More than funders: The roles of philanthropic foundations in marine conservation governance

Environmental governance scholars have overlooked philanthropic foundations as influential non‐state actors. This omission, along with the continued growth in funding from private foundations for conservation issues, presents important questions about what foundations do in governance spaces. To address this gap, we examine The David and Lucile Packard Foundation's involvement in Fiji and Palau in the context of the Foundation's “Western Pacific Program”—a series of coastal and marine‐related investments made from 1998 to 2020. We describe and analyze six governance roles that the Packard Foundation contributed to: funding, influencing agendas, capacity‐building, convening and coordinating, facilitating knowledge, and rule‐making and regulation. In documenting the Packard Foundation's governance roles, we provide scholars and practitioners a conceptual framework to more systematically and strategically think about foundations as more than funders. This research helps move the conversation around conservation philanthropy beyond binary conceptions of “good” versus “bad,” and, instead, toward deeper considerations about what foundations currently do within governance systems, how they engage with diverse practitioners, as well as what they can and should do to advance conservation goals.


| INTRODUCTION
With considerable financial resources and unparalleled organizational autonomy, philanthropic foundations are important-yet understudied-players in the environmental sector (Delfin & Tang, 2006;Rogers, 2015). Philanthropic funding for marine conservation has grown significantly in recent years, from US$252 million in 2010 (California Environmental Associates, 2017 to over US$615 million in 2016 (Berger et al., 2019). Marine conservation grants from philanthropic sources and official development assistance (ODA) were roughly equal in 2015(California Environmental Associates, 2021. While total financial flows (grants, loans, and export credits) of ODA still exceed philanthropic funding for marine conservation (California Environmental Associates, 2018), the continued upward trend in marine philanthropy-an estimated US$8 billion over the past 10 years (Sumaila et al., 2020)-demonstrates foundations' expanding role in marine conservation around the world Wabnitz & Blasiak, 2019).
Scholars have paid limited attention to philanthropic foundations as environmental governance actors , a significant omission given foundations' increasing prominence in the environmental arena. Environmental governance refers to the institutions, structures, and processes through which decisions are made about the environment (Bennett & Satterfield, 2018). Scholars and practitioners acknowledge that governing complex environmental challenges involves a diverse range of actors, including local, regional, and national governments; non-governmental organizations (NGOs); companies and business associations; international organizations; and scientific networks (Armitage et al., 2012;Lemos & Agrawal, 2006;Newell et al., 2012). Foundations are unique among environmental governance agents in that they are situated in a "hybrid" sphere that operates across and between the state, market, and civil society (Spierenburg & Wels, 2010). In addition, they have considerable autonomy compared to other governance agents in that they are not accountable to voters, consumers, or shareholders and are financially selfsufficient (Quinn et al., 2014;Reich, 2016). Jung and Harrow (2015) contend that this might allow foundations to exercise "hyper-agency." In this paper, we draw on the environmental governance literature and an empirical case study to explore what foundations do as governance agents, "where an agent is an actor that steers institutions, structures, and processes through which decisions are made about the environment" (Betsill et al., 2021, p. 2). Specifically, we seek to make visible the roles they play in delivering diverse governance functions. In a recent review of the environmental governance literature, Betsill and Milkoreit (2020) identified 20 different functions, including providing funding; making, implementing, and enforcing formal and informal rules; generating knowledge and sharing information; setting goals; and establishing norms. Different types of agents contribute to the performance of these functions in different ways. For example, while states often have final authority to make formal rules and regulations, businesses, NGOs, and Indigenous communities can influence the rule-making function through lobbying state decision-makers, staging protests, and mobilizing social movements (Hale & Roger, 2014;Marion Suiseeya et al., 2022). Nasiritousi et al. (2016) contend that an actor's role in environmental governance is linked to the power resources they can leverage as well as recognition from other agents in the environmental governance system. Foundations' position in the global political economytheir operational autonomy, hybrid state-civil-market agency, and financial self-sufficiency-may contribute to unique roles as environmental governance agents.
Public and academic debates about philanthropic foundations tend to be highly polarized and simplistic: foundations are cast as either malevolent drivers of neoliberal values and agendas or righteous do-gooders who can save the world . Given the complexity and urgency of environmental problems today, and inadequacy of funding to achieve effective conservation (Xu et al., 2021), we argue that there is a need to move beyond binary discussions of foundations as "good" or "bad" and toward more deliberative discussions about why they exist, what they can and cannot do, and what they should do. Interrogating seemingly disparate discourses about philanthropy-that the institution solidifies plutocratic rule through an "exercise of wealthderived power" with "minimal democratic controls" (Barkan, 2013, p. 635), versus the belief that foundation money is "donated to help the less fortunate through motives of pure altruism" (Fleishman, 2009, p. 48)-can foster tough, honest conversations that recognize the inherent contradictions and trade-offs of philanthropy (Curnow & Helferty, 2018;Delfin & Tang, 2008).
We employ the conceptual toolbox of environmental governance to deepen the conversation around how private foundations leverage their unique position to address complex issues. Specifically, we examine The David and Lucile Packard Foundation's (hereafter Packard) role in Fiji and Palau in the context of what the Foundation calls their "Western Pacific Program"-a series of coastal and marine related investments made across eight countries in the Asia-Pacific region. In Fiji between 1998 and 2017, Packard invested $11,952,000 through 71 grants to 22 organizations for conservation work. The Foundation invested $8,168,000 through 52 grants to 13 organizations in Palau over a similar period. We describe and analyze six governance roles that Packard contributed to in Fiji and Palau during this period: funding, influencing agendas, capacity-building, convening and coordinating, facilitating knowledge, and rule-making and regulation. In so doing, we make three contributions. First, we make an empirical case for environmental governance scholars to pay more attention to philanthropic foundations as influential governance agents. Second, we provide scholars and practitioners a language to more systematically and strategically reflect on the range of roles that foundations can conceivably play to achieve public goals, and with what effects; in other words, to conceptualize foundations as more than funders when they intervene in conservation. Lastly, we believe our research promotes a shared understanding of what foundations do from which donors, grantees, scholars, and other stakeholders can deliberatively engage in pressing normative questions, like: what is the right role of foundation money in marine conservation governance?

| METHODOLOGY
This study focuses on governance roles, which emerged as a major research theme in our 2018 participatory research design process . We focus on Packard's Western Pacific Program work in Fiji and Palau because of our co-authors' experience in those countries and Packard's specific interest in understanding the impact of their long-term engagement and recent exit there. We acknowledge that our own and interviewees' multi-faceted relationship with Packard as both funder and research subject has the potential to introduce real and/or perceived bias into this study. We have managed this with the help of a research advisory committee tasked with holding us accountable in pursuing culturally appropriate research, with a multi-faceted and balanced perspective on foundations. Through biannual meetings, committee members provided feedback on the relevance and usefulness of this paper's topic, the theoretical conceptualization of the paper, and its results. One committee member also reviewed and provided helpful feedback on an early draft of this paper.
Our analysis draws on 65 semi-structured interviews conducted between 2018 and 2021 with 8 current or former Packard staff members (referred to as "donors" herein) and 69 marine conservation practitioners from diverse sectors in Fiji and Palau (Table 1 and Appendix S1). Semi-structured interviews were preferred over other methods because it allowed a diverse range of governance roles to emerge from the data and provided interviewees the space to elaborate on how each role was performed (Kvale, 2013). Eleven of the individuals interviewed were non-grantees. We supplemented interview data with published, publicly available reports and internal documents that summarize the scope and scale of Packard's investments in marine conservation. We conducted thematic analysis of detailed notes and verbatim interview transcripts using QSR NVivo, coding for governance roles drawing on Betsill and Milkoreit (2020), while allowing for additional roles to emerge from the data. We refined and condensed codes through discussions between co-authors and with Packard staff to arrive at six primary governance roles.
We do not assume that our findings can be generalized to all foundations operating in different governance arenas. In fact, practitioners in Fiji and Palau routinely observed that Packard is a unique funder in the ways it interacts with grantees and other stakeholders. Rather, we aim to use the Packard case for analytic generalization where "researchers strive to generalize from particulars to broader constructs or theory" (Polit & Beck, 2010, p. 1453). Our goal is to lay the groundwork for future studies and reflections about the roles foundations play as more than funders-and can and should play-in their contributions to governance functions.

| PACKARD'S GOVERNANCE ROLES
Our analysis revealed six governance roles that Packard played in Fiji and Palau (

Rule-making and regulation
Developing formal and informal rules and institutions for environmental governance specific functions identified in the environmental governance literature. We do not mean to imply that Packard's was solely responsible for each function, nor do we believe this is an exhaustive analysis of the Foundation's roles. Instead, we analyze what we found to be the primary ways that respondents spoke about Packard contributions to delivering different types of functions through complex governance roles that evolved over time. While funding may have been Packard's most obvious role, the Foundation leveraged its "more-than-financial" powers to perform additional roles-evidence of foundations' unique and dynamic relationship to the state, market, and civil sector in environmental governance (Beer, 2022).

| Funding conservation
Foundations assume their agency primarily through grantmaking . This section discusses how Packard used grantmaking to fill systemic gaps in access to material resources, specifically money. We begin by centering the influence of colonialism on traditional management systems in Fiji and Palau, and, most recently, how "imported" Western governance, political, and economic systems created conditions for a funding vacuum that enabled Packard's role as a conservation funder (Carlisle & Gruby, 2019b, p. 529). We then describe how Packard filled a contemporary gap, noting that the Foundation's grantmaking approach was unique relative to other types of funders. The evolution of village life in Fiji, from initial European contact in 1643 to political independence in 1970, coincides with a complex transfer of decision-making power from hereditary chiefs and community councils to government officials and judiciary (Frazer, 1973). Today, iTaukei (Indigenous Fijians) control fishing rights in I'qoliqoli areas (traditionally managed fishing grounds), while the state maintains ownership of the seabed and overlying waters (Jupiter & Egli, 2011). Veitayaki et al. (2011, p. 40) write, "While traditional roles and resource-use systems within the communities are still well defined, leadership structures, protocol, respect, practices and beliefs are changing and are increasingly questioned by the people." In Palau, responsibility for coastal management has also evolved. Over nearly 100 years of foreign occupation and their transition to independence in 1994, Palau's governance arrangement has moved toward a more diverse, "polycentric" system with an increasing reliance on government and other agents, such as NGOs (Carlisle & Gruby, 2019a). Today, in the context of Palau's transition to a globalized market economy and "hybrid" form of traditional/democratic governance, "many responsibilities that were once community concerns are now viewed as the job of government" (Carlisle & Gruby, 2018, p. 232). With mixed emotions, practitioners in Palau acknowledge how conservation has become increasingly professionalized and monetized: Now there's an expectation of money for doing conservation, when it used to be just sort of a traditional conservation, an ethic, that we all do as part of our service to our community. (Palau 035) Within Palau and Fiji's post-colonial 1 context, limited funds can be a primary barrier for professional conservation work. There were few private conservation donors when Packard began working in Palau in 1998, while there was far more private investment in Fiji at the time (Packard Foundation, 2018). Even so, practitioners in both countries emphasized that Packard was a source of money when government or other funders were limited and, in some cases, helped overcome initial financial barriers to pursue conservation projects. In Fiji, this included significant investment in efforts to strengthen customary and Indigenous management of I'qoliqolis through the Fiji Locally Managed Marine Area network. In Palau, for example, Packard funded long-term ecological monitoring of one of their most important tourist sites (Jellyfish Lake); something that the government was unable or unwilling to do: …resources are very scarce. And there's not a lot of public funding in the government to fund work in Palau. I think the funding from private donors are very helpful in terms of making sure that the work on the ground is sufficiently funded. (Palau 037) In contrast with other donors, such as international aid sectors and some other foundations, Packard's funding role was unique in two significant ways. First, Packard has had a consistent commitment to funding over a 22-year period, which enabled practitioners to pursue more ambitious conservation goals over longer timeframes because they knew they had the necessary funding to do so: But the other thing with Packard, their funding is consistent, it doesn't just end within timeframes. We're not limited to deliver on 2-3 year timeframes like other foundations because we know Packard is going to consistently support the process. (Fiji 069) Second, practitioners expressed how Packard funding required less administrative work, including fewer and less complicated project reports. In other words, Packard funding was not as "tie[d] … up in bureaucracy" (Fiji 033). This enabled smaller, grassroots organizations with less administrative capacity to access conservation funding. In addition, Packard funding was described as more flexible, adaptable, and allowed grantees to take risks with their projects where other funders, including public funds from the government, multilateral aid from Global Environmental Facility, and development assistance from the World Bank and United Nations, remained more conservative and rigid: I think that's the beauty of private funders. I think they take a lot of risks in doing the stuff that has been innovative and hasn't been done before. So, for the Northern Reef, I think we piloted the stock assessment analysis. And Packard was willing to fund that work to make sure that we had baseline information to understand what was happening to the resources in the north. And then pilot this project in the Northern Reef even though we weren't quite sure if it was going to work. But with the community's support, it worked out pretty well. And I think without Packard, taking that risk, I don't think we would have done that work. (Palau 037) While Packard's grantmaking approach expanded opportunities for grantees to pursue conservation work, their role as a funder is embedded within broader structural and cultural shifts around who does conservation work-and why. Today, customary marine tenure systems and associated communities in both countries are embedded in broader networks of formal state policies and philanthropic-backed, NGO-supported conservation initiatives (Carlisle & Gruby, 2018;Johannes, 1981;Useem, 1950). In Palau, these shifts raised questions about the right role for money in conservation. What are the consequences of foundation-supported conservation for customary or community-based conservation that historically relied on an ethic of responsibility and Indigenous land and sea relations rather than money? Is foundation funding filling a need it helped create? It is beyond our scope to answer such questions here, but they should be considered in future work.

| Influencing conservation agendas
Packard's role in influencing conservation agendas refers to how they shaped and/or supported the priorities and goals of agents across multiple social and geographic scales (see also Enrici et al., in review). Packard's influence involved a dynamic relationship between program staff and grantees wherein priorities were set through an emergent collaborative process as well as sometimes more strategic, top-down relations of decision-making where the Foundation or program officers played a heavier hand in which conservation strategies and projects were pursued.
Most practitioners spoke positively about how Packard involved grantees in the grantmaking process, especially compared to other foundations. For example, Packard conducted multiple in-country brainstorming meetings that encouraged open and frequent communication between Foundation staff and grantees, as well as government officials, during the initial planning and development phase of the Western Pacific Program. One donor discussed how, while Packard considers itself a strategic funder-where clearly defined goals, set by the Foundation, define the priorities and scope of grant work-they also do "things that are kind of the opposite of strategic philanthropy … like find really outstanding people, [and] give them unrestricted funding just to kind of pursue a strategy that they think could be a breakthrough" (Donor 090).
Importantly, this process sometimes operated alongside less collaborative approaches where the Foundation leveraged its financial power to influence grantees' priorities and agendas. For example, in both countries, Packard added fisheries management to their long-standing focus on area-based conservation in 2012. In response, some grantees and other practitioners reported shifting their priorities or re-framing their work to access grant money. However, even this was not simply a case of the Foundation dictating priorities and agendas. Respondents pointed out that the health and sustainability of local fisheries had always been culturally, politically, and economically important. While some grantees had concerns about Packard's shift in strategy and its effect on their work, others described using the money to act on contextually relevant issues as they negotiated the contours of the fisheries program with Packard. One practitioner described the negotiations related to gender: But the stuff that I was quite pleased about is the gender in fisheries work because [Packard] didn't see the relevance initially and had asked me to take that out. But actually, it was one where I kind of stood my ground and just kept arguing back, why I thought it was important. Gender matters when it comes to fisheries-it's not just about the fish. (Fiji 033) Grantees also reported that Packard played an important role in elevating marine conservation within national-level policy agendas by aligning funding with existing national and/or grantee objectives. This alignment developed through frequent field visits, long-term relationships, and collaborative forums. For example, one practitioner described how Packard funding supported, rather than redirected, the existing agenda of protected areas work for the Palau Conservation Society (PCS), a local NGO primarily staffed by Palauan nationals. Here, Packard collaborated with PCS to understand the organization's priorities for ongoing protected area work around the Rock Islands; together, they co-developed grants that helped establish an effective conservation model for a national network of protected areas. By aligning grants with grantee agendas, Packard funding supported the development of a national agenda around protected areas that was eventually codified into national legislation-Palau's Protected Areas Network (PAN): …it's really helped shape where the Protected Areas sort of model is now because it really supported the work that was done, for a long time and for different aspects of the work that the PCS did at the time … I think [Packard funding] helps sort of provide support for protected areas that were really ready and active at the time. (Palau 050) Mallin et al. (2019) argue that US-based philanthropies, like the Packard Foundation, have been performing an agenda-setting role in ocean conservation since the early 2000s. However, a lack of research on how agendas are set has contributed to relatively binary conclusions that situate foundations as either coercive agents that coopt agendas, or, passive bystanders that simply provide money to do conservation work . Instead, we found that the Packard Foundation's involvement in agenda-setting in Fiji and Palau is a complex, dialectical relationship between grantees and donors where conservation agendas are co-constructed and acted upon in context-specific ways within the Foundation's bigger-picture priorities, programmatic strategies and ultimate decision-making power. The case demonstrates the importance of more closely investigating how foundations' grantmaking style and donor-grantee relationships can and should influence agendas, priorities, and goals throughout different stages of the grantmaking process.

| Capacity-building
Packard performed a capacity-building role through the development and investment in individuals' and organizations' skillsets, including technical knowledge, networks and partnerships, communication, and fundraising, ultimately contributing to the development of a professionalized civil society in the marine conservation governance arena that spilled over into the public sector. The Foundation's goal was to address an understood lack in information, capacity, money, and infrastructure by building the skills of local conservation practitioners who could support effective, and sustainable marine management.
At an individual level, the Foundation funded conservation practitioners to attend workshops and skills trainings. Trainings focused on foundational skills like project management and funding procurement, as well as skills in data collection and analysis, like scoping economic assessments to optimize marine resources for sustainable management. From this, some individuals emerged as leaders in the field, employing their skills and expertise to address gaps in local management, like conducting fishery stock assessments, and influence decision-making at national and sub-national levels. Overall, Packard funded individual capacity-building across organizations and sectors, supporting a network of local, professional conservation practitioners that could advance sustainable marine management: [Packard] facilitated this network of people that even after their exit, they've created this network of individuals at different organizations, not just in Palau, but I think in Micronesia, of really capable individuals … these people are now, one of them is our Minister of Environment. So, they produce these key leaders or key champions for marine conservation moving forward. (Palau 070) At an organizational level, Packard funding supported the development of local conservation NGOs and networks. A good example is the Palau Conservation Society (PCS). PCS used grants from Packard's Organizational Effectiveness Program-a program dedicated to capacity development-to develop new strategic plans for accomplishing management goals, build their technical expertise, and hone skills like managing budgets and personnel. Packard funding supported PCS to bring in consultants, hire new staff members and support staff through leadership trainings. As a result, PCS broadened their scope and capabilities, as well as built "their reputation up as an organization" (Palau 044). Perhaps most importantly, practitioners discussed how Packard funding enhanced the confidence of PCS as an organization. Rather than relying on outside expertise, PCS built on their conservation work from within their organization and local communities. After Packard funding ended, they had the reputation and administrative expertise to apply for and access diverse funding mechanisms: [Packard funding] has helped PCS grow and has helped empower PCS … Packard has given us confidence at PCS, confidence to believe in ourselves and say, we do have this talent and we can do it ourselves. We don't need somebody from the outside to come in and do it for us. (Palau 045) This case opens the door for deeper discussions around foundation led capacity-building programshow they were developed and for whom. Packard's approach to capacity-building was one of the most well received aspects of their funding program: "the capacity building was probably [some] of the most critical work that they've done, you know, just upgrading and upscaling people of all organizations" (Fiji 067). The role of Western institutions in development, empowerment, and capacity-building, is contested in the literature, with critical and Indigenous scholars rightly lamenting Western-led capacity-building efforts that champion "modern" conservation practices over customary institutions (Dawson et al., 2021;Paredes et al., 2019). Avelino (2021 p. 13) argues that such efforts inevitably result in poor conservation and social outcomes while also "creat [ing] a dependence relationship which, by definition is disempowering" and "only reinforces the dualism of powerful-powerless, thus ultimately maintaining the superior position of the powerful." The positive reception of Packard's capacity-building programs may be explained by Dawson et al.'s (2021, p. 1) work on justice in conservation: "equitable conservation, which empowers and supports the environmental stewardship of Indigenous peoples and local communities represents the primary pathway to effective long-term conservation." While Packard's capacity-building efforts did not dismantle asymmetrical power dynamics, we believe, along with many of the grantees, that the donors' self-reflexivity and willingness to listen helped program officers more effectively address the concerns of scholars like Paredes et al. (2019) and Avelino (2021). For example, one Packard donor said:

It's a totally different situation if [a donor]
shows up and acknowledges the power dynamics and acknowledges the oppressive origins of philanthropy, and says I want to do whatever I can to help in this moment with the power I have as an individual that's sitting in a place with the means to support movement on this issue, right? (Donor 085)

| Convening and coordination
Environmental governance scholars acknowledge the specific contribution of agents who convene (provide a platform for actors to interact) and coordinate (engage actors to align activities toward shared goals and create synergies) (Betsill & Milkoreit, 2020;Fujisaki et al., 2016;Schroeder & Lovell, 2012). We describe how, through the structure of grants and promotion of inter-actor platforms, Packard performed a coordination and convening role that linked objectives and work across scales, looked to understand and fill gaps in existing conservation work, and encouraged new synergies and relationships between practitioners to accomplish initiatives together.
Packard coordinated work with other donors and across grantees in a deliberate effort to ensure that grantfunded initiatives were complementary and addressed existing gaps in conservation work: We have so little money-to imagine that $5 million is going to have any impact across six countries alone is ridiculous … If we aren't doing this in concert with other donors, what are we trying to achieve? I just thought it was important that we collaborate … (Donor 086) Packard staff members also facilitated the coordination of conservation activities among diverse grantee organizations, encouraging partnerships and networks to collectively pursue broad objectives, such as coastal fisheries management: So it's not a case of just we'll give a little bit of funding, and then hope that it all kind of comes together. They've been really sort of trying to push, and I think a lot of that came through the role [Packard Staff Member] played … looking to provide that support and encouragement to ensure that [grantees] are talking and communicating and coordinating and collaborating. (Fiji 028) A prominent example of Packard's convening and coordination role was their support for Fiji's Locally Managed Marine Protected Area (FLMMA), a network that encourages "collaboration among government ministries and departments, NGOs, private or business sector, communities and individuals to better manage the 'I'qoliqolis' of Fiji" (Govan & Meo, 2011, p. 6). Packard provided funding for the initial establishment of FLMMA as well as longer term support for the network and ongoing efforts to maintain collaboration among members: "[Packard] fund[ed] people's time to get together and talk about this and develop a constitution of how we work together, our promise to each other" (Fiji 030). FLMMA filled a gap in the existing structure of conservation work by facilitating a community of conservation practitioners and creating a platform that "organized the conservation movement as a whole in Fiji toward similar objectives and goals through the FLMMA network" (Fiji 024). Packard worked with grantees to promote FLMMA as a network based on trusting relationships, communication, community-based and traditional management practices, and coordinated conservation efforts: The other thing was that [FLMMA is] holding everyone together so that they work better together or they're able to start communicating with each other. I think at some point it was mostly organizations just going off and doing their own thing. And I think Packard's enabled or emphasized the need for their grantees to communicate with each other. (Fiji 063) This case provides additional evidence for the role foundations can play in facilitating collective action and highlights the need for more research on the opportunities and risks of philanthropy in collaborative governance (Gilson & Garrick, 2021). Some practitioners in Palau, for example, observed that coordination may have unintended consequences: [Coordination] creates a pool of people that we can tap in as resource persons, but it can also create competition between organizations. And then it makes it kind of, people are applying, and they don't want to share information. So, it's kind of like you have to balance that. (Palau 062) Gilson & Garrick (2021, p. 190) argue that, while funding can enable collective action, "foundations should explicitly consider and address legacies of exclusion for marginalized actors and groups." While recognizing the crucial role of Packard in bringing conservation actors together, the recent closure of the Western Pacific Program raises questions about the continued support and efficacy of coordination efforts, like FLMMA, as well as the impact of exits on established grantee networks (Le Cornu et al., in review).

| Facilitating knowledge
Packard performed a "facilitating knowledge" role through grants that contributed to the development and use of knowledge concerning marine conservation in Fiji and Palau. That is, "facilitating knowledge" includes both the role of building a knowledge base and the role of turning built knowledge into action-often described as evidence-based decision-making. These efforts were often directed at informing practitioners and government decision-makers to influence policy, as well as broadly contributing to marine conservation knowledge and bestpractices.
While Packard grants supported conservation science, such as socio-economic valuations of marine "resources," they also helped elevate traditional knowledge and governance by supporting "bottom up activities from the community side focusing on iTaukei communities" (Fiji 031): They didn't say, hey, maybe you should use the Great Barrier Reef approach of spatial planning. That is what we're doing is, we're not picking a more hard science-based approach to marine conservation. But the foundation has been built by Packard on those traditional approaches … Packard was quite open to traditional approaches. (Fiji 027) In Palau, foundation funding also enabled agencies to invest in expensive tools to collect scientific data on the health of coral reefs, mangroves, and commercial fisheries that ultimately influenced government decision-making: "So I think a lot of their work has done good in providing the information that [government officials] use for decision-making (Palau 052)." In addition, grantees drew on Packard-funded baseline research to push for conservation action from government, as in the case of inshore fish populations in Fiji: The set-size is probably a really good example that's come through supported by Packard funding … They've been able to support gathering the information needed in order to drive that as a priority … it was always an understanding that size limits were inadequate. But there was never that understanding about what next? And if these aren't adequate, what structures can we look at? How can we monitor it? So Packard funding has been able to support reaching that level of understanding, and then also to say this is a priority, this needs to happen. (Fiji 028) This example speaks to Packard's second mode of knowledge facilitation: animating knowledge through translation processes. One Packard staff member described their theory of change as focusing first "on the science and data and spatial and fisheries management work," and then on "look[ing] at capacities across sectors, including government" to ensure "awareness turned into policy" (Donor 086). This demonstrates Packard's commitment to promoting evidence-based decisionmaking, described by some practitioners as "science to action" (Fiji 024). Importantly, Packard's knowledge facilitation role established baselines for future knowledge creation activities. These ways of knowing and creating knowledge influence how grantees design and conceptualize their work long after funding ends: "I think [ecosystem-based management] has had a legacy and continues to be utilized, employed in how we think about project development, how we think about the different partners that need to be involved" (Palau 044).
This case demonstrates the need for continued discussions about how foundations contribute to the legitimization of knowledge, specifically how foundations perform a crucial role in defining whose knowledge counts for environmental decision-making. Packard's role of knowledge facilitation demonstrates how philanthropy can both reinscribe Western-based management strategies and promote the recognition and legitimacy of traditional knowledge in decision-making. This point is important because funder engagement with diverse ways of knowing can have serious implications for conservation agendas and who is empowered to shape them (Avelino, 2021;Baker & Constant, 2020;Dawson et al., 2021;Hickey, 2020;Martin et al., 2019;Temper & Del Bene, 2016). The types of knowledge bases that are legitimized and valued-for example, within Western ideas of managing nature versus I 0 qoliqolis-define what ideas act as truth and allow some forms of governance to make sense, while others do not (Liboiron, 2021;Van der Molen, 2018).

| Rule-making and regulation
Rule-making and regulation is most often associated with state agents and intergovernmental organizations (Betsill & Milkoreit, 2020) with the authority to enact "hard" control measures. While Packard did not have the authority to directly set the governing rules and regulations in Fiji and Palau, we found that they contributed to this governance function through indirect means of steering that included mobilizing political support and galvanizing other governance agents and networks through their grantees. For example, several early grants advanced the implementation of marine protected areas, outlining best practices for conservation and laying the groundwork for developing protected area systems. This catalyzed the development of national legislation to establish a protected areas network (PAN) that, in turn, shaped normative conditions and institutional processes across Palauan states regarding how to manage protected areas (Gruby & Basurto, 2013 Through its capacity-building role, Packard also influenced the pool of decision-makers that ultimately shaped state regulations. In Palau, for example, PCS, a grantee highlighted as an example in the "capacity building" section above was described thus: [PCS] is a very influential organization here in Palau … they have a tremendous amount of influence in how political decisions are made, or even what direction states take in terms of conservation initiatives. They have a lot of clout. And I don't know if it's primarily because of Packard, but I certainly would say that the funding enabled that strength and that influence to prosper further. (Palau 044) In Fiji, the Foundation dedicated resources to support the creation of a new regulatory body tasked with establishing rules and regulations to fill gaps in coastal fisheries management. This funding supported the creation of a standalone division within the Fijian national government to improve inshore fisheries management: "Packard investment to multiple grantees in Fiji was to create some [interest], without lobbying, to get the Ministry of Fisheries to create a coastal fisheries division" (Fiji 031).
The Packard Foundation did not enforce rules or regulations directly. Rather, the work of its granteessupported by Foundation grants for capacity-building, knowledge generation, policy analysis and reform, etc.influenced marine conservation rules, regulations, and regulatory bodies in Fiji and Palau.

| CONCLUSION
Environmental governance provides a novel lens to examine the complex and dynamic roles of philanthropic foundations in conservation. Our research documents six governance roles the Packard Foundation contributed to in Fiji and Palau through the Western Pacific Program. These roles evolved over time as Foundation staff interacted with grantees and other stakeholders and as Packard built on its diverse portfolio of funding activities and relationships. Foundations are a significant blind spot in environmental governance scholarship, and likewise, philanthropic practitioners don't typically see themselves as governance agents. Our aim is to change that by firmly situating philanthropic foundations as governance agents that do more than fund. This paper is thus a starting point for leveraging the full potential of environmental governance scholarship to better understand and inform philanthropic practice and develop a more comprehensive theory of how agents govern the environment, as well as who has a right to govern the environment. For example, Packard does not have the power to directly set governing rules and regulations of any sovereign state or government. Instead, the Foundation's power to influence governing norms and institutions is an indirect means of steering that involved filling monetary gaps that mobilized and galvanized other local governance actors and networks, a process sometimes referred to as "orchestration" (Abbott et al., 2016). Other environmental governance literatures on field-building (Bartley, 2007;Phillips, 2018), agency (Betsill & Milkoreit, 2020), meta-governance (Hooge et al., 2021), polycentricity (Carlisle & Gruby, 2019a) and environmental justice (Bennett, 2018;Bennett et al., 2021;Parsons et al., 2021) are ripe for engaging important questions of rights, sovereignty, and power in conservation philanthropy.
Our study also has implications for practice. By conceptualizing foundations as governance agents, we offer a language that practitioners and donors can use to reflect on who foundations are, how they fit into complex conservation networks, and the tools at their disposal to achieve their conservation goals. For example, our research identified governance roles that go well beyond funding and the categories Packard used to organize its grantmaking: capacity building, policy reform and analysis and applied science, public education and media, and site-based conservation and fisheries management. Literature on environmental governance roles and functions can help foundations reflect on what it is they are really doing in conservation, and what it might mean for their strategy development, monitoring, evaluation, and learning to conceptualize grantmaking and outcomes in different terms. Finally seeing foundations as governance actors may unlock new ways of thinking about their potential to address complex environmental problems, as well as their responsibilities as governance agents. This research helps move the conversation around conservation philanthropy beyond binary conceptions of "good" versus "bad," and, instead, toward deeper considerations about what foundations do within governance systems, how they engage with diverse practitioners, as well as what they can and should do (see also Breeze, 2021;Villanueva, 2021). For example, our results highlight the contradictions of the philanthropic relationship: grant relations can encourage collective action, legitimize local conservation knowledge, and empower communities, while still reflecting and reinforcing the same power and wealth asymmetries that contribute to many environmental problems in the first place (Beer, 2022;INCITE!, 2017). We argue that the way forward is to explicitly acknowledge the paradoxes of giving and collectively reflect on how to manage them. Environmental governance scholarship offers a conceptual roadmap to guide this effort. With a shared understanding of what roles foundations perform, donors, grantees, scholars, and other stakeholders will be better equipped to engage in deliberative dialog around pressing normative questions: what is the right role of foundation money in marine conservation governance? We see this as promoting the necessary conditions for holding foundations accountable in ways that have thus far been elusive (Reich, 2016).

CONFLICT OF INTEREST
Consistent with a knowledge co-production approach, we have embraced a multi-faceted relationship with our funders who also serve as research participants and research subjects at various stages of the project. We have convened an external research advisory committee to help manage any conflicts of interest that arise from this arrangement.

ETHICS STATEMENT
The purpose of the interviews and the use of data were explained to all interview participants. All interview participants agreed that the content of the interviews could be used for scientific analysis and published in an anonymous form. Interviewees listed in Appendix S1 gave explicit permission to be acknowledged for their time.
ORCID Jeffrey E. Blackwatters https://orcid.org/0000-0001-6135-5882 ENDNOTE 1 We recognize that the "post" in "post-colonial" is problematic, because colonization is not a concrete historical moment, but rather an ongoing set of land and sea relations that maintain imperial state-formations, settlement, and capitalist development. We used the term "post-colonial" to center how Palau and Fiji have and continue to disrupt colonial relations and maintain selfdetermination (see Liboiron, 2021).