Document(s) 23 of 24

Introduction

The research projects underlying this series of cases proceeded from the supposition that the meaningful participation of poor farmers or fishers might produce more effective and sustainable innovations than technical interventions driven by outside experts. The common framework underlying this action research approach is outlined in Chapter 2. In light of this starting point and the common framework, what is surprising about the cases is how much the authors have come to emphasize engagement with policy. Their starting point was the understanding of local resource degradation trends and management problems, with the emphasis on improving productivity and sustainability through the meaningful participation of local actors. Yet this local engagement in understanding and testing collective action for resource management has generated a much greater sensitivity to and engagement with NRM policies. In some cases, research effort in the field became specifically oriented to informing policy.¹

In this chapter we explore how that happened, and assess successful strategies for policy influence. We adopt a broad view of policy, which recognizes not only its important content character, but also its procedural and contested character. This perspective includes both the discursive nature of policy (underlying concepts, principles and values) as well as its practical implementation. The process offers considerable opportunity for influence and interpretation.² This view of policy allows us to engage a perspective which is missing from most policy studies: what does policy look like through the lens of local experience? The collection of cases in this volume illustrates in a number of important ways how field-level views of policy can be influential in reforming policy substance, discourse and implementation.

Policy-making everywhere is somewhat of a black box: very few people really understand how it happens.³ Most researchers, even those directly engaged in policy-relevant work, have rather naive expectations about policy influence and how research results connect to policy systems (Lindquist, 2001). Yet we conclude here that these participatory research projects have had a substantial degree of policy impact.

There is a broad literature on participatory approaches to the study and design of rural development interventions. The work of Robert Chambers has figured prominently in it during the past 20 years (Chambers, 1983, 1997; Chambers, Pacey and Thrupp, 1989; Pound et al., 2003). Efforts to link these participatory methods to policy have demonstrated the value of the tools in generating policy-relevant information, but have also concluded that this information is seldom applied successfully to policy change (Holland, Blackburn and Chambers, 1998). The rhetoric of participatory development can also be appropriated, more perversely, by the state itself as a means to implement policies which actually may undermine the empowerment and learning intentions of participatory methods (Sarin et al., 2003). For example, Vandergeest (2003) describes how a national programme of land reform in Laos used participatory local planning and formalized collective forms of resource tenure, yet arguably resulted in greater centralization of control and dispossession of local communities.

In most of the cases in this volume, technological innovations to improve resource productivity are introduced and then implemented by households for resources they control. As the cases demonstrate, the more challenging innovations were those directed at the development of functional local institutions for collective action and management of common-pool resources (or CPR). This is not only a difficult research problem, which requires attention to ecological, indigenous and scientific knowledge, power relations and social differentiation in the community. It is also an area which lacks good policy guidelines.

There has been widespread experience with the devolution of forest management throughout Asia. However, reviews of this experience reveal little evidence that such policies have led to improved equity or local economic benefit (Edmunds and Wollenberg, 2003). As a result, there is little consensus on what constitutes an effective policy regime for CBNRM. As these cases continue to demonstrate, meaningful devolution or co-management of CPR remains elusive despite the rhetoric of many governments in Asia.

By starting, not with arguments for policy reform, but with the situation of local resource users, these cases shed light on to what is working and what can work at the field level. They provide insights into the failings of good policy in implementation, as well as point to workable mechanisms in the field. These can serve both as models to challenge assumptions and transform policy discourse, and as examples from which to design policy reforms. By looking at policy issues from the field and from the perspective of the resource users most directly affected, these participatory research cases demonstrate ways in which the strategic interaction of resource users, local governments and researchers can influence policies of decentralization and NRM. This chapter explores the factors from the case experiences which enabled such a policy impact. We conclude that both context and research approaches have an important bearing on the policy effects of participatory local NRM research.

The chapter proceeds by exploring what policy is about, and proceeds to discuss how policies are typically shaped by a variety of influences both in their design and implementation at different levels of government. We present the experience of the cases in two ways: first, we link the opportunities for CBNRM to the national decentralization policy framework of several of the countries represented here. Next, we demonstrate the types of policy influence demonstrated by the cases in this volume, and illustrate the strategies the participants adopted to exploit opportunities for policy influence. We then show how these strategies and influences arose from the context and methods applied. These comparative assessments demonstrate the importance of social learning, and the transformative effects of local initiative and innovation on policy change. Finally, we examine the new roles which researchers have adopted through these experiences, and the ways in which these roles have been instrumental in the policy change outcomes.

What is policy?

It is common to conceive of policy as being the explicit statement of government priorities as interpreted in action and reflected in laws, operational directives and regulations. The local impact of policy, however, is as much related to the way it is put into practice (or perhaps filters down) as to the letter of the central government documents (Shankland, 2000). Policy also often changes during the process of its implementation. Therefore, we will be interested in the formulation and formal content of policy as well as its implementation and local administration. This perspective brings out the process nature of policy: it is formulated and implemented, but also modified, deflected, interpreted, contested and resisted. This also draws attention towards the various actors involved in policy at different levels, the roles they play, the ways they relate to each other, and their networks of information exchange and learning.

In this chapter, we focus on the role in the policy process of researchers who are engaged in community-level PAR. In this role of considering researchers as actors, sometimes unexpectedly the researchers became part of the policy community (Lindquist, 2001; Lindayati, 2002). This type of community consists not only of one or more government departments, legislators and/or state leaders. Depending on the national context, such a community also includes journalists and commentators, political parties and donors, as well as other international organizations, consultants and lobbyists, and various interest groups. A policy community can be small, closed and opaque, as in China. There, major recent NRM initiatives such as the logging ban and the upland conversion policy were initiated by key central and provincial leaders. They were subsequently worked out in central government agencies and implemented top-down through the associated line departments. In contrast, Cambodia is characterized by a much more open and diverse policy community, even if key decision-making power is concentrated in few hands. Here we find major multilateral and bilateral donors, international environmental (NGOs, domestic NGOs, foreign and domestic corporations, and researchers who are all playing key roles alongside government agencies.⁴

Policy communities are not static. Lindayati (2002) shows how in Indonesia, during the period of Suharto's New Order, the forestry policy community was small and dominated by the bureaucracy. The political upheaval of the late 1990s brought a major change, particularly with respect to parliament, local governments and NGO-based societal interest groups, which all began to play a larger role in the policy process. However, the extent and type of linkages researchers can make with policy processes depends on the kind of policy community in the country and its dynamics.

The ways in which the various actors interact in the policy process has been extensively analysed in terms of policy networks, advocacy coalitions and other kinds of groupings.⁵ In the wider policy community, different actors can rally around certain issues or promote certain directions of change. They do not necessarily share the same interests or values, however, and may come from different backgrounds and play different roles in the network. Often more than one policy network is active in a policy community, working to maintain or change certain policy directions. As we will see below, many of the research teams became part of or helped initiate specific policy networks. With the authority that came from international support and the credibility provided by field-based research evidence, the teams were able to enrol senior policy-makers and others into their networks through a variety of mechanisms, which we discuss below. Straddling local community and decision-making levels, research projects uniquely brought the voices of local men and women into the policy networks in diverse ways.

Focusing on policy as a process and on the role of different actors within it also illustrates how policy-making is multidirectional. In other words, policy does not simply come down. It is constructed in practice and the bureaucracy and other organized stakeholders as well as local people all play a role in its construction. In this chapter, therefore, we are not only concerned with the ways in which researchers and local people managed (or failed) to influence policy (in the sense of its formulation and implementation), but also with how they used policy in the local context for a variety of purposes. Seen in this way, the policy process is also intimately intertwined with politics and power.

Finally, policy can be regarded as a discourse or narrative: a story about issues, their causes and solutions, as well as basic assumptions underlying them.⁶ Such policy discourses can be quite powerful and pervasive, with people being hardly aware that their actions are guided by them. In the context of the case studies in this volume, most research teams faced entrenched policy narratives about environmental degradation (caused by local and indigenous people's NRM practices), about the need for state control over natural resources and about the inability of communities to conserve these resources.

In practice, the basic assumptions of mainstream policy discourse often blind actors from different groups to realities that do not conform to the model. In the process, certain groups become marginalized. For example, in many settings, NRM policies are based on the experiences of lowland, irrigated, smallholder agriculture. However, policies become extended without further consideration or adjustment, and applied to upland areas which have radically differing systems of land use and access rights (see the Ratanakiri case study, Chapter 3).

Different discourses also mean that certain kinds of knowledge become privileged. For example, narratives associated with Western modern and scientific forestry are still dominant throughout most of the world and explicitly legitimize the knowledge of technically trained foresters over that of the forest dwellers. We will see this type of legitimacy repeated time and time again in the scientific forest inventories and management plans that are required before communities can be granted management rights. One key role that the CBNRM research teams in these cases have played is to provide alternative policy ideas and narratives. Like the mainstream discourse, these consisted of a story about issues, causes and solutions, but instead, these policies emerged from and were substantiated with contextual field evidence.

Policy positions and narratives often show a remarkable tenacity, even after changed circumstances or an evident lack of effectiveness. This is primarily because policy discourses become embedded in the routines and practices of organizations and are internalized by their staff through training (Shankland, 2000).⁷ Therefore, promoting policy change in practice often means a struggle with the bureaucracy. However, despite its stickiness, policy does change. Such change can be brought about by officials in government agencies who after learning more become champions for change. Learning can also be more broadly shared social learning within the institution in which they work (Korten and Sy, 1988). In some countries like the Philippines, changes in departmental leadership as a result of elections can be an important factor in uprooting entrenched perspectives. Sometimes, shock events trigger policy change, with environmental crises like major floods being a common and prominent factor in NRM policy changes. Whatever the specific drivers of change, it is clear that policy is more responsive to external influences in certain situations and at certain times than others. This phenomenon of important coincident factors, sometimes referred to as windows of opportunity (Lindquist, 2001), leads us first to examine the substantial policy context in which the case studies took place. We discovered all of these cases are characterized by windows of opportunity that were vital to enabling the eventual policy impacts.

Governance reform and decentralization

In this section, we will examine the policy context in which the case studies are situated. We will show how broad processes of national policy reform around decentralization provided opportunities to which the researchers responded adroitly. The types of impact and strategies they deployed in these windows of opportunity are analysed in the next section. Most of the research teams soon realized that certain policies were obstacles to CBNRM, or that certain policy processes could help push along local change. This change towards policy orientation was most pronounced and explicit in the case of the IIRR. Chapter 13 describes how their learning process quickly took the research team away from a focus on community capacity-building towards policy issues. They realized that the main bottleneck to community forest management was not the community's capacity, but the disincentives inherent in the wider political economy.

Decentralization of NRM is occurring in many countries around the world (Larson, 2003), including most of the countries covered in the case studies. This trend has been under way for some years in Asia, with mixed results (Tyler, 1995; Dupar and Badenoch, 2002; Edmunds and Wollenberg, 2003). Most experiences have been sectoral. In particular, there have been extensive government programmes in the forestry sector to engage communities in forest management, from the very widespread adoption of forest user groups in Nepal (Kanel, 2003) to the much less comprehensive results in Southeast Asia (Fisher, 2003). In the Philippines and, more recently, in Vietnam there has been some devolution of coastal fisheries management to local government authorities (Ferrer, Polotan de la Cruz and Domingo, 1996; Tuyen et al., Chapter 4). To the extent that these experiences have been reviewed critically, the main conclusions have been that despite many years (sometimes decades) of formal policy implementation, they remain contested because of the difficulty of wresting significant management authority from the state to better serve the interests of the local poor. It should also, however, be noted that most of these policy reforms were introduced centrally as a response to internal or external political pressure, but not in response to specific local NRM initiatives or examples.

Such programmes of governance reform sometimes involve the decentralization of functions and responsibilities to local governments in general, or reforms specifically related to natural resources, or both. Although the situation of Cambodia in many ways is unique and anomalous, it well illustrates these trends and the responses of researchers.

In the Cambodian context, researchers responded to take advantage of this fluid context and shifted the focus of their work in response to policy opportunities. Detailed illustrations are provided in the section below. However, the community forestry research project (Chapter 11) also provided inputs to the formulation of the community forestry sub-decree and organized consultations with communities around constructing a draft language for this document. The CFRP team has presented experiences with field processes such as participatory inventories and management plans to the FA to assist it in formulating guidelines for such processes.

Although the specific historical context in Cambodia is unique, many elements of this picture of decentralization as well as the responses of researchers in Cambodia are common to the other case studies as well. First and most important, the countries covered in the case studies are all undertaking decentralization reforms. In the Philippines, this also involves general decentralization to local government units and special provisions related to indigenous peoples and community forestry. In Mongolia, Laos and Vietnam (and in China, considerably earlier), user rights over natural resources were also devolved away from collectives or state entities, to households and local communities.

Second, policy-making is often in a state of considerable flux throughout the period covered by the case study. In Cambodia, both policies and institutions were being created from scratch or were being recreated. While the extent of such policy fluidity may differ, it is also found in most of the other countries. Often, promulgation of legislation is not the end of policy-making but the beginning of complex processes of policy exploration and negotiation. We see this, for example, with the new land law in Mongolia, with legislation on indigenous peoples' land rights in the Philippines and with the formalization of a water policy in Bhutan. Policies must find the right fit with reality on the ground, and CBNRM research projects have been instrumental in facilitating exactly such processes.

A third commonality suggested by the Cambodian cases is that the CBNRM teams pursued local research objectives that were broadly consistent with the government's policy goals under the decentralization programme. Rather than taking an oppositional approach to government policies, research teams contributed to their improvement and implementation. However, while broad government policy initiatives pointed in certain directions, sectoral agencies and local governments usually had little guidance and few ideas on how to implement them. But with their CBNRM outlook and extensive field experience, the researchers had concrete ideas on how to translate broad policies into practice on the ground. They were able to frame interventions to benefit local communities in the context of these policy opportunities.

In one sense, this is not surprising. Decentralization transfers authority and responsibilities to the local level, which is exactly where the CBNRM projects were active. But the connection between the field research and decentralization in practice was conditioned by the nature of the resources which are at the core of CBNRM work (see Chapter 2). Water, forests, grassland and swidden land are all CPRs. They have multiple users and some degree of collective action is required to prevent overuse and degradation. For this reason, as well as their economic and strategic value (such as national security or biodiversity conservation), governments usually assume responsibility for their management. This combination of local use and state authority directs CBNRM projects to activities that combine rearrangement of local tenure and roles with policy advocacy. As such, this is in contrast with agriculture, for example, where land rights are already at least quasi-private, and where management is universally acknowledged to be at the household level. For example, the Hong Ha commune project in central Vietnam includes work related to agriculture and to forestry (Chapter 5). In contrast to the forestry work, which almost immediately drew the project into questions of national policy in relation to forest access rights, the agricultural component has been quite successful without needing to concern itself with policy at all.

Where these research cases merged with decentralization policies, they worked to establish or strengthen community claims to resources. This usually involves the delineation of boundaries and the definition or change of access and management rights by different user groups. For rights and claims to be secure, they need to be officially recognized and supported by the state. Therefore, a considerable amount of researcher advocacy relates to such state recognition (see the discussion of legitimacy below).

However, state legislation of resource rights often ignores the complexity of local property relations and access arrangements. Perhaps the clearest example of this is the Ratanakiri case, where the indigenous people practise swidden agriculture. Here, land use patterns cannot be easily classified in terms of agriculture and forestry. Customary land use rights consist of a complex mix of private and collective arrangements. When the national government initiated a land titling drive based on PLUP, the Ratanakiri experience influenced it to develop special procedures for indigenous areas. In other cases, state policy on resource rights may lead to deliberate or inadvertent enclosure of CPR, resulting in the selective exclusions documented in the Laos and Tam Giang lagoon cases (Chapters 7 and 4 respectively). In the Philippines, the ancestral domain case (Chapter 12) illustrates how the hasty conferment of indigenous land rights on local administrative units led to a mismatch between the authority responsible and the customary organizational forms for managing these rights. The matching of local institutions and rights to resource use and management are at the core of what the research cases do, but such reforms are also central to the effective expression of decentralization policies.

What does policy look like from the field?

As demonstrated in these cases, participatory research provides insights which allow the characterization of policy alternatives or policy implementation more starkly in the lived experience and quality of life of local men and women. This provides an immediacy that is more arresting than quantitative analysis, and is more likely to reveal contradictions in underlying assumptions than detailed analyses that take such assumptions as a starting point. Indeed, conventional policy research often fails to take account of local realities. Priorities and measures of success for policy-makers (as well as standards of evidence, assumptions and systems of belief) are not necessarily the same as those of community members.

In these cases we see that local resource users, community leaders and researchers, once engaged in the action research enterprise, display a broad range of explicit or implicit strategies by which they engage, manipulate and make use of policy opportunities to strengthen the legitimacy, power and effectiveness of new CBNRM schemes. The actions of the various local actors are motivated by a keen desire to improve conditions for local men and women.

It is the local perception of policy which is most important in guiding and motivating interventions for change. What does policy look like from the field? In many of the projects, it was the contradictions between the stated intent of the policy and its actual implementation (or unintended side-effects), which were key to shaping the project innovations. So, in the IIRR case (Chapter 13), an experimental approach to shared learning was developed to address the conflict between the stated intent of the community forestry policy and the systemic barriers to implementation which were revealed in preliminary field investigations. In the Tam Giang lagoon in Vietnam (Chapter 4), national policies to foster aquaculture development led to unexpected transfers of resource rights and to the exclusion of poor resource users, which the project tried to address. And in Mongolia, the dismantling of traditional forms of community pasture management under state socialism, followed by the introduction of economic reforms which encouraged the growth of private animal herds in the 1990s, led to serious degradation of grazing land (Chapter 6). In all these cases, the local view of policy change was decidedly different from what had been intended by the drafters of the original policies. In recognizing this contradiction, the research teams began to act strategically to influence policy.

Recent studies of how research influences policy suggest that while there are no universal guidelines, some generalizations can be drawn about factors which strengthen or weaken the potential for such influence (IDRC, 2004). One of the most important elements in policy influence from research was the relationship between governmental needs for knowledge and researcher interests and capability in providing it. This relationship can be characterized along a spectrum where, at one end, policy-makers and analysts in government know that they need knowledge about a particular policy issue and are faced with a need for urgent action. In this situation they are actively seeking knowledge and insight. At the opposite end of this contextual spectrum, governments may be hostile to addressing an issue which is in the public domain, leading them to not only ignore, but also to bury research on that issue. Most of the research cases in this volume can be situated in the middle of this spectrum, where the government is at least nominally committed to addressing the issues (e.g. decentralization, poverty, environment and resource degradation) and has only a limited sense of how policies should be framed, or what specifically could be done to implement them. In this context, researchers must act strategically to assure policy attention to and impact from their work.

The kind of knowledge which is provided through PAR, such as the cases presented in this volume is rich, but often perplexing to policy analysts. Participatory research generates messy evidence: local interactions between resource users and the environment are shown to be complex and shaded by culture, history, social relations, the ways power is used, by ecological context and by external forces. This picture contradicts the simplifying assumptions about behaviour, incentives and causation made by policy-makers who necessarily operate at a high degree of generalization. Such studies also reveal that local interests in resource use overlap and conflict. New, hitherto unrecognized interests are frequently exposed, to further complicate situations in which power and political trade-offs are often central. None of this makes policy decision-making easier. It also tends to make policy-makers respond defensively to the conclusions of such exercises, or to dismiss them entirely (Freudenberger, 1998). Despite the apparent mismatch between the methodology and policy impact potential, these cases demonstrate how researchers can highlight key elements of this complex field reality and successfully link field learning to focused policy change.

The success of delivering policy impacts from participatory research depends crucially on the skills and initiative of the researcher. We have seen how those cases which engage most directly with policy reforms took place in a context of policy fluidity and under national policy trends of decentralization coupled with the devolution of NRM. However, this context merely structured the opportunity: it was up to the researchers and their colleagues in government and other agencies to identify and seize such opportunities. They proceeded from local research insights to policy influence in a variety of ways.

Direct research input to policy formulation

The most obvious way in which the participatory research affected policy in these cases was when the research results provided direct input into the formulation of new policies on resource tenure or resource management. In a number of cases, the research team has been in a position to draft or influence the framing of policy reforms associated with the decentralization of natural resource governance. In such instances, important provisions of the new policies are based on the results of the participatory local research work. These are cases in which the research is led by staff from a key national government agency. In the Mongolian case, the research project leader is a senior policy adviser in the Ministry of Nature and the Environment. But this is an unusual situation: research project management and policy advisory units are normally separate, and in the cases presented, the research may even be undertaken in a completely different agency from that charged with direct policy responsibility for management of the natural resources in question.

Two of the Cambodian cases in this volume demonstrate a direct influence on national policy formulation. The community forestry case (Chapter 11) demonstrates how participatory research in multiple field sites was specifically structured to provide evidence to support policy reform. In this case, the evidence was intended to show the feasibility of participatory local demarcation and control of forest utilization to a sceptical forestry agency. This was effected by developing practical tools for local forest planning and management, by demonstrating congruence with other government policies (particularly local government reforms), and by demonstrating the potential for improving local livelihoods. The research was led by the Ministry of Environment, not the Ministry of Agriculture, Forestry and Fisheries, which controls policy jurisdiction. This situation made the policy linkage particularly delicate, in order to avoid jurisdictional rivalries. We discuss relevant strategies below.

In the Ratanakiri case (Chapter 3), the early research work of the project influenced the process of framing new national land tenure legislation. The legislation was guided under the influence of various donors and with multiple objectives in mind. However, the research project provided the critical evidence needed to justify the creation of a new category of collective land and resource tenure which was appropriate to the customary practices of upland ethnic communities in Ratanakiri and other provinces. The significant aspect of this example was that the project was based in a provincial not national government office. The office had no policy responsibilities and was only responsible for implementing a large donor-funded programme throughout the province. With careful strategic effort, the research work done by the provincial office was used to leverage national policy change (Muny, 2001).

In the case of Bhutan, the team worked in a national policy context favouring decentralization. However, during the course of their fieldwork the team uncovered evidence proving that customary water allocation principles were inequitable. The research helped explain and overcome these problems, and helped form the structure and content of a new national water policy. This occurred even though the research group was not part of the line department responsible for water resources management.

These examples suggest that one way for the participatory research to provide direct input to policy formulation is for government agencies that have access and policy credibility to be involved in the research itself. Local NRM issues create conflicts, and government is not a neutral player (Tyler,1999). For these reasons, it may be difficult if not ill advised to engage the line agencies with direct administrative jurisdiction over the resource base in participatory learning at the community level. Nonetheless, these examples also show that the researchers do not have to come from the responsible resource management agencies in order to be effective in influencing the policies adopted by those agencies. This is a delicate situation because jurisdictional issues and inter-agency rivalries frequently hinder relations between the different groups.

To overcome frictions, it is important that government researchers do not attempt to claim jurisdiction or to provide specific policy recommendations, because such strategies would inevitably be interpreted as interference. Researchers' influence stems from being able to provide unique knowledge and insight from within the government system. We can see in the inter-agency dynamics of these cases (particularly in Cambodia) that research partners who are in the government system are better able to attract the attention of senior political figures, convene formal or informal inter-agency working groups and assign staff across departmental boundaries for special projects.

It is also significant to note that policy insights which arose from research were not due solely to the technical expertise of the researchers. Technical expertise and authority reside in the line agencies whose staff are trained in forestry, fisheries or water resources. While the natural science background of most of the researchers was an important element in their work, policy insights came from their use of innovative participatory methodologies. It was the unique knowledge and insights generated from these participatory learning processes, rather than special technical qualifications, which gave legitimacy to the researchers, enabling them to provide advice on policy formulation. Such insights were not available to the technical specialists in line agencies, although once engaged they were often able to validate conclusions from their own field experience.

Framing policy implementation

An even more widespread mechanism by which these cases influenced policy was through the shaping of policy implementation at the local level. This kind of policy influence is very important to successful policy reform. The experience of failed development programmes throughout the world illustrates that while policy is crucial to effective change, it is not sufficient. Carefully designed policy and programmes aimed at reducing poverty frequently fail to meet expectations at the implementation stage (Pritchett and Woolcock, 2004). It is the intersection between policy and practice which forms the locus of the work described in these cases. By identifying specific interfaces between policy and field innovations, the research teams went beyond policy recommendations to testing policy in field experiments.

For example, the ancestral domain case in the Philippines (Chapter 12) defined one of its goals as demonstrating and persuading the government of the utility of using the indigenous ili as the unit for ancestral domain titles. This project also facilitated the development of ADMPs, which formed a focus for work on the interface of policy and practice. On the one hand, they entailed a series of concrete activities in the field, in which communities analyse, plan, devise rules and finally develop a specific management plan that also guides their future behaviour. On the other hand, such participatory management plans provide examples of how policy could be implemented. This is because they are the objects of formal recognition by state agencies charged with implementing the new ancestral domain policies.

The cases illustrate many similar kinds of examples, where the research practice in the field provided concrete illustrations of how policies could be implemented to better achieve their stated aims (typically related to poverty reduction, equity and resource conservation): PLUP protocols developed by researchers in Ratanakiri (Chapter 3); the rearrangement of waterways and net-enclosures facilitated by researchers in the Tam Giang lagoon (Chapter 4); the forest management plans developed in Cambodia for the community forestry case (Chapter 11) and the villagers' NTFP cultivation activities which required legalizing access to forest resources in Hong Ha commune (Chapter 5). The Mongolian pasture management project has been involved both in the drafting of national legislation as well as in the local details of how it might be implemented (structuring leases, developing participatory management plans, providing for oversight and approval mechanisms and so on). In the Lao case study (Chapter 7) research reveals how policies can be implemented in unexpected ways, as community and government staff framed exclusionary interventions in terms of the new policy orthodoxy.

The practice of participatory research addresses local policy conflicts and contradictions between intentions and outcomes, by building shared insight and trust between men and women resource users and local government. This is a fundamental element of improving governance and policy. It allowed the local people and researchers to move beyond typical complaints ('there needs to be more participation') to jointly developing the mechanisms that embody solutions (participatory planning, consultation processes, resource user groups). This changes the way that policy is implemented, and allows opportunities for local interests to be better served under existing policy, through more sensitive and appropriate implementation strategies.

It can also help overcome resistance to policy change which is already under way. In China's Guizhou province, researchers found that despite supportive policies on poverty reduction and local participation in extension and resource management, intermediate levels of government are locked in narrow technocratic approaches and are unable to take advantage of higher-level policy reforms. Implementing agencies are frequently resistant to policy change, even when mandated by new legislation. PAR can help to build common understandings and move away from irreconcilable conflicts.

In most of these cases, the participatory research process involved engaging local governments in experiencing learning with local men, women and researchers. Typically, rights over and use of natural resources were disputed. These represent important sources of local conflict and sensitive elements of policy. A frequent criticism of decentralization policy reforms in NRM and other sectors is that local officials are not provided with the training and resources necessary to implement the policies effectively (Cheema and Rondinelli, 1983; Tyler, 1995). Several of the cases specifically refer to the confusion of local officials who had little idea how to implement new policy guidelines for local resource management. Because their research offered new insight into problems, new ways for local people to communicate with governments and new approaches for responding to and engaging local people, these participatory action research projects were able to overcome policy implementation pitfalls.

The cases report a wide range of innovations which improved the ways that policy could be delivered. Most of these were achieved by transforming perceptions and possibilities. In the Philippines, despite the comprehensive reforms recognizing indigenous peoples' rights, government agencies had not considered ways to implement the legislative requirements through indigenous management institutions. Similarly, forestry officials had not paid attention to communities' management competence and people's frustration with official intransigence. In Cambodia, Vietnam and Mongolia, the cases show how research is providing officials with evidence of plural, rather than homogenous, local interests, pointing to the need for more inclusive consultation and decision processes. In all the cases, the research helped to transform assumptions by allowing local participants (farmers, government officials and researchers) to experiment with new roles and build common platforms of shared information. As a basis for making policy (and governance in general) more effective, these represent significant mechanisms of change.

Introducing new concepts through provisional legitimacy

Without changing policy, researchers can help foster new ways of looking at rural development by demonstrating successful local interventions and seeking formal government sanction for the innovative processes which generated them. Even if it is provisional, such formal recognition and sanction changes the nature of policy discourse by introducing new concepts and gaining commitment from policy actors to testing them. This also serves to build political interest in the outcomes of the research.

The situation is that when local resource users develop innovative responses to resource management problems which are not yet sanctioned by official policies and precedents, it is often crucial for them to seek political support and legitimacy using ad hoc mechanisms. Ratanakiri researchers introduced and tested a wide range of new local planning processes with the support of the provincial governor, before the framing of the Seila programme procedures. It was their experience with these provisional processes which led to the definition of more formal and widely applied processes in the national programme.

In some cases, because the work on the ground was ahead of formal policy formulation, it operated in a kind of policy vacuum. This happened in the PMMR project (Chapter 8), partially due to the remote location of Koh Kong province. It was only in 2003, after the local commune council elections, that the Seila programme began to be implemented in this coastal area. Formal national regulations for community fisheries have not yet been developed. In this context, the team was able to mobilize key politicians such as the provincial governor to support and authorize the CBNRM experiments. The team (Marschke and Nong, 2003) observes that the policy vacuum in which the project activities took place may have been an advantage, because the work was not constrained by particular models of resource management. Therefore a lot of experimentation could be done. They also argue, however, that over the longer term formal legal recognition of a community's right to manage resources is needed.

In conditions of policy fluidity and with appropriate facilitative support, this kind of contingent support for local cases can have a strong influence on national policies, both by allowing the development of local experience with innovative alternatives, and by providing policy actors with new concepts, tools and vocabulary for them to use in policy negotiations and debates. There is some interplay here with the politics of decentralization. CBNRM offers a practical tool for local political leaders to demonstrate their capacity for more independent action, in the context of dynamic tension between local and national authorities over the details of decentralization processes. When there is no policy in place, sometimes it is easier to lead locally by experimentation.

Learning from failures

Authors of the cases mostly highlight the success of their attempts to influence policy and government programme implementation. However, there are some examples of how researchers have also learned from their failures and the failures of others to effectively influence policy. In the case of Guizhou province in China, for example, the research team expresses their frustration with the unwillingness of county-level officials to reform programme management and field guidelines. This occurred despite supportive policies for poverty reduction and participatory rural extension, the encouragement of senior (provincial) government agencies and positive reports from local field officers. Although their vertical scaling approach had limited success, the researchers found much better results from horizontal efforts to scale up lessons and influence government programming through farmer-to-farmer and local government agents in adjoining jurisdictions.

In Vietnam, the Tam Giang lagoon research team learned from the failure of conflict management efforts in Phu Tan commune. When the government imposed a solution to contested fishing rights, violence ensued after it had failed to consult all the parties affected. Such a failure was instructive to both the research team and the government, and helped convince authorities to try participatory multistakeholder co-management processes which gave much more attention to marginalized fisher groups.

Researchers have recognized the need for persistence in these efforts. Some initiatives have been successful, others have not. In particular, researchers struggle to gain recognition and support of line agencies with formal jurisdiction over the resources. Even under favourable policy regimes, as in Bhutan and Philippines, the problems facing CBNRM continue. This is partly because of the challenges associated with interpreting and implementing a supportive policy, once a regime is approved, and partly due to the contingent nature of all policy change. Legislative reform can change the relative power of different actors, and can address certain issues. However, this action frequently just shifts the focus to other issues. Therefore, the need for research effort and the continued refinement of policy is evident. The impact of effective action learning in the field can be substantial.

Strategies for achieving policy influence

The participatory research cases in this volume influenced policy formulation, implementation and discourse. Researchers found ways to legitimize local innovations even without the existence of formal enabling policy. It is worth examining the strategies used by the researchers and their local research partners (village leaders, farmers, local government) to achieve policy influence in these ways. These strategies are rooted in the PAR framework adopted by the researchers. Therefore, they provide insights for researchers and practitioners in other contexts on how they can better shape enabling policy conditions for CBNRM.

Opportunity and initiative

Policy influence is born from both preparation and opportunity. Opportunity cannot always be planned into a long-term, site-based, complex research project. And participatory research is unpredictable because local men and women play an active role, not only in contributing information, but in setting priorities for learning and innovation as the project unfolds. Successful cases of policy impact were those in which researchers were prepared to respond to dynamic opportunities in the policy system. Good examples of this include opportunities created while new legislation is being proposed; when there is uncertainty associated with large-scale projects; or when conflicts or deadlocks are difficult to resolve under old policies. Because these opportunities often arise unexpectedly, researchers need to use accumulated experience from long-term research, together with strategic preparation, in a flexible and responsive way.⁸

For example, in the Ratanakiri case (Chapter 3), research evidence was applied to the development of new legislation (the land law), the formalization of local governance reforms (with the Seila programme) and the development of new forest management approaches. Each of these influences required project responses to emerging opportunities, the timing and structure of which could not be predicted in detail. Their success also relied on the cultivation of networks of information sharing and exchange. In the case of ancestral domain in the Cordillera region (Chapter 12), researchers were familiar with the customary social and governance structures of indigenous societies because of their earlier extensive fieldwork. They were able to respond to the changing policy context by developing a participatory local planning process. It brought together customary and statutory governance bodies to address resource management planning. These and other successful cases demonstrate two important prerequisites for policy influence. One is that the research teams, or their close collaborators, need to identify the emerging policy opportunity; and the other is that the teams need to be able to respond in a timely fashion based on accumulated evidence at hand.

In most policy opportunities, timing is crucial. In conditions of fluidity, widespread debate or contested policy implementation, early examples of successful solutions are influential. These projects demonstrate how innovations in CBNRM, such as examples of the first innovative leasehold rights or the first participatory community forestry management plans, can provide compelling models for change.

Communication of credible results

The impact of the local participatory research projects on policy design or implementation also depends on the strength of their evidence, analysis and the presentation of results. The research had to generate new ideas and concepts in action, not merely tell a story about complicated local conditions. It is not the complexity of local conditions which is helpful in policy decision-making; instead it is the practicality of precisely how innovations can respond to such complexity at various levels. Good documentation, of the process as well as the analysis and outcomes of the research, is essential for research to influence policy. But documentation alone is insufficient.

Policy influence requires timely and effective communication. It is important for research insights and lessons to be communicated using language and tools appropriate to the local culture. The research cases reported here all built on extensive international work in related fields (including such aspects as participatory research, agricultural extension, common property and tenure). However, because research teams were able to engage local participants in the local language and produce documentation that local officials could read, they were able to have a much greater impact. Many of the cases also make use of direct farmer-to-farmer and local-government exchange visits (particularly the Hong Ha and Guizhou cases, Chapters 5 and 9). In addition, the PMMR, CFRP and Ratanakiri projects went to great lengths to ensure that senior national politicians had an opportunity to visit the field site. These are particularly effective mechanisms for communicating outcomes because they put practical men and women with analogous problems together to critically assess proposed solutions. First-hand experience to verify outcomes is vital to overcoming scepticism when most policy actors are continually subject to promotional hyperbole from various advocates.

Government officials learning with researchers

In several of the cases, government staff from the responsible agencies have been strategically engaged in the project as researchers or advisers from the outset. This has been particularly significant as a tool for building trust between local people and government officials, and for transforming commonly held assumptions of government officials about causes of local problems. However, this strategy also has risks (Freudenberger, 1998). Government staff typically have very little time to devote to project work, particularly when it involves lengthy field visits, and their attention is continually drawn away by the interventions of their superiors. Reorganization or reassignment of staff is common and can result in the loss of key project supporters, so that the research team has to start all over again with new (and perhaps less sympathetic) personnel. It is also frequently difficult to overcome the deeply embedded assumptions and behavioural models of government staff (see above section on governance reform and decentralization). Initial responses are likely to be defensive, particularly by line-agency officials charged with NRM and policy implementation.

The research teams in these cases adopted a variety of strategies for enrolling government officials with direct responsibility for resource management in their projects. In the case of the community forestry project in Cambodia, for example, FA officials responsible for community forestry were invited to serve on the project advisory committee, and through this role became involved in fieldwork and became familiar with site-based research success. The PMMR project in Cambodia (Chapter 8) was led by national government staff. However, to implement the project they had to fully engage provincial level staff, not only in the Department of the Environment, but also other line agencies. These officials have become key supporters of the innovations introduced by the project. This strategy has perhaps been most clearly applied in the case of the Tam Giang lagoon project. Here, the provincial Department of Fisheries officials served as members of the research team in the project's early phases. As well, the department's director is strongly supportive of the new participatory planning approach to lagoon management (see Chapter 4).

Note that in a number of cases, the researchers themselves were government staff in the line agencies responsible for introducing innovations (Mongolia, Bhutan). However, even in these cases, the research work was clearly separate from the agency's regular policy responsibilities and was normally carried out by different staff in separate groups. The Bhutan RNRRCs, for example, did not have line responsibilities for resource management. And in the Cambodia cases, both the community forestry and PMMR projects were led by staff from the Ministry of the Environment, which has line responsibility only for protected areas, not forestry or fisheries. Similar approaches were employed in the Mongolia case. So even when the research was led by government staff, a deliberate strategy of inclusion and building line agency buy-in was needed.

In all cases, the benefits of this strategy arise from social learning: the transformation of perceptions and understanding through interaction with other people and the shared interpretation of evidence. Such interaction went beyond formal field engagement to social interaction in the field after working hours, which helped break down formal inter-agency divisions and barriers. Research team members did not merely undertake independent studies and write analytical reports. They also shared field experiences and interpreted and debated what they saw.⁹ Research planning, fieldwork, communications with villagers, strategies for engaging government, reviewing and analysing data, interpreting outcomes and reporting were all shared through interactive experience. The participation of government staff in many of these elements helped them to buy in to the conclusions of the research work and to recognize how those conclusions had been reached. The experience of the researchers engaging government staff has shown them new ways of collaboration to ensure the relevance and impact of applied research.

Networking and alliance-building

The literature on policy-making refers extensively to the importance of networks of influence and information exchange, through which policy agendas take shape (see reviews in Lindquist, 2001; Neilson, 2001). Many of the projects have been quite deliberate and strategic in both enrolling in such networks and in recruiting influential network members. This point has already been explained, for example, in relation to the credibility and communication of research results, and the engagement of government officials in the projects.

This strategy has been presented most clearly in the PMMR case study (Chapter 8). It focuses on the networking element of the project and demonstrates the attention paid by project leaders and members in engaging a variety of local, national and international networks.

Part of this strategy is also the construction of alliances with sympathetic individuals inside or outside government. Even large organizations which might be inimical to policy change are seldom monolithic, and different individuals or groups in the government or donor agencies can approach policy issues from diverse perspectives and strategic interests. This kind of alliance-building is not limited to government officials. Indeed, international NGOs and donors can form important strategic partners to introduce and support new ideas with local governments. Once convinced of the efficacy of local research results, large international donors have many channels for coordinated influence on policy which are not available even to well-placed government officials. This strategy has been employed by the Ratanakiri research team, including an early decision to locate the research project as part of a large, donor-funded governance reform programme (Muny, 2001).

Overall, networking has been adopted by almost all the projects represented in these cases, often with a high degree of sophistication. Research teams have contributed to alliances and coordinating groups, sometimes even creating these where they did not exist but where the need was evident. For example, development coordinating committees in the Philippine Cordillera region were designed to implement ancestral domain claims. Another set of examples where research has helped to build alliances and networks for change is the linking of local resource users to local government structures. This example merits discussion in its own right.

Linking resource user groups to local government

An inherent element of the PAR framework applied in these cases is the simultaneous strengthening of local resource user groups and the linking of them to local government structures. This is strategically important in policy influence, particularly in the fluid environment of policy change which characterizes many of these cases. CBNRM processes and local resource user groups rely on the sanction of senior government officials to enforce local planning and resource decisions (as well as technical support and inter-agency coordination). Therefore, they interface directly with reforms to fundamental processes of governance. One aspect of this, the learning on the side of communities, is borne out by a quote from the Mongolia chapter: 'CBNRM is a process where herders learned how to represent themselves to the local governors, and learned about democratic procedures by participating in decision-making on pasture and NRM.'

CBNRM projects have been criticized for the way in which, under the sanction of powerful external donors and NGOs, they have created parallel institutional structures which can disempower local government authorities (Ribot, 2002). However, most of the cases reported in this volume appear to have unfolded somewhat differently.

The CBNRM action research projects in these cases created platforms or mechanisms for cooperation and dialogue of communities with the institutions of local government, as for example the sum-level co-management teams in Mongolia, and the combined project-line agency activities of GAAS in Chapters 6 and 9 respectively. In other cases, non-formalized channels for functional relations between communities and local government are fostered in the resource planning and delivery of services such as extension, credit or infrastructure. The Cordillera, Ratanakiri and Tam Giang lagoon cases are all good examples of this.

Operating in legal and policy vacuums, in many cases the projects create new structures, particularly at the lowest level. Instead of disregarding existing or evolving local governance bodies, however, these projects interact intensively with those bodies. Rather than being set up as alternatives to the formal institutions of government, they operate in a complementary and collaborative way. Researchers are keenly aware of fostering such relationships and inclusive arrangements:

The field research team helped the community to form their own organizational structure. A community forestry management committee was formed with local participation, including local authorities like the village, commune and district chiefs, through free election. The new community forest management structure has to stay under the existing Village Development Committee, in which community forestry planning can be integrated in the commune development plan. Doing this can address the issue of overlapping roles and responsibilities among commune council members and provides the community forestry committee and Village Development Committee with an opportunity to build relationships and work together with the other fields of the development program. (Royal Government of Cambodia, 2004: 19; italics added)

Of particular note here are those few cases in which the interaction of user groups and local government gave prominence to gender-based differences in needs and interests. In Mongolia, women's interests in pasture management and livelihoods were represented explicitly through women's discussion groups. In the Tam Giang lagoon, researchers made special efforts to recruit women to the lagoon management committees. In Guizhou, women emphasized to village leaders the importance of water supply as a local resource management priority. The research projects provided entry points for identifying gender-differentiated development needs in local government processes.

Instead of simply being standalone new creations, CBNRM institutions become part of local political decision-making, and in some cases report to accountable policy-making bodies. This is one key aspect of the decentralization context exploited by these cases. Specifically, instead of paralleling existing governance structures, CBNRM institutions mesh with them, renegotiating roles and rights. This is possible because of the fluidity entailed in the ongoing decentralization initiatives. Where governance structures are well established and policy flux less pronounced, as in Guizhou province in China, the integration of CBNRM processes and structures with those of the government is considerably more difficult.

Strengthening local voices

Participatory research which builds local NRM institutions strengthens community organization and leadership and gives local resource users voice in their dealings with policy decision-makers. Researchers can create opportunities for local men and women to engage with decision-makers indirectly through the profile of their projects or contacts. However, experience suggests that the message is far more powerful if it comes directly from the people, stated in their own terms. By building the capacity of local actors to understand, integrate and articulate the policy implications of local NRM alternatives, PAR builds essential policy feedback opportunities in governance systems.

As Holland, Blackburn and Chambers note (1998: 85), 'it is difficult for decision-makers to challenge the views of the poor'. This is particularly true when the poor are in the same room, when their views are presented in an articulate and confident way, and when their opinions are substantiated with local evidence.

This is especially important in the case of marginalized groups which otherwise would not have any reasonable opportunity to communicate their experiences to policy-makers. As explained in the Ratanakiri case:

Direct meetings of the indigenous people with policy-makers were found to be more effective than contacts mediated by NGO representatives or through other official channels. Policy-makers at all levels seemed to appreciate this, probably because it was clear that the indigenous people were speaking for themselves rather than telling the government what they had been primed to say by development organizations. These direct discussions also resulted in breaking down negative stereotypes about minority peoples which were widely-held by national authorities (John and Phalla, Chapter 3, p. 42).

Through such processes, community members were also able to review draft policies and provide feedback on how these might affect them. As noted in the Mongolia case (Chapter 6): 'Discussions with herder groups on the drafts of national level policy and legal documents was another way to promote feedback on strengths and weaknesses. Changes were made in drafts because of local inputs.'

Several other cases in this volume demonstrate this effect explicitly. For example, the PMMR project made a point of engaging policy-makers directly with local people. Because of their position, even though project team members could not assume public positions on controversial issues, they could identify and connect powerful decision-makers to opportunities for personal inspection and familiarization with the field situation. They could make introductions to local leaders, and turn up the volume of local voices quite a bit. The whole intent of the IIRR policy linkage workshop (Chapter 13) was to provide a forum for village voices to be heard effectively by government officials. Even in the Tam Giang lagoon case (Chapter 4) an important role of government officials in the participatory local planning scheme was to validate and approve local lagoon reallocation schemes, thus demonstrating that they had heard local concerns and endorsed their priorities. These opportunities for direct communication of local concerns in NRM are vital elements of effective governance when natural resources form such an important part of local livelihoods.

New roles in policy reform

Through most of these policy-influencing strategies, the cases also demonstrate how conventional roles for the key actors in resource management policy are changing. Researchers, for example, have conventionally been considered as objective analysts of empirical evidence, publishing results to foster greater discussion which build on existing disciplinary frameworks. However, in these cases researchers served a variety of roles in building local awareness and the capacity to respond to, or even initiate, policy reform. By applying new models of learning which engage local men, women and government officials in generating shared insights, researchers are connecting local practice to national policy. They are facilitating multistakeholder exchanges and negotiation through social learning.

For example, the Cambodia community forestry project (Chapter 11) structured its research teams, site selection and field interventions in ways which facilitated policy-relevant learning for forestry officials, partner NGOs, newly constituted commune councils and village development committees. Local forest user groups and village development councils are experimenting with their new roles as resource managers. Commune councils are experimenting with new roles which allow them to sanction village resource planning and enforcement. FA officials are learning how they can be technical resources and guides rather than focus on regulatory enforcement. Researchers are learning facilitative, rather than analytical, roles from field experience and from interaction with policy stakeholders at multiple levels.

Other cases provide similar evidence. In Bhutan, researchers confronted their traditional roles and practices in the face of learning from participatory research experiences. In the Guizhou case, researchers are explicit about the challenges of encouraging government officials to adopt new roles. The IIRR case is concerned with how to help government officials rethink their conventional resource management roles. In Mongolia, the participatory research is defining new individual and collective roles for pastoralists as resource managers.

However, the case which most clearly demonstrates the ways in which multiple actors have adopted new roles through participatory research is probably the Tam Giang lagoon case (Chapter 4). The authors elaborate how each set of actors has adopted unconventional roles in lagoon resource management as a result of their positive experience with PAR on CBNRM.

Challenges

The successes which a number of the cases report in strengthening the voices of local resource users in communications with policy decision-makers will evolve further. Some reforms have been introduced; new processes are beginning to institutionalize consultation and accountability mechanisms. But while these reforms may have strengthened democratic mechanisms, there remain many governance issues to be addressed. Even strong democracies with a long tradition of open public access and complaint to decision-makers (e.g. India, Philippines) have failed to effectively devolve resource rights and management responsibilities to the benefit of local communities (Sarin et al., 2003).

The successes reported here belie the complexity and non-linearity of the project experiences. A close reading of the cases demonstrates that most of these successes were several years in the making. For example, the Tam Giang lagoon research team had five years of research experience, which enabled them to diagnose the inequitable impacts of aquaculture policy. But it was only after the enabling co-management policies were introduced that their participatory research experience provided a platform for a robust community response and a new model for implementing this policy. Most of the other cases similarly tried many ways of consulting and sharing their research lessons, often with limited success, before they identified the key policy issues or the appropriate linkages and mechanisms to leverage policy impacts in their own context. Through more systematic analysis and comparison of these lessons, field practices can be strengthened, experiences shared and project efforts directed more effectively.

Many of the research teams are also only beginning to explore the latitude for using decentralized NRM to better respond to the needs of women resource users. While resource decentralization policies of one form or another were widespread and attracted considerable attention, policies which linked resource management to gender and equity concerns were not, and hence provided less opportunity for interpretation and implementation. Few of the projects identified barriers to the security of women's resource assets and rights which might be strengthened through policy reforms. However, several cases built women's awareness, knowledge, skills and confidence, as well as their capacity to articulate such interests.

Conclusions

The essential argument of this chapter is that the research cases presented in this volume have had a surprising degree of influence on policy formulation and implementation. In part, this points to the growing recognition by the project research teams that local innovations required enabling policy changes, and that local innovations alone would not be sufficient to address the problems communities face in NRM. But it also appears that many of their strategies for policy influence were successful precisely because they started with local PAR. These cases provide examples of how PAR was able to generate local successes and thereby to influence policy reforms.

It is helpful for CBNRM practitioners to understand the potential and mechanisms by which field experience can shape policy, as a guide to strengthening their own practice and building a more supportive context for sustainable and equitable resource use. We argue that there are several important reasons for the outcomes reported here, and that CBNRM practitioners can use these lessons to better diagnose and intervene in policy contexts. One of the factors is the extent of contextual policy flux, particularly around decentralization initiatives. These are ubiquitous, but they offer contexts for CBNRM experimentation of varying potential. Local practice which engages resource users and government officials in social learning about innovative solutions to essential livelihood problems or conflicts will challenge the assumptions of many of the actors involved. Innovative local practice can have a striking influence on decentralizing policies in three situations in particular: when policy is still taking shape; when the local evidence demonstrates that policy commitments are not being fulfilled; or when local government is interested but lacks the capacity to implement policy effectively.

The connection between local participatory innovations and policy change is not a simple cause–effect relationship. Policy processes and the influences which feed into them are complex, idiosyncratic, opportunity-driven and nonlinear. Yet the introduction of new information, especially when it challenges fundamental assumptions or contradicts long-standing simplifications, can open doors to new perspectives.

These cases influence policy in several ways. While the most obvious is that, in some cases, the research provides direct inputs to policy formulation, perhaps the most important influence is on the actual implementation of policy. Here, participatory local research which engages government officials and local resource users in social learning offers potential for improving governance and increasing the effectiveness of policy reforms. These research cases illustrate strategies which respond to contextual opportunity, relying on astute leadership and documented insight from local learning more than careful planning or analysis.

However, some strategies can be fairly widely transferred, such as the engagement of government officials in the learning process, and the strengthening of connections between the research (learning) team and networks of policy influence. Effective participatory research also builds both the confidence and the voice of local resource users in their own interactions with government agencies. An important element of facilitating policy change through participatory research in these cases is in the adoption of novel roles by key actors in the process. By exploring and practising these new roles, development actors are redefining conventional relationships which constrain policy options and are thereby opening doors for innovative alternatives.

These cases demonstrate the difference between investing in research as a product (that is, as analytical conclusions and reports) and investing in research as part of a process of action and change. As others have also concluded, it is the process, and the strengthening of specific actors within that process, which leads to policy influence, more than the product itself (Holland, Blackburn and Chambers, 1998).

These cases provide evidence to policy-makers, inform strategy for advocates and direct policy implementation by government agents. Indeed, it is likely that effective policies fostering the emergence of CBNRM can be designed only through evidence from PAR. This approach provides a unique set of tools for small-scale policy experimentation and for social learning of participants around issues of contradiction and conflict. PAR provides practical examples that can stimulate enabling policy reforms and transform the roles of researchers, officials in line agencies and local government, and men and women who rely on the resources. This combination of impacts allows new collaborative resource management processes to take root and grow in diverse contexts.

Document(s) 23 of 24