras funding. With the Canaticu Fisheries Agreement process, fishers and partner organizations aimed to transform degraded local fisheries into a new regime of sustainable self-governance supported by state authorities.

DATA AND METHODS

We used project documents and key informant interviews to trace the Canaticu Fisheries Agreement process and to describe transformations in fisheries governance, practices, and resources in the six years after promulgation of the agreement.

Positionality

The first and second authors were born in Ourém, a city in northeastern Pará on the Guamá River. Their parents were smallholder farmers, and their upbringings included close associations with the river and artisanal fishing. In their professional careers, both have worked extensively with traditional Amazonian communities in public sector, NGO, and academic capacities. The first author began working with Ribeirinho communities in Marajó in 2011 as project coordinator for a regional NGO, working on initiatives related to environmental governance, non-timber forest products, and artisanal fishing. The second author lived in Marajó for two years and has for the past decade advised Ribeirinho communities in Marajó linked to social movements around extractive reserves. The third author was born in the USA and has conducted research on environmental governance and rural livelihoods in Pará since 2013.

The first author participated in the Canaticu Fisheries Agreement process as project coordinator for Instituto Peabiru, which led the Marajó Viva Pesca project from its conception in 2012 through the project’s closure in 2015. In 2019, the first author entered a postgraduate degree program, and in 2021, they returned to the Canaticu as a researcher to evaluate the fisheries agreement’s effects during its first six years of operation. The first author’s prior work with fisheries actors in the Canaticu facilitated access and rapport with interviewees, while the informed consent process clearly communicated their role as a researcher. Participants’ frank criticisms of the Marajó Viva Pesca Project’s legacy and triangulation of interview data with document analysis and personal experience give us confidence that the first author’s prior role with the project did not significantly bias interview data.

Document analysis

To trace the fisheries agreement process, the first author in 2021 conducted a close reading of the Marajó Viva Pesca project dossier (which included technical reports, meeting minutes, participant lists, and correspondence). This dossier was compiled by Instituto Peabiru and filed with Ideflor-Bio for recognition of the fisheries agreement (file 2015/205963), constituting an official archive of the agreement process. The first author’s firsthand experience of the project aided contextualization and interpretation of these documents. We used data from this archive to describe the co-construction of the fisheries agreement and the principal interests and concerns expressed by actors involved.

Interviews

In 2021, the first author conducted interviews in the Canaticu on the fisheries agreement’s effects during its first six years of operation. Due to limitations from the COVID-19 pandemic, the açaí harvest period (July-December), and costly and time-consuming fluvial transport in the Canaticu, we chose to conduct fieldwork in February-March 2021 through semi-structured interviews with a purposive sample of key informants comprising members of the fisheries agreement’s managing committee and community leaders who possessed uniquely specific, firsthand knowledge of the development of the agreement and its implementation. We sought to interview 19 local members of the fisheries agreement managing committee. The managing committee has 22 members. We excluded three members representing RESEX Terra Grande-Pracuúba (Sector 1, see Fig. 1), because this area is administered by the federal government and the fisheries agreement, whose effects we sought to isolate, is administered by the state conservation agency, Ideflor-Bio.

The first author successfully interviewed 15 managing committee members. Of the four members not interviewed, three had moved away from the Canaticu, while one was not present at the time of fieldwork. We succeeded in interviewing at least two members from each of the seven sectors of the agreement under investigation (cf. Fig. 1). We also interviewed one community leader who was not a committee member but who participated in the agreement process as president of the Community Associations Union, giving 16 interviews in total. The semi-structured interview guide contained open and closed-ended questions in four principal sections regarding: (1) interviewee profile; (2) perspectives on the original development of the agreement; (3) functioning of the agreement; and (4) compliance with agreement rules. Human subjects research approval involved presentation of the research to community leaders in a group meeting, at which permission was given for research in the communities, and interviews were conducted with informed consent of participants in accordance with Brazilian legislation and ethics rules established by the Federal University of Pará. Interviews were audio-recorded with permission and transcribed, then analyzed through inductive thematic analysis with coding in N-VIVO (Michelat 1987, Braun and Clarke 2006, Emerson et al. 2011, Cavalcante et al. 2014). Interview section 4 on rule compliance comprised 23 structured (yes/no) questions. Responses to this section were quantified, and agreement rules were then scored according to the level of compliance reported by respondents.

The 16 interviewees included 1 woman and 15 men with ages ranging from 28 to 62 years old. The four committee members not interviewed comprised two women and two men. The gender imbalance in our interviewees is a function of the gender imbalance on the managing committee. Two identified as members of an evangelical church (Assembly of God), while the remaining 14 identified as Catholic. All interviewees were born and raised in the Canaticu watershed. Regarding livelihood, 2 interviewees do not receive the federal off-season stipend for artisanal fishers (seguro-defeso) because they are affiliated with a rural workers’ syndicate, while the other 14 receive the stipend (13 through the Curralinho Fishers’ Syndicate Z-37 and one through a different local fishers’ association).

RESULTS

Tracing the Canaticu Fisheries Agreement process, 2014–2015

The Marajó Viva Pesca project was implemented through partnerships among multiple organizations, including the Community Associations Union, Curralinho Fishers’ Syndicate, CODETEM, Curralinho municipal government, and Instituto Peabiru. The project’s three principal axes were: (1) participatory development of rules for fisheries agreements among residents of the watershed; (2) biological research on major fish and shrimp species in the Canaticu fisheries; and (3) environmental education to disseminate study results and information about the agreement process. We focused on the project’s first axis related to the Canaticu Fisheries Agreement.

The 2002 IBAMA Regulatory Order established procedures for the development of fisheries agreements eligible for state recognition, which guided Instituto Peabiru’s work with Canaticu communities (Potiguar and Costa 2015). Table 1 summarizes this process based on document analysis of the Canaticu Ideflor-Bio dossier. The process was developed through six stages that enacted core SRSG legal and institutional design principles.

In Stage 1, acting on the legal authority granted communities under IN IBAMA 29/2002, and with tangible support from the Marajó Viva Pesca project, participants defined the boundaries of the social-ecological dilemma, namely overfishing in the Canaticu watershed (cf. Huitema et al. 2009, DeCaro et al. 2017). Project documents identified logistics as a principal challenge because river communities are separated by substantial distances with little local infrastructure for supplying the project team’s fuel, lodging, and food, yet the regulatory order required substantial discussion and popular participation for fisheries agreement development. A May 2014 meeting of leaders from the 29 Canaticu community associations and project partners discussed how to conduct a representative process that would engage all communities with limited project resources. Participants decided to define user groups by dividing the river into eight sectors aligned with Catholic Church administrative areas (Fig. 1).

In Stage 2, the agreement process advanced through sectoral meetings to define fisheries problems through participatory evaluations, and each sector elected representatives who became the 22 members of the fisheries agreement managing committee.

Stages 3–5 comprised rule development. In Stage 3, the managing committee worked with support from project partners to consolidate local practices with existing legislation to create the first set of proposed rules for the agreement (since agreements must be compatible with state regulations). This stage constituted a process of analytic deliberation aimed at devising rules congruent with the social-ecological conditions of the Canaticu (cf. Dietz et al. 2003). Project documents described four committee meetings: one for training and agenda-setting, two for development of proposed rules (reflecting discussions held by committee members with resource users in their respective sectors), and a final meeting to produce an action plan for monitoring the fisheries agreement after its formalization by state authorities. In Stage 4, rules proposed by the managing committee were discussed and validated through a participatory decision-making process comprising 22 additional community meetings across the 8 sectors to produce the final agreement text, which in Stage 5 was validated by the managing committee.

In Stage 6, the agreement was submitted to the relevant government agency, Ideflor-Bio. Formal recognition of the agreement via Regulatory Order IN Ideflor-Bio 01/2016 vested the agency with a legally binding responsibility to enforce the agreement (cf. DeCaro et al. 2017).

The Canaticu Fisheries Agreement co-produced through this 2-year process contained 17 articles, including prohibition or limitation of certain fishing practices and gear. For example, the agreement established minimum spacing between strips of matapi shrimp traps. Matapi traps are cylinders containing two inverted cones through which shrimp enter the trap and are unable to escape. The traps are made with strips of local palm species and used to capture Amazonian river prawns (Macrobrachium amazonicum). To maintain prawn stocks, the spacing between strips on the trap must be sufficiently wide to allow smaller individuals to escape. Because prawn fishing is one of the only fisheries that generates income for Canaticu families, the fisheries agreement process involved heated debates over minimum spacing for matapi strips. Agreement article 9 establishes “regulation of spacing for matapi and other gear used in shrimp fishing at 1 cm between parallel strips,” but recognizes in section 2 that communities on the Sorva and Tartaruga Rivers agreed to minimum spacing of 0.5 cm (IN Ideflor-Bio 1/2016 https://www.semas.pa.gov.br/legislacao/files/pdf/191.pdf). The agreement also contained full prohibitions on catching certain species (e.g., turtles, caimans), restrictions for capture of other species (tucunará and aracu fish), and place-specific access or practice restrictions (especially for headwaters).

Although it attempted to encompass the diversity of target species, conditions, and practices along the length of the river, the fisheries agreement included an expectation that rules would be refined locally through continued discussions (aligning with principles of reflexive law). Articles 10 and 11 expected communities to identify nursery areas where shrimp fishing will be prohibited and to establish a local limit on the number of matapi in accordance with local stocks. Article 14 mandated participatory monitoring of fisheries agreement rules and Article 16 stipulated that violators would be subjected to legal sanctions, while Article 15 provided for iterative decision making, stating the agreement can be revised at any time by request of the managing committee. Iterative decision making was further instituted under the 2002 Regulatory Order, which stated that agreements should be accompanied by monitoring of indicators for annual evaluation to determine the need for any rule changes. In the Canaticu, a March 2015 workshop produced a monitoring plan and a vision to orient the managing committee. Workshop documents, and the resulting action plan, emphasized the importance of training managing committee members in administration, environmental monitoring, and environmental legislation and securing resources to support the committee’s activities, as well as developing environmental education and conflict-resolution training for Canaticu youth and families.

In sum, the Canaticu Fisheries Agreement emerged through a co-production process between local communities and state and NGO partners aiming to transform a predatory fishing regime through new governance arrangements. The agreement process was structured by state guidelines and involved substantial social, political, and financial support to facilitate community self-organization. Formally, the agreement contained legal and institutional features identified as crucial for effective state-reinforced self-governance, including a participatory decision-making process, monitoring and enforcement provisions, designation of a responsible state agency, and iterative decision-making mechanisms. As Chaffin et al. observed, however, “There is potential for seemingly transformative policies and institutions to be merely aspirational, incomplete, or short-lived” (2016:416).

Transformations in Canaticu fisheries, 2016–2021

Interview data produced three primary findings regarding effects of the fisheries agreement: (1) the agreement changed some norms and practices, improving some fisheries resources; (2) lack of monitoring, enforcement, and conflict resolution undermined agreement effectiveness; and (3) absence of tangible support for iterative decision making weakened adaptive capacity and collective action.

1. The fisheries agreement changed some community norms and practices of fishing and self-governance, improving certain social-ecological characteristics of fisheries resources.

Commons governance theory emphasizes the importance of clearly defining user groups, resource boundaries, and harvest limits for sustainable resource management (Ostrom 1990, Dietz et al. 2003). For Brazilian fisheries, legal exclusion of “outsiders” from fishing access is not possible due to the status of fish and waters as public goods (D’Almeida 2006, Cerdeira and Camargo 2007, Oviedo et al. 2015). The Canaticu agreement, like other Amazonian fisheries agreements, addressed this limitation by determining that outsiders accessing the resource must comply with agreement rules. Setting catch limits for Amazonian fisheries, meanwhile, is complicated by the mobility of target species and a challenging terrain for monitoring and enforcement (Oviedo et al. 2015). Rather than imposing catch limits, the Canaticu agreement comprises prohibitions and limitations on use of specific fishing practices and equipment, harvest of particular species, and access to specific areas.

Six years after the agreement’s promulgation, the majority of interviewees (n = 11) reported that at least some Canaticu fisheries rules were being obeyed. According to one interviewee:

Timbó [poison fishing plant], which in the past they used a lot, now they also have awareness about that [tomaram consciência]. Today there’s no more of that timbó (participant 13).

Before, people came from Curralinho, they recounted:

they came and they invaded here, everyone who came had nets, they were going to night-fish with lights [lanternar], but after that happened, that they began to prohibit it, it stopped (participant 13).

Another reported:

There’s just one thing that... until today, we managed to maintain, since the time of the fisheries agreements, which is the part about spearfishing [espingarda de mergulho], right? Which we don’t allow in this sector here, and till today we still have plenty of tucunaré here in this area, because no one allows it (participant 15).

Espingarda de mergulho is an artisanal harpoon used principally for tucunaré (peacock bass) fishing. This practice, though selective, is highly predatory because it targets the female tucunaré when she is caring for her young. The first author recounts:

During fieldwork on the Croari River in Canaticu, I was speaking with residents about spearfishing and was told that the new rules had a positive effect, because tucunaré had “returned to the river.” As evidence, they pointed to the water to show me the tucunaré. I looked, and saw nothing but the river. I asked them where the fish was, and everyone on the dock (two interviewees, the boat pilot, and the guide) pointed to the spot: “there!” I looked again, and saw nothing. Seeing that I still hadn’t spotted the tucunaré, they all again pointed to the same spot in the river. At last I noticed something, a straight line of bubbles in the water lilies. “Oh, there!” I exclaimed, “It’s those bubbles!” An interviewee explained that the bubbles were from the young that stay near the surface with the mother directly below them, and for this reason spearfishing is so destructive: the lines of bubbles lead the fisher directly to the reproductive female.

Interviewees also recounted ongoing self-organized efforts to combat violations and bring practices in line with the agreement. One interviewee described numerous meetings and discussions that finally succeeded in protecting a nursery area (participant 5). According to another:

There was a time, last year, because the aracu come to the riverbanks [to spawn], right? So people had a meeting and they stopped catching them, because the aracu were disappearing. So they stopped catching the fish that were coming to the riverbanks (participant 12).

Fifteen of the 16 interviewees affirmed that after the agreement process there was a change in resource users’ behavior, diminishing predatory practices and improving the condition of the fisheries. In particular, they emphasized how the agreement process strengthened local compliance with pre-existing state regulations. Sustainable practices

stuck, yes... from the agreements. Even from before the agreements, right? But there was a commitment with the agreement... but the rule was already there beforehand, then we just implemented it and managed to maintain it, right? It became stronger (participant 15).

As the interviewee indicated, prior to the fisheries agreement, there was clear understanding in the Canaticu about predatory fishing practices: many were already legally proscribed, such as poison fishing and use of nets with small mesh sizes, but prior to the agreements there was insufficient collective organization to restrain those practices.

Look, another observed, when I was coming up, fishing here, I grew up with my father, his fishing method was to block streams and poison fish [bater timbó; laughs]. Not anymore, we’ve stopped that, we stopped stream-blocking, stopped timbó, now we’re fishing with nets and lines, so there was already more change (participant 16).

Man, Lord have mercy, brother, another exclaimed, Because we came, and we hit the buzzer, right? We sounded the alarm alright, those who participated [in the agreement] sounded the alarm, and it seems like some people started to take notice of the situation. Man! And then the fish started to come back, tucunaré. You wouldn’t believe it, man! All the big fish! (participant 5).

Similarly, responses to the rule-compliance questionnaire indicated that after 6 years, out of the 23 key rules included in the fisheries agreements, 16 were generally respected, while 5 were generally not respected and 2 elicited a split response (Table 2).

These results indicate compliance with a substantial number of fisheries agreement rules, but several contextual details are essential for interpretation. Fishing with explosives and electric shock fishing were not understood to be present in the Canaticu prior to the agreement, according to minutes and workshops of the agreement process, but they are banned by the Fisheries Law (Law 11.959/2009), and so those prohibitions were included in the agreement. Close reading of managing committee and community validation meeting minutes reveals six primary concerns in fisheries agreement discussions: fishing nets [malhadeira]; aracu; tucunaré; cast nets [tarrafa]; river-blocking [tapagem]; and garbage. Transformation of these issues should thus be central to our evaluation of the fisheries agreement.

The agreement stipulates that use of fishing nets should obey rules determined locally by communities or users of a particular area, and those rules should be posted on signs in their respective areas (Article 3). During the agreement process, debates over rules for fishing nets were long and contentious: some wanted to ban them entirely, others wanted to maintain the status quo, but there was consensus on complying with legal restrictions (e.g., mesh-size minimums). Participants decided to decentralize fishing net rulemaking to localities to produce rules adapted to the diversity of the watershed. Questionnaire responses showed little progress occurred on this most contentious issue: only one interviewee reported the existence of local regulations on fishing nets. An interviewee expressed this ambivalence about the agreement’s effects:

Look, in my opinion, ...fishing practices before [the agreement] were practices that, in my view, weren’t conscientious, right? People fished with any type... yeah... even today that’s the case, just that before there was that question of timbó, which thankfully doesn’t happen anymore, but... today there are still many people fishing improperly, with fine-mesh nets, fishing during the spawning season ban, so I think that there was a small change just in that question of timbó, but with regard to fishing nets, cast nets, those things, I think that almost nothing changed, especially here in our community (participant 2).

Another similarly reflected on the fishing net problem:

In truth, from then to now, since the agreement, we haven’t had any more rules... because really the Fishers’ Syndicate was going to enforce that, right? - there was supposed to be someone to enforce it... But I see that before, fine-mesh nets were prohibited, right? I remember, from the fisheries agreement, those 0.25 [fine-mesh] nets weren’t allowed, right? And today they’re being used... There are many people using those small, fine nets (participant 15).

Governance of cast net fishing, banned by the agreement except in two specific areas (Article 7), was likewise ambivalent, and some interviewees reported persistence of this practice.

The agreement further stipulated that after its promulgation, aracu fishing would be suspended for one year (Article 5), and then could be resumed based on locally established rules. Documents from the fisheries agreement process identified aracu as locally threatened due to predatory practices. During the second managing committee meeting (27–28 September 2014), committee members raised the problem of predatory aracu fishing, occurring when the fish migrate to headwaters to spawn (which happens multiple times during the year, or during the full moon, according to fishers). One of the participants affirmed:

before aracu was caught in large quantities (12 dozen aracu), today you don’t catch a dozen...

Six years later, questionnaire responses indicated little progress on these rules: three-quarters of respondents still reported lack of effective aracu governance.

Interviewees also reflected on this mixed record of aracu fishery transformation.

I remember that one thing that improved is that I remember that in the agreement we had one year where we would not fish aracu, right? participant 2 recalled.

And during that period, I remember that we had a great result here in our community on that... But after, it seems that people started to disobey the agreement again. So for me the agreement is something that has to be in the community, they affirmed, speaking to the need for ongoing self-organization to complement formal agreements.

In that vein, another participant recounted a community meeting resulting in an agreement to protect spawning aracu near riverbanks (participant 12).

Regulations on tucunaré fishing appear more effective, however. As with aracu, the agreement imposed a one-year prohibition on tucunaré fishing to be followed by reopening under locally determined rules (Article 6), though spearfishing would remain prohibited (Article 7). Questionnaire responses indicated a substantial reduction in predatory tucunaré fishing and residents reported increased abundance of the species.

River-blocking (tapagem) is banned by the agreement (Article 7), and questionnaire results indicated substantial compliance with this prohibition.

People got the message, because before, it [river-blocking] happened, one interviewee reflected. They really hurt the fish, it was a time of river-blocking, and today we don’t see that anymore. It was bad for the fish, you know? (participant 6).

A second interviewee observed:

Mainly when it was spawning season, you know, people would go up and river-block, they’d go up with nets to catch the fish that were arriving to spawn, right?...That changed, you know, after the fisheries agreement, it changed, yes, we had more control, but then, more recently, things are trying to go back to how they were, and we’re hanging on, you know, trying to organize again, the people, because the group that did everything [with the agreement] complied, but there’s a new group now, that’s trying to take control of the fishery (participant 1).

This commentary emphasized that the river-blocking ban, like other rules, was a contingent achievement, requiring continued community commitment.

Proper waste disposal was another key concern of Canaticu residents, and the agreement mandated proper disposal of garbage by all river users (Article 12). Questionnaire results indicated substantial success on this item, though the degree to which this result can be attributed to the fisheries agreement is dubious because at the time of fieldwork the Catholic Church was campaigning against irregular waste disposal in the river.

Timbó fishing was less discussed during the agreement process, largely because there was already broad consensus on its prohibition, and following the agreement, respondents unanimously affirmed that poison fishing was no longer occurring. Although only this rule received universal compliance, the majority of agreement rules are respected, according to the majority of respondents.

2. Lack of monitoring, enforcement, and conflict resolution undermined agreement effectiveness.

Commons governance theory suggests monitoring and enforcement in small-scale commons are best conducted by resource users, who fulfill these functions more effectively and at lower cost than external actors (Ostrom 1990). In the Canaticu, however, formal monitoring of the fisheries agreement is sparse or non-existent, and although six interviewees stated that rule-breakers are sanctioned by communities, seven respondents indicated that sanctioning depended on external actors like the Curralinho Fishers’ Syndicate and the Municipal Environmental Secretariat (SEMMA), while another three did not see any sanctioning for violations.

Look, said one interviewee, truth be told, that [monitoring and enforcement] now just depends on the communities, you see? Because for a long time we don’t have any presence [of external actors]. For example, I was sometimes going to SEMMA, but it’s been more than a year since I’ve been there to discuss things [like violations] (participant 9).

Another interviewee stated that the Curralinho Fishers’ Syndicate’s enforcement of the spawning season [piracema] fishing ban was the only significant enforcement of fisheries rules, saying:

The stronger rule is during the piracema, outside of piracema there is no rule that can punish people who are doing something wrong in the fishery (participant 2).

Indeed, one interviewee recounted syndicate enforcement in a case of river-blocking during spawning season:

Look, it’s happened when they do river-blocking [tapagem] here, during the spawning season, people from the syndicate came, and they took the nets, they took the gear, opened up the blockage, and there was punishment (participant 16).

In the absence of formal monitoring, however, fisheries rule enforcement, when it occurs, largely depends on individuals reporting violations to authorities. In the fisheries agreement process, monitoring or stewardship of the agreement was addressed by the managing committee principally in terms of training and financing, with relatively little attention to enforcement. That focus reflects both the reservation of legal sanctioning powers to state authorities, and the challenging logistics (e.g., large distances, high fuel costs) for monitoring Amazonian fisheries. The interviewees and members of the managing committee did not associate monitoring and enforcement activities with their committee or its action plan. According to federal fisheries agreement regulations, the managing agency for the agreement, in this case Ideflor-Bio, has responsibility for sanctioning agreement violations, but Ideflor-Bio has been physically absent from the Canaticu since the agreement and has not maintained communication with the managing committee. Thus, rule enforcement is not exercised by the relevant authority and rarely happens communally but rather relies on individual residents reporting violations to external actors.

Community members may be reluctant to denounce a neighbor, however, so the enforcement regime not only lacks active monitoring, but also may not be activated for violations known to resource users. An interviewee explained:

So they have that body that monitors it, just that they wait for someone to make a report, right? ...For example, if they see someone here, the guy is a fisher and is doing something wrong, there has to be someone to report it, right? To the syndicate, so the syndicate can call on that person to fix things. Either you follow the rules or your [off-season] stipend will be suspended, will be canceled... But that’s rare... It almost doesn’t happen among us here... because we’re not going to hurt someone [by reporting them], right? We talk to the person and advise them, you know, tell him what he should or shouldn’t do (participant 2).

Another interviewee complained of this reactive, command-and-control enforcement approach:

I always tell people, even when I go to SEMMA, I always question that: ‘Look, you can’t come just when the fire is burning, you have to go there when the person has the fire in their hand, right? Before they set fire to the forest.’ Why am I saying that? Precisely because there are many people who just want to punish, just want to impose fines, but that’s not the way. You have to invest in dialogue, you have to tell people, ‘Hey, you know if you do that you’re going to pay for it, you may have to pay a fine, may go to jail...’ (participant 9).

In short, there is little collective self-organization of resource users for monitoring and enforcement of fisheries agreement rules, and Ideflor-Bio, the relevant governmental authority, has been absent. Soon after the agreement was promulgated, some managing committee members reported violations to external authorities, but the managing committee has not assumed a formal enforcement role, and sanctioning violations depends primarily on individual community members reporting rule breakers to the Curralinho Fishers’ Syndicate or SEMMA.

This weak enforcement framework has implications for conflict resolution. The 2015 Managing Committee Action Plan included conflict resolution training for the committee, but that training never occurred, and in practice, fisheries conflicts have been resolved either on an individual basis between the parties or through existing social channels. There are no dedicated collective mechanisms for resolving fisheries conflicts beyond the community level, leaving the coercive power of state or parastatal agencies (e.g., Curralinho Fishers’ Syndicate) as a final option, and fishers may therefore accept suboptimal resolutions to avoid involving external authorities (which involves both procedural costs of a formal complaint and potential costs to social cohesion). One interviewee described the fraught process of enforcing rules and resolving conflicts in the absence of formal institutional support:

One time I called out some people that were fishing brooding tucunaré, you know? And they weren’t obeying [the rules], and I had already talked with them twice. So then I called for us to go for a mediation, for guidance really. I remember that at the time it was M— [president of the Community Associations Union], I believe, who was there. Then there was a disagreement, because they said that they couldn’t afford to go hungry, but after hearing both sides they understood. In fact, after that they started to help me combat the others [who were fishing tucunaré improperly], who were everywhere, they were coming from the other river, from Tartaruga [River], people were coming here wanting to fish, it was terrible. Then the guy who was there, M—, said, ‘Look, if you don’t work together, preserving today, tomorrow you won’t have anything.’ I think that [going through that process] was worth it, because I would have stayed silent, because they were already angry with me, and, thank God, it didn’t improve as much as it could have, but for what it was [it improved]... (participant 9).

In sum, lack of monitoring and enforcement has resulted in low compliance with some fisheries agreement rules. Continued predatory practices degrade resources and generate social conflicts for which resolution mechanisms are lacking. Over time, these negative governance feedbacks may undermine social cohesion and reduce social and ecological resilience to new conflicts or resource pressures.

3. Absence of tangible support for iterative decision making weakened adaptive capacity and collective action.

Despite a fisheries agreement provision allowing for revisions upon request of the managing committee and an expectation that the committee would conduct yearly evaluations of the agreement, in practice, ongoing collective deliberation on fisheries rules has been minimal. Eight interviewees reported that since promulgation of the agreement, there have been no follow-up meetings for evaluating rules, while four reported occasional collective discussions of fisheries linked to entities like the Curralinho Fishers’ Syndicate and Catholic base communities, and just four reported routine collective discussions of fisheries issues in their areas. Interviewees described declining collective mobilization since the agreement and end of the Marajó Viva Pesca project. In a typical account, one interviewee stated:

I think that since, since we stopped [the project], we always meet, you know? But today, I think that for the last two years or so we haven’t met to talk about fishing issues. So since the guy said: ‘the project’s over,’ we didn’t get together anymore, didn’t discuss those fisheries agreement things anymore (participant 2).

No, another interviewee concurred, just when you [the project] came back, then we all met, we got together to discuss and after that, no more (participant 3).

More recent discussions of fisheries in Canaticu have occurred not through the fisheries agreement, but through activities of the Curralinho Fishers’ Syndicate or local churches, such as a yearly community visit from a syndicate representative to discuss the spawning season fishing ban (participant 11), but respondents emphasized that organized collective deliberation on fisheries governance depended on the tangible support of external actors during the fisheries agreement process.

Although they feel free to self-organize around fisheries management, interviewees also indicated a desire for orientation and support from external actors. Scholarship on Amazonian fisheries agreements emphasizes that successful agreements depend on support for resource users from multiple partners (Raseira et al. 2006, Seixas and Kalikoski 2009, Vidal 2010, Santos 2014, 2019, Costa 2017) and lack of support increases the likelihood of failure (Cerdeira and Camargo 2007, Costa 2010, Oviedo et al. 2015). In the Canaticu, organizations that supported the fisheries agreement process (Curralinho Fishers’ Syndicate, Instituto Peabiru, Community Associations Union, and Curralinho Municipal Environmental Secretariat) are no longer actively supporting fishers to implement the agreement; indeed, they ceased support with the closure of the Marajó Viva Pesca Project in 2015, before the fisheries agreement formally took effect. Moreover, there is no evidence of physical presence or regular communication by Ideflor-Bio, by law the managing agency for the agreement. Withdrawal of tangible support means iterative evaluation and updating of the agreement has not occurred, weakening adaptive capacity and collective action for fisheries governance (see Table 3).

DISCUSSION

Legal and institutional design of the Canaticu Fisheries Agreement

DeCaro et al.’s (2017) SRSG framework effectively captures many governance dynamics of the Canaticu Fisheries Agreement. Applying legal design principles focused on reflexive and iterative decision making; allocation of legally binding authority and responsibility; and tangible state support along with Ostrom’s institutional design principles helps explain the mixed governance outcomes observed in this case. Table 3 links these SRSG principles with the institutional design of the Canaticu Fisheries Agreement and subsequent governance actions and social-ecological transformations in Canaticu fisheries.

Balancing state legal systems and tangible support with self-organization of resource users is a core challenge for SRSG. In the Canaticu, despite decades of collective action centered on religious communities and the açaí economy, external actors were necessary to the fisheries agreement process and local collective action for fisheries governance dissipated after the Marajó Viva Pesca project. Effective monitoring and sanctioning are essential to transformative governance, but rather than embedding these functions in communal institutions, Brazilian fisheries agreements maintain the state as the legal enforcement authority. While this model enhances communal management relative to previous state-centric governance arrangements (Seixas and Kalikoski 2009), it also institutionalizes cross-level contradictions in governance functions between communities that have legal authority to create but not enforce rules and state agencies that maintain legislative and enforcement powers but lack resources to conduct monitoring and enforcement. In the Canaticu, although managing committee members initially experimented with a more active role in reporting violations, Ideflor-Bio, the state agency legally responsible for monitoring and enforcement of the agreement, was wholly absent.

Lacking tangible external support or institutionalized collective structures for monitoring and enforcement, Canaticu fisheries reverted to an enforcement regime dependent on external state or parastatal actors (SEMMA, Curralinho Fishers’ Syndicate) responding to individual complaints. Other Amazonian fisheries agreements have addressed this governance gap by creating Volunteer Environmental Agents, bridging actors who are drawn from resource-user communities and monitor agreement rules, referring infractions to the government managing agency for punishment (Cerdeira and Camargo 2007, D’Almeida 2006, Vidal 2010, Oviedo et al. 2015, Santos 2019). These agents are regulated on a case-by-case basis; they were not foreseen in the federal fisheries agreement regulation (IN 29/2002) and were not included in the Canaticu Fisheries Agreement. In the Canaticu, the managing committee was expected to be a bridging actor, with the committee mobilizing community members to monitor the agreement and report violations for referral to Ideflor-Bio, but without tangible support after the Marajó Viva Pesca project, the managing committee’s action plan was never implemented.

Under Brazilian law, fish stocks and waterways are public goods: communities cannot exclude fishers from other areas, but they may use fisheries agreements to control fishing practices and gear, fishing seasons, protection of spawning grounds, and catch volumes (Santos 2019). In the Canaticu, as in most Amazonian fisheries agreements, catch limits or minimum size requirements for individual species were not included because these regulations would require intensive monitoring (Oviedo et al. 2015), and agreement rules focused instead on regulating practices, seasons, and fishing areas. These rules approximate a reflexive law approach by avoiding rigid harvest limits in favor of more general fisheries standards and ground rules (DeCaro et al. 2017). Six years after the Canaticu Fisheries Agreement, the majority of rules are respected by resource users, though several rules lack broad compliance, notably regarding use of fishing nets (which were heavily debated in the agreement process) and shrimp fishing (which is practiced commercially). For several species (e.g., tucunaré, aracu), interviewees reported improvements in fish stocks following the agreement.

Presently, there is little ongoing collective deliberation on fisheries rules among residents and no conflict resolution mechanisms internal to the fisheries agreement. Adapting and improving fisheries governance for broader transformation of the resource system would require sustained collective action to review and deliberate on rules and governance arrangements and effective mechanisms for cooperative conflict resolution. The current scenario of relatively low social capital in the fisheries agreement contrasts with self-organization and collective action by Canaticu communities before and during the agreement process, suggesting that with tangible support from government and NGO partners, the fisheries agreement could be revitalized. Although fishers’ appeals for support indicate a certain dependency on external actors for fisheries governance, that support, especially from state actors, is indispensable when sanctioning and social benefits (off-season stipend) depend on the state, and co-management arrangements (i.e., the fisheries agreement) require substantial ongoing institutional investments. Participation by Ideflor-Bio, officially the managing governmental agency for the agreement, is essential, but hampered by low capacity in this agency.

Lessons for Amazonian fisheries agreements

The Canaticu case shares with other Amazonian fisheries agreements challenges of weak monitoring, enforcement, and government support (Aquino 2007, Oviedo et al. 2015, Gurdak et al. 2019, Arantes et al. 2022). Volunteer Environmental Agents (VEAs) for fisheries agreements are often hampered by lack of enforcement power and weak government support (dos Santos 2005, Santos 2019), but they can nonetheless fulfill crucial governance functions and their absence in the Canaticu is notable. Published examples of fisheries agreement VEAs are in partnership with IBAMA; the Canaticu case suggests Ideflor-Bio should support VEAs for fisheries agreements under its responsibility.

Property relations in Canaticu fisheries provide new insights on PAEs and the role of fishers’ syndicates. Creation of PAEs near Santarém was based on existing fisheries agreement territories and fisheries agreements were incorporated into PAE utilization plans so that PAE formation reinforced fisheries governance institutions (McGrath et al. 2015). In the Canaticu, PAEs were formed based on Catholic Church administrative areas prior to the fisheries agreement process, and PAE boundaries did not necessarily respect social-ecological realities of communities or their fisheries. The Curralinho Fishers’ Syndicate, meanwhile, has generally prioritized compliance with federal legislation (based on open access principles) over implementation of the community fisheries agreement. Projetos de Assentamento Agroextrativista (PAEs) and fishers’ syndicates are state or parastatal institutions that have the potential to reinforce sustainable fisheries governance (for example, by making social welfare benefits contingent on fisheries agreement compliance), but the Canaticu example emphasizes that those institutions must be harmonized with the territorial scale and legal principles of fisheries co-management.

From an institutional design perspective, some aspects of floodplain lake fisheries appear more propitious for effective governance than Canaticu tidal river fisheries. Dry season lakes have clearly defined boundaries, making it easier to exclude predatory outsiders, and biological characteristics of pirarucu (obligate air-breathers) facilitate population monitoring (Castello et al. 2011). Difficulty excluding outsiders does not explain Canaticu agreement outcomes, however, because fisheries degradation was driven by local fishers, not industrial vessels. Likewise, there is sufficient local knowledge to potentially monitor key Canaticu fish populations such as tucunaré (in which locals can identify locations of reproductive females) and shrimp (cf. Costa 2010). Rather than biology, tangible support from external actors appears to explain the development of monitoring systems, such as support from the Mamirauá Institute for pirarucu monitoring (Castello et al. 2011, Gurdak et al. 2019) and the NGO FASE for shrimp monitoring (Costa 2010). Geography and biology do not overdetermine governance outcomes but rather must be understood in interaction with socioeconomic and political characteristics of the resource system.

Mixed outcomes for Canaticu Fisheries Agreement rules call attention to the heterogeneity of fisheries covered by the Canaticu agreement, which encompasses 51 communities targeting diverse species from the Canaticu headwaters to its mouth. This breadth contrasts with the smaller territories or number of species covered by many other agreements. For example, the Gurupá Fisheries Agreement was specific to shrimp and began among 10 families in 2 communities (Costa 2010). Our results indicate some community-level differences in Canaticu fisheries governance. Participant 12 described meetings in one community that led to an agreement protecting spawning aracu, for example, but most interviewees report non-compliance with aracu fishing rules. Our research focused on aggregate agreement outcomes, and we do not have sufficient data to explain why certain rules were respected in certain communities, but the clearest variation in our data appears not at the level of communities but rather at the level of specific fisheries. Notably, aracu and shrimp fishing rules had low overall compliance, while tucunaré fishing rules had high compliance. Explaining these differences would require examining variation across biological, geographic, technical, and socioeconomic variables. This heterogeneity reinforces Castello et al.’s argument that variation in small-scale fisheries across communities means that “each fishing community must be considered a distinct management unit” (2013:563), requiring specific, localized rules. Amazonian fisheries agreements thus illuminate questions of scale as a crucial consideration for SRSG.

Lessons for state-reinforced self-governance: scale and social capital

Our Canaticu Fisheries Agreement analysis highlights that social capital and the social construction of the self-governance scale mediate between SRSG legal frameworks (fisheries agreement regulations) and the character of the resource (fish biology, biogeography, social organization of resource users), and these dimensions need to be more fully integrated into the SRSG framework. Literature on commons governance and resource politics does not always critically examine the construction of the local community level at which self-governance is institutionalized (cf. Brown and Purcell 2005), but in the Canaticu, mixed governance outcomes for a watershed characterized by resource heterogeneity, difficult logistics, and social differentiation raises the question of whether fisheries self-governance started at the wrong scale. Indeed, Marajó Viva Pesca project partners envisioned a second phase of the project to focus on specific sectors and communities to further develop and implement local rules for individual fisheries, but the second phase was not implemented due to the end of external funding. Consequently, fisheries agreement rules largely remained at the more general level of the watershed, in which social organization was weaker relative to the sector or community level. Although the fisheries agreement applied reflexive law principles for reinforcing local decision making, for example by decentralizing decisions on fishing net rules to the local level, these provisions did not result in local rulemaking after the Marajó Viva Pesca project’s end.

Grassroots governance at the community or sector level might prove more enduring and effective than governance at the watershed level by targeting a more homogenous management unit at a scale with higher scale-specific social capital. In Gurupá, for example, a fisheries agreement for a single species (Amazonian river prawn) began with two small communities, added communities over the course of several years, and was eventually formalized at the municipal level (Costa 2010), suggesting a bottom-up pathway for SRSG. Scalar trade-offs are always present, however. Local specificity cannot ignore connectivity (fishing practices in one community affect others in the watershed) and funders and NGOs often have institutional incentives to implement projects at larger scales (Thaler 2021). In short, research must examine the dialectical relations between state legal structures and scales of self-organization for effective SRSG. State-reinforced self-governance is thus always a multilevel governance problem and complementing the SRSG framework’s appreciation of polycentricity with deeper engagement with multilevel governance concepts (e.g., Brondízio et al. 2009, Mwangi and Wardell 2012) could enhance the framework’s approach to scale.

Social capital is an important variable in commons governance literature (Bebbington 1997, Ostrom 2009), but it is especially crucial for multilevel governance problems at the heart of SRSG, in which institutions must facilitate cross-level coordination and communication (Brondízio et al. 2009). The Canaticu Fisheries Agreement constituted a new level of fisheries governance, creating watershed-level institutions to link communities and the state, but the agreement suffered from weak social capital mobilization at the watershed level to bridge community and state institutions. Fisheries agreement proponents attempted to leverage watershed-level social capital among religious communities (e.g., for determining agreement sectors) and in the açaí economy (e.g., building on NGO relationships from açaí projects), but the social mobilization of the fisheries agreement process proved ephemeral. Aspects of the açaí economy that may have facilitated collective action are missing in the fisheries sector, namely, the commercial importance of açaí to household incomes and the presence of an external opponent (açaí middlemen) against whom to organize. However, more research is needed into how social capital may transfer across resource systems and how social capital intersects with scale to determine SRSG outcomes.

CONCLUSION

In the Brazilian Amazon, state-centric fisheries governance promoted open access principles without effectively monitoring and enforcing corresponding regulations, leading to social conflict and resource degradation. Community fisheries agreements first arose through grassroots collective action to transform systems degraded by commercial extraction, and in interaction with state agencies these community initiatives spawned a new SRSG legal framework for fisheries co-management. Community fisheries agreements have been effective in numerous cases in reducing conflict, protecting fish stocks, and supporting Ribeirinho livelihoods. Reported improvements in fisheries resources after the Canaticu agreement corroborate other studies reporting positive effects of fisheries agreements on fishery sustainability (McGrath et al. 1998, de Castro and McGrath 2001, Aquino 2007, Almeida et al. 2009, Gurdak et al. 2019).

State-reinforced self-governance is a central feature of transformative and adaptive governance of complex, multilevel resource systems, but our understanding of legal and institutional design principles for effective SRSG is still nascent. We have highlighted Amazonian fisheries agreements as important cases for testing and refining SRSG principles, and our research on the Canaticu Fisheries Agreement gives new insights into fisheries agreements and SRSG. In particular, iterative decision making requires ongoing collective mobilization, and maintaining collective mobilization in multilevel governance systems demands enduring tangible support from state and non-state actors. To ensure tangible support for SRSG, legally binding responsibilities must be linked with accountability mechanisms, which for Amazonian fisheries agreements might include long-term budgetary commitments from state and NGO partners, greater use of virtual communication technologies for linking remote areas with state authorities, and oversight by the Public Ministry, a government body of independent prosecutors that helps guarantee environmental law enforcement (McAllister 2005).

The need for long-term external support to maintain self-governance in the Canaticu case highlights a mismatch between social-ecological system transformation and the temporal and spatial dynamics of the government and NGO project cycle. The project cycle emphasizes discrete outputs “at scale,” such as a fisheries agreement for the entire Canaticu watershed concluded within two years as an output of the Marajó Viva Pesca project. Transformative governance is not a product, however, but a process that requires ongoing investment in social capital across multiple levels. Accordingly, SRSG initiatives and analyses must give special attention to questions of scale and social capital. To move beyond a pathology of state-reinforced transformative governance agreements without state-reinforced implementation, state and NGO actors must adopt a more incremental, process-oriented approach to participatory co-management that maintains long-term partnerships and strengthens social capital and collective mobilization at locally relevant scales to achieve more adaptive and enduring state-reinforced social-ecological transformations.

LITERATURE CITED

Acheson, J. M. 1988. The lobster gangs of Maine. University Press of New England, Hanover, New Hampshire, USA. https://doi.org/10.2307/j.ctv1xx99v9

Almeida, O. T. 2006a. A indústria pesqueira na Amazônia. ProVárzea/IBAMA, Manaus, Brazil. https://apps.oraotca.org/aro/documentos/files/DR%20EE%20Industria%20Pesqueira.pdf?token=dGVzdGNsaWVudDp0ZXN0c2VjcmV0

Almeida, O. T. 2006b. Introdução. Pages 17–24 in O. T. Almeida, editor. Manejo de pesca na Amazônia Brasileira. Peirópolis, São Paulo, Brazil.

Almeida, O. T., K. Lorenzen, and D. G. McGrath. 2009. Fishing agreements in the lower Amazon: for gain and restraint. Fisheries Management and Ecology 16:61–67. https://doi.org/10.1111/j.1365-2400.2008.00647.x

Almeida, S. S., D. D. Amaral, and A. S. L. Silva. 2004. Análise florística e estrutura de florestas de Várzea no estuário amazônico. Acta Amazonica 34(4):513–524. https://doi.org/10.1590/S0044-59672004000400005

Aquino, A. S. 2007. Avaliação do processo de discussão e implementação dos acordos de pesca no estado do Amazonas. Relatório Pró-Várzea. IBAMA, Manaus, Brazil.

Arantes, C. C., L. Castello, X. Basurto, N. Angeli, A. Sene-Haper, and D. G. McGrath. 2022. Institutional effects on ecological outcomes of community-based management of fisheries in the Amazon. Ambio 51:678–690. https://doi.org/10.1007/s13280-021-01575-1

Bebbington, A. 1997. Social capital and rural intensification: local organizations and islands of sustainability in the rural Andes. Geographical Journal 163(2):189–197. https://doi.org/10.2307/3060182

Begossi, A., P. H. May, P. F. Lopes, L. E. C. Oliveira, V. da Vinha, and R. A. M. Silvano. 2011. Compensation for environmental services from artisanal fisheries in the SE Brazil: policy and technical strategies. Ecological Economics 71(15):25–32. https://doi.org/10.1016/j.ecolecon.2011.09.008

Benatti, J. H., D. G. McGrath, and A. C. M. Oliveira. 2003. Políticas públicas e manejo comunitário de recursos naturais na Amazônia. Ambiente e Sociedade 6(2):13–154. https://doi.org/10.1590/S1414-753X2003000300009

Braun, V., and V. Clarke. 2006. Using thematic analysis in psychology. Qualitative Research in Psychology 3(2):77–101. https://doi.org/10.1191/1478088706qp063oa

Brondízio, E. S., E. Ostrom, and O. R. Young. 2009. Connectivity and the governance of multilevel social-ecological systems: the role of social capital. Annual Review of Environment and Resources 34:253–278. https://doi.org/10.1146/annurev.environ.020708.100707

Brondízio, E. S., and A. D. Siqueira. 1997. From extractivists to forest farmers: changing concepts of caboclo agroforestry in the Amazon estuary. Research in Economic Anthropology 18:233–279.

Brown, J. C., and M. Purcell. 2005. There’s nothing inherent about scale: political ecology, the local trap, and the politics of development in the Brazilian Amazon. Geoforum 36:607–624. https://doi.org/10.1016/j.geoforum.2004.09.001

Campos-Silva, J. V., and C. A. Peres. 2016. Community-based management induces rapid recovery of a high-value tropical freshwater fishery. Scientific Reports 6:34745. https://doi.org/10.1038/srep34745

Castello, L., D. G. McGrath, C. C. Arantes, and O. T. Almeida. 2013. Accounting for heterogeneity in small-scale fisheries management: the Amazon case. Marine Policy 38:557–565. https://doi.org/10.1016/j.marpol.2012.09.001

Castello, L., J. P. Viana, and M. Pinedo-Vasquez. 2011. Participatory conservation and local knowledge in the Amazon Várzea: the Pirarucu Management Scheme in Mamirauá. Pages 261–275 in M. Pinedo-Vasquez, M. L. Ruffino, C. Padoch, and E. S. Brondízio, editors. The Amazon Várzea: the decade past and the decade ahead. Springer, New York, New York, USA. https://doi.org/10.1007/978-94-007-0146-5_18

Cavalcante, R. B., P. Calixto, and M. M. K. Pinheiro. 2014. Análise de conteúdo: considerações gerais, relações com a pergunta de pesquisa, possibilidades e limitações do método. Informação e Sociedade: Estudos 24(1):13–18.

Cerdeira, R. G. P., and S. A. F. Camargo. 2007. Gestão participativa da pesca na região do Maicá, em Santarém, PA: reflexões juríd́icas e ambientais. Anais do XVI Congresso Nacional do CONPEDI, Belo Horizonte, 4489–4507. http://www.publicadireito.com.br/conpedi/manaus/arquivos/anais/bh/regina_gloria_pinheiro_cerdeira.pdf

Chaffin, B. C., A. S. Garmestani, L. H. Gunderson, M. H. Benson, D. G. Angeler, C. A. Tony, B. Cosens, R. K. Craig, J. B. Ruhl, and C. R. Allen. 2016. Transformative environmental governance. Annual Review of Environment and Resources 41:399–423. https://doi.org/10.1146/annurev-environ-110615-085817

Costa, P. M. 2010. Acordos de pesca: desafios de implementação e consolidação em áreas de várzea do município de Gurupá, Pará, Brasil. Thesis. Universidade Federal do Pará, Belém, Brazil. https://repositorio.ufpa.br/bitstream/2011/3502/1/Dissertacao_AcordosPescaDesafios.pdf

Costa, P. M. 2017. Um pesqueiro real: entre acordos e manejos na RESEX Marinha de Soure. Dissertation. Universidade Federal do Pará, Belém, Brazil. https://drive.google.com/file/d/1rcgygOtxZ_LxS_IU60sPjO7vNqHGtyvx/view

Cox, M., G. Arnold, and S. Villamayor Tomás. 2010. A review of design principles for community-based natural resource management. Ecology and Society 15(4):38. https://doi.org/10.5751/ES-03704-150438

D’Almeida, B. G. 2006. Os acordos de pesca na Amazônia: uma perspectiva diferenciada de gestão das águas. In XV encontro preparatório do CONPEDI, Recife, Brazil. http://www.publicadireito.com.br/conpedi/manaus/arquivos/anais/recife/direito_ambiental_bruna_almeida.pdf

DeCaro, D. A., B. C. Chaffin, E. Schlager, A. S. Garmestani, and J. B. Ruhl. 2017. Legal and institutional foundations of adaptive environmental governance. Ecology and Society 22(1):32. https://doi.org/10.5751/ES-09036-220132

de Castro, F. 2002. From myths to rules: the evolution of local management in the Amazon Floodplain. Environment and History 8(2):197–216. https://doi.org/10.3197/096734002129342648

de Castro, F., and D. McGrath. 2001. O manejo comunitário de lagos na Amazônia. Parcerias Estratégicas 12:112–126. https://ipam.org.br/wp-content/uploads/2019/07/o_manejo_comunit%C3%A1rio_de_lagos_na_amaz%C3%B4ni.pdf

de Oliveira, H. J. S. 2018. Redes, aprendizados e ativos territoriais na Amazônia: o processo de qualificação do açaí do rio Canaticu, arquipélago do Marajó, Pará. Thesis. Universidade Federal do Pará, Belém, Brazil. https://repositorio.ufpa.br/jspui/handle/2011/12195

Dietz, T., E. Ostrom, and P. C. Stern. 2003. The struggle to govern the commons. Science 302(5652):1907–1912. https://doi.org/10.1126/science.1091015

dos Santos, M. T. 2005. Aprendizados do projeto de manejo dos recursos naturais da várzea - ProVárzea. Ministério do Meio Ambiente, Brasília, Brazil. https://apps.oraotca.org/aro/documentos/files/Série%20Estudos%2004%20-%20ProVarzea%20(AMA).pdf?token=dGVzdGNsaWVudDp0ZXN0c2VjcmV0

Emerson, R. M., R. I. Fretz, and L. L. Shaw. 2011. Writing ethnographic fieldnotes. Second edition. University of Chicago Press, Chicago, Illinois, USA. https://doi.org/10.7208/chicago/9780226206851.001.0001

Fassoni-Andrade, A. C., F. Durand, A. Azevedo, X. Bertin, L. G. Santos, J. U. Khan, L. Testut, and D. M. Moreira. 2023. Seasonal to interannual variability of the tide in the Amazon estuary. Continental Shelf Research 255:104945. https://doi.org/10.1016/j.csr.2023.104945

Furtado, L. G., W. Leitão, and A. F. Mello. 1993. Povos das Águas: realidade e perspectivas na Amazônia. Museu Paraense Emílio Goeldi, Belém, Brazil. https://repositorio.museu-goeldi.br/bitstream/mgoeldi/777/1/POVOS%20DAS%20AGUAS%201993%20ALENCAR.pdf

Furtado, L. F. 2006. Origens pluriétnicas no cotidiano da pesca na Amazônia: contribuições para projeto de estudo pluridisciplinar. Boletim do Museu Paraense Emílio Goeldi. Ciências Humanas 1(2):159–172. https://doi.org/10.1590/S1981-81222006000200013

Gelcich, S., T. P. Hughes, P. Olsson, C. Folke, O. Defeo, M. Fernández, S. Foale, L. H. Gunderson, C. Rodríguez-Sickert, M. Scheffer, R. S. Steneck, and J. C. Castilla. 2010. Navigating transformations in governance of Chilean marine coastal resources. Proceedings of the National Academy of Sciences 107(39):16794–16799. https://doi.org/10.1073/pnas.1012021107

Gillingham, S. 2001. Social organization and participatory resource management in Brazilian Ribeirinho communities: a case study of the Mamirauá Sustainable Development Reserve, Amazonas. Society and Natural Resources 14(9):803–814. https://doi.org/10.1080/089419201753210611

Gurdak, D. J., C. C. Arantes, L. Castello, D. J. Stewart, and L. C. Watson. 2019. Evidence of recoveries from tropical floodplain fisheries: three examples of management gains for South American giant Arapaima. Pages 267–295 in C. C. Krueger, W. W. Taylor, and S.-J. Youn, editors. From catastrophe to recovery: stories of fishery management success. American Fisheries Society, Bethesda, Maryland, USA. https://doi.org/10.47886/9781934874554.ch11

Huitema, D., E. Mostert, W. Egas, S. Moellenkamp, C. Pahl-Wostl, and R. Yalcin. 2009. Adaptive water governance: assessing the institutional prescriptions of adaptive (co-)management from a governance perspective and defining a research agenda. Ecology and Society 14(1):26. https://doi.org/10.5751/ES-02827-140126

Instituto Peabiru. 2016. “O peconheiro”: diagnóstico das condições de trabalho do extrativista de açaí. Instituto Peabiru, Belém, Brazil. https://institutopeabiru.files.wordpress.com/2017/09/160915-o-peconheiro-diagnostico-acai.pdf

Jimenez, E. A., R. S. L. Barboza, M. T. Amaral, and F. L. Frédou. 2019. Understanding changes to fish stock abundance and associated conflicts: perceptions of small-scale fishers from the Amazon coast of Brazil. Ocean and Coastal Management 182:104954. https://doi.org/10.1016/j.ocecoaman.2019.104954

McAllister, L. K. 2005. Public prosecutors and environmental protection in Brazil. Pages 207–229 in A. Romero and S. E. West, editors. Environmental issues in Latin America and the Caribbean. Springer, Dordrecht, The Netherlands. https://doi.org/10.1007/1-4020-3774-0_9

McGrath, D. G., L. Castello, O. T. Almeida, and G. M. B. Estupiñán. 2015. Market formalization, governance, and the integration of community fisheries in the Brazilian Amazon. Society and Natural Resources 28(5):513–529. https://doi.org/10.1080/08941920.2015.1014607

McGrath, D., F. Castro, E. Câmara, and C. Futemma. 1998. Manejo comunitário de lagos de várzeas e o desenvolvimento sustentável da pesca na Amazônia. Novos Cadernos NAEA 1(2):1–23. http://dx.doi.org/10.5801/ncn.v1i2.4

McGrath D. G., F. Castro, C. Futemma, B. D. de Amaral, and J. Calabria. 1993. Fisheries and evolution of resource management on the lower Amazon floodplain. Human Ecology 21:167–195. https://doi.org/10.1007/BF00889358

Mello, A. F. 1985. A pesca sob o capital: a tecnologia a serviço da dominação. Universidade Federal do Pará, Belém, Brazil.

Michelat, G. 1987. Sobre a utilização de entrevista não diretiva em sociologia. Pages 191–212 in M. Thiolent. Crítica metodológica, investigação social e enquete operária. Fifth edition. Polis, São Paulo, Brazil.

Montagnini, F., and N. Muñiz-Miret. 1999. Vegetation and soils of tidal floodplains of the Amazon estuary: a comparison of várzea and terra firme forests in Pará, Brazil. Journal of Tropical Forest Science 11(2):420–437. https://repositorio.catie.ac.cr/bitstream/handle/11554/4309/Vegetation_and_soils_of_tidal_floodplains.pdf?sequence=1&isAllowed=y

Murrieta, R. S. S., D. L. Dufour, and A. D. Siqueira. 1999. Food consumption and subsistence in three Caboclo populations on Marajó Island, Amazonia, Brazil. Human Ecology 27(3):455–475. https://doi.org/10.1023/A:1018779624490

Mwangi, E., and A. Wardell. 2012. Multi-level governance of forest resources. International Journal of the Commons 6(2):79–103. https://doi.org/10.18352/ijc.374

Oliveira, V. M. S. 2012. Governança territorial: diretrizes para o fortalecimento do núcleo gestor da pesca do Rio Canaticu, município de Curralinho. Thesis. Universidade Federal do Pará, Belém, Brazil. https://repositorio.ufpa.br/jspui/bitstream/2011/11344/1/Dissertacao_GovernancaTerritorialDiretrizes.pdf

Olsson, P., V. Galaz, and W. J. Boonstra. 2014. Sustainability transformations: a resilience perspective. Ecology and Society 19(4):1. https://doi.org/10.5751/ES-06799-190401

Ostrom, E. 1990. Governing the commons: the evolution of institutions for collective action. Cambridge University Press, New York, New York, USA. https://doi.org/10.1017/CBO9780511807763

Ostrom, E. 2009. A general framework for analyzing sustainability of social-ecological systems. Science 325(5939):419–422. https://doi.org/10.1126/science.1172133

Oviedo, A. F. P., M. Bursztyn, and J. A. Drummond. 2015. Agora sob nova administração: acordos de pesca nas várzeas da Amazônia brasileira. Ambiente e Sociedade 18(4):119–138. https://doi.org/10.1590/1809-4422ASOC985V1842015

Pinho, P. F., B. Orlove, and M. Lubell. 2012. Overcoming barriers to collective action in community-based fisheries management in the Amazon. Human Organization 71(1):99–109. https://doi.org/10.17730/humo.71.1.c34057171x0w8g5p

Potiguar, M. R. S., and P. M. Costa. 2015. Viva pesca: guia de construção dos acordos de pesca do Rio Canaticu. Instituto Peabiru, Belém, Brazil. https://institutopeabiru.wordpress.com/wp-content/uploads/2014/01/viva_pesca_guia_de_construcao_de_acordos_-de_pesca_canaticu_web.pdf

Raseira, M. B., E. P. L. Câmara, and M. L. Ruffino. 2006. Gestão participativa dos recursos pesqueiros na várzea amazônica. Agriculturas 3(1):32–35. https://aspta.org.br/files/2014/10/Artigo-9-Gest%C3%A3o-participativa-dos-recursos-pesqueiros-na-v%C3%A1rzea-amaz%C3%B4nica.pdf

Resilience Alliance. 2010. Assessing resilience in social-ecological systems: workbook for practitioners. Revised version 2.0. Resilience Alliance. http://www.resalliance.org/files/ResilienceAssessmentV2_2.pdf

Santos, V. B. 2014. Acordos de pesca e “capital social Ribeirinho” na Amazônia Oriental: a dimensão cultural em concepções contemporâneas de desenvolvimento e sustentabilidade. Anais do 38^o Encontro Anual da ANPOCS, Caxambú, Brazil.

Santos, V. B. 2019. Desenvolvimento territorial e capital social em acordos de pesca na Amazônia Oriental. Cadernos Cepec 8(1):67–88. http://dx.doi.org/10.18542/cepec.v8i1.7765

Sarker, A. 2013. The role of state-reinforced self-governance in averting the tragedy of the irrigation commons in Japan. Public Administration 91(3):727–743. https://doi.org/10.1111/padm.12011

Seixas, C. S., and D. C. Kalikoski. 2009. Gestão participativa da pesca no Brasil: levantamento das iniciativas e documentação dos processos. Desenvolvimento e Meio Ambiente 20:119–139. http://dx.doi.org/10.5380/dma.v20i0.12729

Smith, N. J. H. 1985. The impact of cultural and ecological change on Amazonian fisheries. Biological Conservation 32(4):355–373. https://doi.org/10.1016/0006-3207(85)90023-0

Thaler, G. M. 2021. Ethnography of environmental governance: towards an organizational approach. Geoforum 120:122–131. https://doi.org/10.1016/j.geoforum.2021.01.026

Vidal, M. D. 2010. Manejo participativo da pesca na Amazônia: a experiência do ProVárzea. Ciência e Natura 32(2):97–120. https://periodicos.ufsm.br/cienciaenatura/article/view/9488/5637