Document(s) 20 of 24

Introduction

CBNRM advocates often take the concept of community for granted (Agarwal and Gibson, 2001: 8; Mansuri and Rao, 2004: 13) as the basic form of social organization in the countryside in regions like Asia (Lynch, Talbott and Berdan, 1995; Vandergeest, 2004). At the same time, however, some of the more convincing criticisms of CBNRM, those based on detailed fieldwork, argue that one of the most serious practical problems inherent in CBNRM practice revolves around unexamined assumptions about the nature of rural communities (Li, 1996, 2002; McDermott, 2001). These critiques have had some impact in policy circles and among development practitioners (Mansuri and Rao, 2004), calling into question the value of community-based approaches to natural resource management and development.

The effectiveness of the projects described in this volume and the continued importance of the concept of community in more people-centred development work, however, point to a need to go beyond the critiques, so as to develop a more robust and realistic concept of community than what currently circulates among uncritical advocates of CBNRM. A more robust notion could better highlight its potential as an approach that is not just about micro-level intervention, but which also contributes to restructuring broader development practices, power relations and the distribution of economic benefits. This stronger and nuanced concept of community is, therefore, an essential element of more effective practice in development work.

In this chapter, I examine the concept of community as it emerges from the case studies in this volume. I will pay particular attention to how these communities were made, how they were structured and what they did. The accounts of community contained in this volume on CBNRM in practice can be compared and contrasted with how community is typically understood in the academic and practitioner literatures on CBNRM.

I will show that that the studies described in this volume may not give much self-conscious attention to the concept of community (with a few exceptions). Nonetheless, they present a complex notion of how communities are made, what they do, and how they facilitate collective action. I will highlight three aspects of these complex communities:

1. these projects brought together both local and trans-local networks of actors to make communities;
   
   
2. the communities can be usefully understood as ways of mobilizing collective action around common projects; and
   
   
3. by doing this, communities become collective actors, active both locally and in broader networks.

My analysis points to how CBNRM can do much more than simply organize interventions for the management of natural resources. Understood broadly, CBNRM can also make a substantial contribution to a broader rethinking of sustainable development, one that takes account of ways that rural people can act collectively. This broad perspective also points to the way that the communities described in this volume were the product of hard work by rural people, project teams and government actors. Because of this, they represent fragile accomplishments that need continued effort to be sustained.

I will use the term 'CBNRM community' to highlight how the communities discussed in these case studies were the product of CBNRM research projects. I should be clear that I am not arguing that there was a vacuum before organizers came in and made a new community. The point is that the projects remade or transformed existing institutions and networks into what I call CBNRM communities, based on a combination of trans-local methods and local practice. In other words, the studies described in this volume produced specific kinds of communities, whose characteristics I will attempt to summarize.

Critical perspectives

CBNRM and participatory development are often framed as alternatives to top-down development planning. Over time, this vision has been undermined somewhat as large development institutions have mainstreamed community-based development (Cooke and Kothari, 2001; Mansuri and Rao, 2004). Proponents of alternative approaches link development based at the local level with sustainability, social justice and resistance to top-down, mainstream development. These views are often expressed as the prioritizing of local communities in development practice. That is, alternative development is often based on the notion of the community as a social group bound by locality, with the idea that localities should be the starting point for development.

CBNRM programmes can have many different motivations, too many to review here. If we limit our attention to alternative development, however, at its core CBNRM means finding ways of increasing the participation of rural communities in resource management, especially in environments where many resources are claimed as state property by government resource agencies: forestry, water, fisheries. The argument is that rural communities have a more intimate knowledge of their localities than state resource agencies. They also have a greater stake in managing resources sustainably, because their livelihoods depend on it. These arguments complement ethical arguments about the rights of communities to use resources in places where they may have lived for many generations. CBNRM, along with the protection of common property institutions, is promoted as a way of helping the poor and vulnerable in society to gain more secure access to livelihood resources (Johnson, 2004). In a broader economic and political context in which both socialist collectivization and centralized state NRM have been largely discredited, CBNRM also promises to address a need for finding ways of managing resources through decentralized collective action (see Chapter 2).

What is notable in all this is the way that advocates of CBNRM and common property often frame local communities as existing outside state and market institutions. The more political scholar activists often go further and argue that states and markets are the main threat to local communities and common property management (Li, 2001; Vandergeest, 2004). However, even among those whose views about market integration and joint forest management with state agencies are more positive, the basic assumption still tends to be that local communities are outside these institutions.

Many elements of alternative development theory and practice have been subjected to critical examination (Li, 1996, 2002; Brosius, Tsing and Zerner, 1998; Bebbington, 2000; Agarwal and Gibson, 2001; Hart, 2001). The most useful critiques acknowledge that simplifications are necessary for effective policy advocacy on behalf of rural people whose access to resources may be threatened by state claims or private interests, but go on to cast doubt on whether the particular simplifications associated with CBNRM and common property advocacy might not cause more harm than good for many rural people (Li, 1996, 2002). The critics present evidence that rural people such as migrants to upland areas who do not fit into the kinds of stable local communities assumed by this approach can be marginalized or erased from consideration. They argue that this approach is often characterized by a lack of attention to gender, class, caste or racial differences in rural areas, and that CBNRM can have the effect of diverting resources to local male elites (Beck and Nesmith, 2001). And they point out that advocates of local development often ignore evidence that many rural people desire greater access to both markets and the state (Li 1996; Brosius, Tsing and Zerner, 1998; Vandergeest 2004). Some of the critics suggest that the older language of agrarian reform might be a better approach to policy advocacy for the rural poor. In particular, the problems of rural people might be better addressed through land reform and enhanced citizenship rights than community rights.

As I will outline in detail below, the chapters in this volume suggest that the practitioners who prepared these case studies have developed a better practical understanding of rural communities than that found in much of the literature on CBNRM or in advocacy circles. The critique outlined above, moreover, has been around now for close to 10 years, and a good part of the practitioner network has responded with increased attention to the issues it raises. As I describe below, rural difference is now addressed through gender analysis, wealth rankings and projects targeting the poor. These methods have become part of the repertoire of CBNRM projects that try to address inequalities, although there is cause for some scepticism about the impact of these exercises in countering entrenched local inequities (Chapter 15).

The considerable efforts put into increasing community capacity, building confidence and enhancing social cohesion suggest that in practice the projects in this volume do not take local communities for granted as pre-existing their projects. The case studies also demonstrate that community-based approaches can be effective ways of addressing some of the key problems facing rural people. For example, in situations where entire communities or villages do not have secure property rights, these insecurities are better resolved through community-based collective action than land reform. This does not take away from the relevance of inequality in property rights within communities, but does suggest that programmes to resolve these inequalities need to be undertaken within a broader understanding of the marginalized position of many rural communities vis-a-vis state and other claims on land and resources.

At the same time, the tension between simplifications that seem necessary for effective policy advocacy for the resource rights of local people on one hand, and the inherent dangers posed by ignoring the complexity of rural communities on the other, is not going to go away. The basic issues emerge over and over again. Thus it is not enough to simply expose and criticize the romanticization of the local that is inherent in alternative development arguments. It is also important to find ways of advocating for local or community rights in the face of state claims or private development that are not so susceptible to the problems outlined by the critique. What these studies demonstrate is that the agendas of agrarian reform that focus on rural inequalities and complexity on one hand, and community-based development efforts on the other, are not necessarily incompatible. It depends what is meant by community, what these communities do and how inequalities within communities are addressed.

Constructing communities

Some reflection and a brief review of literatures on communities and community activists show that there are many different ways of understanding community. Some of these include the following.

• People who live together in a specific locality or territory such as in rural villages and urban neighbourhoods (Agrawal and Gibson, 2001: 8–9).
  
  
• The existence of dense or frequent forms of social interaction, a concept now often linked among development practitioners to the notion of social capital (Meinzen-Dick and Di Grigorio, 2004). More recently, this approach has given considerable attention to social networks that can stretch across localities. For example, trans-local migrants often organize their lives across borders; and electronic communications facilitate the formation and maintenance of diverse networks across space.
  
  
• Shared norms, linked to locally made institutions for organizing collective action to enforce these norms (Agrawal and Gibson, 2001: 10–12). Examples include prohibitions on the use of spirit or conservation forests, or collective institutions for distributing access to land for swidden farming.
  
  
• Local administrative institutions created by the state, in particular, administrative villages.
  
  
• Collective imaginaries, or the way that certain collectives imagine themselves having commonalities or collective identities. Examples include national communities, ethnic groups, indigenous peoples and peoples affected by dams.

This list indicates that there is a wide range of ways of understanding community. All have some relevance to the case studies in this volume, and in practice there is considerable overlap. But for the purposes of this chapter, it is worth thinking through the contrast between what might be called the locality-based versus network approaches to understanding communities.

Locality-based approaches to community

Locality-based approaches begin by asking if people living in close proximity for a long period of time can develop shared norms and collective institutions for managing resources. This understanding often allows for the incorporation of the concept of locality into community through the term 'local community'. It can also be tied into territorial definitions of local communities. In other words, a community consists of the people living in a given territory. Finally, it is often linked to the concept of tradition through the assumption that many local communities in rural areas have developed stable norms and institutions by virtue of living with each other and facing collective challenges over a long period of time. At a smaller scale this approach has led to many community studies in the disciplines of anthropology and sociology,¹ although it can be scaled up to the idea of homogeneous national communities and traditions.

Community-based development often draws on this tradition of community when studies assume the persistence of stable and relatively homogeneous rural communities. It can be argued that these traditional communities are capable of undertaking development initiatives for the common good. Many proponents of alternative community-based development add that community stability depends on maintaining some kind of boundary or separation between local communities on one hand, and institutions external to the community, especially states and markets, on the other hand. Boundaries and exclusion are also tenets of common property theory (Chapter 7).

Network approaches to community

The second approach to communities is currently gaining influence in academic research. The focus is less on the boundaries separating local communities from external institutions or other communities, and more on how these are connected and how they construct each other through these connections. Or, put differently, the focus is less on the attributes of communities that emerge out of local interaction, and more on social relationships emerging from networks that span localities (Massey, 1993; Raffles, 1999; Li, 2001). Although such networks are located in space, they are not necessarily contained by spatial boundaries, nor are they located in just one locality. For example, research on transnational migration now pays a lot of attention to how communities can be made across localities that are often separated by national or other borders. The term 'trans-local' is becoming popular as a way of pointing to how these networks cannot escape locality, while simultaneously underlining how they are also not contained by just one locality.

In practice, these case studies suggest that both of these approaches are relevant for CBNRM. Most authors do not discuss whether they understand community in terms of bounded territory or social networks. However, all the studies described in this book began with the assumption that communities were at least partly defined by geography. Most often, community was understood in terms of territorially based administrative units like villages or communes. This is reinforced by the way that the collective NRM practices that give CBNRM its name usually rely on clear and exclusive definitions both of the community and the resources to be managed. The idea that CBNRM is in part about creating boundaries around localities in order to exclude resource users who are not community members comes through clearly in a number of papers in this volume. The most explicit statement was made by Ykhanbai and Bulgan, in their chapter (Chapter 6) on the co-management of pastureland in Mongolia. They write that for their paper 'community refers to a geographical area containing a number of herder households'. A key issue this project had to grapple with was when and how to exclude herders who were not community members. Tubtim's chapter (7) on community fisheries in Laos points out that the idea of collective benefits is widely accepted in rural Laos, but refers exclusively to a single village. In this situation, 'when a particular resource becomes scarce or valuable in the market, village boundaries can be easily brought into play to claim exclusive rights'.

At the same time, the case studies give considerable attention to the way that the projects were implemented through social networks of various kinds, some of them local, but often trans-local. Many of the case studies in this volume emphasize the importance of networking, with NGOs, neighbouring communities, state actors and so on. Even the territorial boundaries of communities can often be understood as the product of social networks, so that the second approach can encompass the first. In other words, administrative villages, herder groups and other territorially defined resource user communities are defined through social interactions with states, or with external donor projects.

My reading of the case studies in this volume suggests another definition of community that might be particularly useful for CBNRM, one that offers a way of incorporating both approaches to communities. All these case studies mention community not only in terms of networks and territories, but also as collective action. The cases outline how the projects facilitated 'voluntary action taken by a group to achieve common interests' (Meinzen-Dick and Di Gregorio, 2004). More specifically, these case studies all describe instances of people acting collectively to manage resources and to achieve a series of other collective goals related to NRM. I should immediately add, however, that interests in practice need not be common for this kind of collective action to occur (Chapter 7), and that this does not imply that all members of a community participate in or benefit from collective action in the same way (Chapter 15).

I am not trying to suggest that the notion of community as collective action is the correct one; only that it is a particularly useful one when thinking about CBNRM communities. Nor does it imply abandoning other ways of understanding community. Rather, we can work them into this way of understanding community by paying attention to the ways that collective action can be facilitated and structured both by the spatial arrangement of people and by social networks. In other words, the focus on collective action resolves the tension between geographical and network approaches by directing attention to a community-building process that can exhibit both network and geographical characteristics. It is also more dynamic and helpful to practitioners, because it draws attention to what people actually can do collectively. It focuses our attention on the ways that collective action can be mobilized to intervene in managing common pool resources (CPR), or secure better access rights to resources that are important for livelihoods. It thus fits well with an approach that thinks about community in terms of networks of actors. Collective action undertaken by communities can be understood in terms of both the local networks comprising the group and their specific roles, and how communities become one actor in a broader network of actors that include state agencies, private interests, development agencies, NGOs and others.

It is through collective action that groups of people become community-based actors. In the section below, I review in more detail what communities do when they become actors. If we take this approach, the relevant questions become reformulated into questions around the specific ways that people get organized to act collectively. For example, how is collective action facilitated by spatial arrangements, local histories and existing institutions for collective action? What is the role of the outsider in collective action? How does external financial support facilitate collective action? How do the links between a rural community and broader networks of actors facilitate or structure collective action? How might these interactions with broader networks affect policy formation and implementation (Chapter 17)? How might collective action be institutionalized and routine? How does it reflect power and inequality within community (Chapter 15)? How might it change these inequalities? How might collective action address broader inequities between communities and other powerful actors?

This approach also helps link CBNRM to practitioners and scholars concerned with identity. When a group of people becomes an actor, these people are creating a shared identity. Usually, identity is framed in terms of community, but collective identities can also emerge from other identities and roles: women's groups, indigenous peoples, the village poor, and farmers. It is worth briefly highlighting one of these broadly speaking ethnic (or racial) identities. Only a few of the case studies in this volume discuss ethnic identities explicitly as having an impact on resource management and property rights (examples include the Philippine Cordillera in Chapter 12; the two Cambodian community forestry cases, Chapters 3 and 11; and upland Vietnam, Chapter 5). Diverse ethnic identifications are a fundamental feature of rural life in Asia, and they are often linked to resource management and property rights. This is especially true in the upland and coastal zones where CBNRM projects are usually located.

In the rest of this section, before I move to what community actors do, I will elaborate upon some of the lessons that can be drawn from the case studies which describe how we think about community. I will discuss a series of characteristics of these communities, some of which overlap, but each of which is worth highlighting as important to developing a more realistic and robust concept of community in CBNRM practice.

Diverse actors

I have already mentioned that critics point to how CBNRM programmes and advocacy often assume that rural areas are comprised of stable communities, and that they often overlook or fail to address gender, class, caste or racial differences in rural areas. With respect to stability, for example, many if not most rural people in Asia have been mobile in the recent past, due to war, economic changes and state policies. It is usually the migrant communities which are blamed for resource degradation, have the most tenuous property rights and are poorest.

Nevertheless, this book suggests that CBNRM can work effectively in communities that have settled in their locations relatively recently. Nong and Marschke (Chapter 8) note that in coastal Cambodia it was the in-migrant fishing villages, consisting of households displaced from other provinces, who were initially most interested in participating in the CBNRM activities. Because they were new to the coast, they had not developed strong local institutions for managing resources. Their prevailing livelihoods (charcoal production) were insecure because they were illegal, and the project offered them a way of gaining legitimacy. The project in Hong Ha commune in Vietnam also worked with recently resettled communities, which similarly expressed a need for improved livelihood and resource management practices in their new environment.

When we think of communities in ways that do not rely on the assumption of the long-term stability of a homogeneous set of people, then we open CBNRM to a much broader range of practice. Now we can include communities that may not exhibit the sorts of characteristics associated with classic conceptions of stable social interaction and collective institutions. In particular, we can incorporate ways of doing CBNRM that target those who often need these activities the most, as Nong and Marschke's case suggests.

In cases where there were community-based institutions for managing resources prior to the research projects, the projects have had varied relations with the prior institutions. Communities were often self-consciously built on existing institutions (China, Bhutan). But it is also possible for government or NGO-led programmes in support of community resource rights to undermine existing community institutions, replacing them with less effective administrative institutions (Agrawal, 2001). In this book, Mendoza et al. describe how the community definitions created by certificates of ancestral domain in the Cordillera existed in tension with the ili, the traditional community. The certificates were awarded to a municipality, a territorially defined administrative community, but the municipality in turn consisted of several ilis. As indicated by the title of the chapter, their project was concerned with harmonizing the actions of these two communities. This chapter makes perhaps the clearest case for paying careful attention to diverse and changing expressions of community, because it was the definition of community that was built into the ancestral domain policy that produced the problems that the CBNRM project needed to resolve.

Similarly, in the Tam Giang lagoon project in Vietnam, the government was forced to think about how to constitute the collective entity that could hold collective resource rights. None of the existing legally recognized collective institutions (Farmers' Union, Women's Union and so on) were appropriate, and eventually one of the project's pilot activities was to form a user group-based community, the Fishing Coalition. These cases illustrate why writers of laws and policies need to work with legal definitions of community that allow for considerable leeway in how communities can be made at the local level. Sometimes they are best based in local government institutions, but sometimes they are not.

There is another sense in which CBNRM involves diverse kinds of communities and other actors. Most of the projects in this book began with administrative villages or communes as the unit around which people organize collective projects. But they very quickly became much more specific about who comprised the collective actors, and how they built on collective actions that preceded the project interventions. For example, the most common actors are committees organized around a particular task, which create management plans or build infrastructure. Sometimes actors are groups involved with many different kinds of activities and goals, such as marketing, crop improvement, credit or management of a specific resource, such as mobile-gear fishers in the Tam Giang lagoon in Vietnam.

This does not mean that such committees or groups actually comprise the community. Instead, committees typically act on behalf of some broader group of people that they call the community. Nonetheless, this focuses our attention on the way that many community-based activities are not the collective action of everyone living in a given area. Not everyone participates equally or in the same way in community actions. As Beck and Fajber make clear in Chapter 15, forms of inequality represented by economics, power, gender and other social relations are part of community. Therefore, careful examination of the local as well as trans-local networks of actors who are involved in collective action takes us beyond assumptions that communities are simply everyone living in a geographic area, although that assumption can remain a useful starting point.

In some countries, CBNRM activities have become so popular that village committees for mobilizing collective action are proliferating, creating problems with coordination, duplication, and organizational fatigue (Marschke, 2004; Mendoza et al., Chapter 12). The Cordillera project addressed this problem through the formation of yet another collective institution, an interagency committee for NRM. Below I discuss how making CBNRM communities requires work and effort. This suggests that it is important for practitioners and policy-makers to pay careful attention to the demand placed on rural people by community-based initiatives that rely on local participation or work. Heavy demands on time, the scheduling of meetings and difficult logistics all represent complicating factors in the effort to include women and the poor in community-based work. For instance, Mongolian women were unable to leave their domestic duties to make 15-km trips in midwinter to attend meetings.

Cooperation and conflict

Some chapters go further, and discuss conflicts among the resource user groups or among individuals who comprise rural communities. The chapter on coastal Vietnam is exemplary in highlighting the diversity of actors and violent conflicts over resource access among different user groups. In these cases, the kinds of collective action that defined community became highly contentious. Other chapters that address conflicts or disputes over resource access within or among communities include those on Laos, Cambodia, Mongolia and Bhutan. These chapters demonstrate that the assumption of harmonious communities is not essential for CBNRM to be effective. Indeed, the most effective projects in terms of facilitating collective action to manage resources, secure tenure access and enhance livelihoods are those that aim to understand how conflicts among rural people can be based on inequalities. Such discrepancies can be in access to resources, or in the ways that economic and political changes may threaten the resource access of some groups while promising benefits to others. We can go so far as to say that collective action does not need to be based on shared interests at all. Instead, as Tubtim explains in her case study in Laos (Chapter 7), CBNRM can be a platform allowing for a single process involving actors with diverse interests and worldviews. In this sense, CBNRM can be a way for diverse groups to find ways of working through conflicts of interest and worldviews, to achieve multiple goals.

Power relations and inequality

If we pay attention to the way that communities are not necessarily based on common interests, but may also be multiple collectives that reflect economic and power differences among rural people, then we are taking an important step towards addressing many of the possible problems identified by critics of CBNRM. Many more mainstream CBNRM projects give this only superficial attention. However, an examination of this volume's case studies demonstrates how the CBNRM repertoire includes methods allowing practitioners to address the need to work differently with various groups in communities. Attention to the gendering of livelihood practices and resource access was common to many of these projects. Sensitivity to differences in wealth was expressed through the wealth-ranking exercises that have become a standard feature of this approach, one used in most of the projects described in this volume, although not always mentioned in the case study chapters.² In other words, the template for doing CBNRM at the village level does not ignore differences within communities. The studies in Bhutan and Mongolia mention that the project formed groups based on relative wealth in order to address the specific needs of each group and as a way of targeting the poor. Both projects in Vietnam devoted considerable effort to identifying and working with poorer people, while the Lao fisheries project also highlighted different impacts the project had on wealthy and poor villagers.

Identifying, organizing and working with the rural poor is one thing, but projects that aim to actually change how inequality can be based on unequal access to resources are another. In the next section on what communities do, I discuss the question of whether CBNRM activities can or do address the causes of inequality in the context of understanding CBNRM communities. Beck and Fajber discuss this more fully in Chapter 15.

The cases paid less explicit attention to issues of power in rural communities. To my knowledge, an assessment of power relations has not yet been worked into standard field repertoires for CBNRM more generally. Therefore, it is not surprising that these cases did not apply standardized techniques to trace and address power differentials, in a manner similar to those addressing inequalities of wealth. But at the same time, the cases in this volume demonstrate it is difficult to avoid engaging with local power relations if researchers aim to target poorer villagers and inequitable gender relations. Case studies that address power relations in CBNRM include those on Laos, coastal Vietnam and coastal Cambodia. Other chapters mention local power indirectly, for example, in discussions of how collective institutions are organized, and most often, in struggles that occurred in communities over specific projects. Tuyen et al.'s chapter (Chapter 4)makes explicit how CBNRM practices in the coastal lagoon involved finding ways of mediating struggles among different user groups. Tubtim (Chapter 7)mentions how in Laos it was normally the village committees that were influential in making decisions, and that not all villagers were fully active in village meetings. In Bhutan (Chapter 10), upstream water users benefited from their advantaged location to gain preferential access to water, and resisted attempts to find more equitable ways of allocating water supplies.

Articulation with government institutions

Many of the studies described in this volume are concerned with local resource rights, often in opposition to the claims of state resource agencies. This should not be taken to mean that rural communities are created in opposition to the state and market. To the contrary, the projects address these sorts of conflicts not by keeping the state out of the community, but rather by inviting the state into them. The goals usually include finding ways of convincing state actors to provide formal recognition of informal management and access rights.

Along the way, community-based institutions for organizing collective action often broaden to incorporate non-community-based collective actors to negotiate conflicts, allocate resources or draw up management plans. The coalition of actors involved typically includes village authorities, village-based user groups, project staff and local governments. For example, in coastal Vietnam, planning for allocating resource access began with a workshop that brought together project staff, fishing user groups, the village leadership, the commune government and representatives of district and provincial departments. As Tuyen et al. write, the 'new participatory planning process ... emerged from interaction among resource users and other concerned stakeholders'. Other chapters also describe how community resource access and management are organized through institutions that bring together CBNRM teams, government authorities and rural people. These cases include those on the Cordillera project, Mongolia, Laos, China and all the Cambodia cases.

The ties between rural communities and the state that characterized most of the projects described in this book were reinforced by the way most projects worked with the village or commune administrations. These institutions are the lowest level of government civil administration, and although they are usually comprised of elected villagers, they nevertheless represent the state in the village, as Tubtim notes in her Lao case study (Chapter 7). In this way, the projects accepted that these hybrid state-village institutions were an important part of how communities organized collective action.

State institutions affected the formation of community in other ways as well. For example, resource management plans were often based on what forestry departments or other government authorities consider legitimate. Although relations between local government officials and village/user groups described in these chapters were sometimes difficult, the cases capture many examples of how villagers, village authorities and government officials worked together to achieve collective ends, or worked together on common projects even when they had different interests.

One reason for the spread of CBNRM is that practitioners have worked out a repertoire of practices and institutions that can make sense both to villagers and to government officials, enabling cooperation even where objectives may differ. Nonetheless, tensions exist. For villagers, for example, enlisting government assistance requires reconciliation with official definitions of community, technocratic criteria for resource management, and in some cases accepting a government role in managing what villagers might previously have considered their own resources. Government officials must recognize that villagers have knowledge and skills that can be helpful in meeting some of their objectives, although other objectives may be compromised. But overall, the close interaction between state administration and rural people described in this volume belies images of rural communities that are separate from – or in simple opposition to – local government (Chapter 17).

A concept of community that allows for conflict also allows us to see that in part, rural communities are made through interaction with state institutions, even when that interaction includes conflict. Without exception, CBNRM communities described in this volume were constructed through complex relationships between government agencies and rural institutions, instead of outside government.

Engagement with markets

The other kind of opposition that sometimes characterizes CBNRM advocacy is one between rural communities and common property on one hand, and private actors and markets on the other. Are local communities opposed to markets? Is common property incompatible with private property? It is true that in many of the cases in this book, communities acted to curtail the expansion of private property rights at the expense of common property resources. For example, one of the goals of the Cambodian community forestry project was to create and formalize community forests so as to stop timber poachers and agrofood entrepreneurs from privatizing and destroying forests. The lagoon project in Vietnam worked to maintain community access in the face of the enclosure of lagoon resources. In Mongolia, community control of pasture was strengthened to help solve problems caused by unregulated private use.

A single-minded focus on these apparent oppositions, however, would misrepresent the way that communities often form around ways of regulating or improving market access. There is no inherent contradiction between market engagement and the strengthening of community control over common property.

Many of the cases in this book describe projects designed to increase the productivity of common property or privately held resources so that villagers can get more income by having more products to sell in local markets. What made the work specifically CBNRM was the application of collective sustainability objectives to activities aimed at increasing market integration. Projects in upland Vietnam, Mongolia, China and the Philippines worked to enhance commodity production and monetary income. Their efforts included the improvement of specific commercial crops, better pasture management, rotating credit schemes and collective marketing. For example, in Laos, the government and village authorities involved in the communal fish pond understood the project represented an investment in increased productivity. Fish caught through communal fishing was sold in the market and the income used to fund a primary school, while privately caught fish were also sold, especially by poor women.

In summary, the projects described in this volume aimed to find collective ways of controlling unregulated market activities and stopping unsustainable use of common property resources. This is not the same as saying that common property and local communities are opposed to markets. On the contrary, it is impossible to understand the formation of communities without addressing the way in which rural people are already integrated into markets and usually want to increase their engagement with appropriate community-based regulations. This implies that advocates and practitioners of CBNRM should devote increased attention to the way that power and benefits are distributed in market processes. It also suggests that a community-based approach to market engagement might include facilitating the collective regulation of commodity production, for example, through community involvement in certification. In particular, practitioners might give more attention not only to ensuring that market processes do not undermine resource access among the poor through privatization of common property, but also to exploring ways of instituting markets in ways that allow for the increased capture of benefits by the poor (Ribot, 1995; Watkins and Fowler, 2003).

Articulation with CBNRM networks

In this chapter, my basic argument is that CBNRM communities are made through CBNRM projects, which are in turn part of broader networks that research, practise and promote CBNRM. Practitioners in these networks develop new techniques and analyses, which then move across sites, to be appropriated and further refined in new sites. All of this produces a repertoire for practice: how to facilitate the collective action that characterizes a CBNRM community. This does not mean that there were no communities before the projects, but simply that the projects draw on models that circulate through CBNRM networks to remake existing institutions for collective action, or in some cases build new institutions. Every case study in this volume, with the possible exception of the policy-oriented writeshops presented by O'Hara (Chapter 13), describes how project teams worked with rural people to build institutions, increase their confidence, plan NRM and so on. What made these projects more than simply another way of imposing new institutions on rural people was the commitment to meaningful participation and local leadership. Participation was organized by the CBNRM project teams, but was also a way for villagers to work with project members to identify forms of collective action appropriate to their circumstances.

This idea can be illustrated by the case studies located in transitional economies. In these sites, the communities created by the research projects often filled voids left by the dissolution of collectivization. Although collectivization was often resisted or failed to take hold, especially in Laos and Vietnam, in other cases it superseded local institutions for managing resources.

For example, according to Ykanbai and Bulgan (Chapter 6), in Mongolia customary management institutions that had been in the control of feudal officials, clans and tribal groups were dissolved in 1921 and replaced with state ownership and centralized management. In 1992, when the government moved to a more market-oriented system, herders were left with very little collective management. In this situation, herders needed the research project's outside intervention to introduce CBNRM as a form of co-management. Similarly in Guizhou, China, the end of the rural commune system in which farmers were organized to work and manage resources collectively left villagers with few incentives and few formal institutions for this purpose. Again, CBNRM project activities helped fill this gap, in part by building on informal collective institutions.

In some mainstreamed approaches, CBNRM networks can affect local action by rigidly imposing a formal model without regard for local context. While this may often be the case when CBNRM ideas are applied broadly by government agencies or larger institutions (Vandergeest, 2003), the case studies in this volume demonstrate how a flexible model can be applied sensitively in response to varying local circumstances. When applied like this, projects often adopt new or unexpected meanings and opportunities. One important dimension of this flexibility involves the many ways that local communities can be made or remade through participatory research and collective action.

Outcomes of hard work

A striking feature of the chapters in this volume is just how hard the project teams, rural people, programme officers, NGO staff and sometimes even government officials worked to achieve the many results described by the authors. CBNRM project staff spent much time traveling to and staying in rural areas. As well, rural people travelled to and attended meetings, did resource assessments and mapping, planned resource management, enforced rules and regulations, performed trials of specific crops, visited government offices and accomplished many more tasks. In the case of Mongolia, community members sometimes had to travel 15 km on horseback in winter temperatures that even most Canadian participants in this collective effort had never experienced. The point is that communities described in this book do not just exist naturally; rather, they are the product of considerable effort or work.

This observation is important not only because the project teams participating in this book should be recognized for their hard work and dedication. It also points to a fundamental characteristic of CBNRM communities: if communities are the outcome of ongoing work, it must be recognized that when people stop working, these communities can disappear. This does not mean that the memories or multilayered institutional legacies of these communities disappear or that they cannot be recovered. What it does mean is that communities are often fragile and unstable. As the kinds of work people put into making communities changes, the characteristics of the communities also change. When projects lose their funding, or when rural people lose their commitment, communities may decline. When the work involved in making communities becomes institutionalized, routine or supported by ritual practices, as described by Tubtim regarding Laos, communities can be stabilized or even become inflexible.

A central concern of many of these projects, which was discussed at length in several cases, is exactly how to stabilize communities so that they endure even without the input of project funding. According to O'Hara (Chapter 13), the projectization of community forest management in the Philippines is more about sustainable development assistance than sustainable forests. These projects are kept going by the incentives created by external funding, and they disappear when the funding ends because the larger problems based on disincentives in the policy environment are never addressed. Even when community-making work becomes institutionalized, routine or turned into ritual, it is easy to overestimate the stability of these practices, as earlier CBNRM advocates tended to do. Thus, the crucial characteristic of the communities described in this volume is that they are almost always the tenuous accomplishment of considerable effort, and because of this, they are unstable and in constant transformation.

CBNRM communities in action

I mentioned above that what CBNRM does more than anything else is facilitate collective action. This raised several questions. What kinds of collective action? What is specific about the collective action that defines it as being CBNRM? Since the emphasis in this chapter is on conceptualizing community, I will keep this discussion short as it is intended primarily to illustrate the broad range of collective activities in which CBNRM communities have engaged in these examples. In fact, CBNRM communities are actors both in relation to local issues and to trans-local networks, which include markets, government and so on. I will stress the latter because that is what is emphasized in many of the chapters, and it helps to counter the idea that the impacts of CBNRM work are limited to micro or local levels.

Although the activities covered in these case studies vary too widely for me to describe them all, some of the more important actions include:

1. managing resources;
   
   
2. holding and exercising property rights;
   
   
3. becoming a vehicle for multiple development activities and other types of collective action;
   
   
4. changing power relations and inequalities; and
   
   
5. becoming a voice for local villagers so that they can communicate to project staff, government and other actors.

Managing resources

All the projects described in this book were part of a broader programme to promote CBNRM; therefore, NRM was usually a core activity. Because the project was about community-based management, projects were oriented toward managing those resources where it made sense to take a collective approach. These included forests in, for example, the Philippines, Bhutan, Cambodia, Vietnam and China; water supplies for irrigation or household use in China and Bhutan; grazing lands in Mongolia; and fisheries in Vietnam, Laos and Cambodia.

The approach to the collective management of resources taken by these projects emerged in part from what we might call a kind of CBNRM methodological template. It circulates through the programming of the IDRC and, more generally, through community and common property advocacy circles (see Chapter 2). Usually this involves creating formal management plans, as well as doing wealth rankings, gender analysis and so on. But the methodological template also emphasizes the importance of local participatory processes to involve communities and work from local priorities. In some cases, management plans hardly involved project staff at all, for example, in the management plan for the fisheries project described by Tubtim. Resource management plans were also oriented towards doing what was necessary to obtain government recognition and thus formal legitimacy for community resource rights, as in Cambodia, Mongolia and the Philippines.

In the chapter (Chapter 11) on Cambodian community forestry, for example, Kamnap and Ramony detail the process leading to the creation of a management plan. They mention electing a community management committee, doing a forest assessment, creating a management plan with rules and restrictions regarding forest use and obtaining recognition from the provincial government. Plans were usually based on existing forest management and use, which were transformed into the kinds of territorialized zones and regulations required by the CBNRM management plan. However, the creative use of the CBNRM methodological template is indicated by the way communities frequently added many regulations to reflect new forest uses. For instance, in the case of lagoon management in Vietnam, the development of a fisheries management plan meant finding ways of resolving conflicts among different user groups, and involved government officials as well as community members.

The CBNRM template should not be considered only as an imposed form. In many cases the community used its formalization of resource management to address problems or conflicts that had emerged, for example, as forests became more oriented towards commercial use. One element of the management plans I have discussed is the inclusion of mapping the resource's territorial boundaries where very often there had been overlapping uses among different communities. In other words, the creation of a formal management plan often has the effect of defining exclusive community rights to some resources, while excluding other communities and reinforcing territorial definitions of community. In this way, CBNRM contributes to redefining communities in a way that brings out territorial characteristics and, arguably, produces enclosures (Tubtim, Chapter 7; see also Tubtim and Hirsch, 2005).

Claiming and exercising property rights

This activity is closely related to the creation of management plans in two ways. First, property rights, including collective or common property, necessarily imply both exclusions and finding ways of distributing rights, obligations and benefits. I have already described how community-based property claims means excluding neighbouring communities. However, in other cases, it has enabled communities to exclude more powerful claimants – loggers, palm oil companies or the military – as illustrated by the two Cambodian case studies on community forestry.

Powerful actors making claims on community resources not only include extractive businesses, but also resource agencies in governments that have historically claimed jurisdiction and ownership of specific resources. The case studies in this volume mention forestry departments (Vietnam, Bhutan, Cambodia and the Philippines), irrigation departments, fisheries departments (Cambodia), conservation departments, watershed management authorities (Vietnam) and mineral departments (Cambodia). Many of these resource agencies are closely linked to private companies that hold concessions to extract resources or to NGOs and other international actors that provide resources and legitimacy to these agencies. The latter are especially important in the case of conservation and watershed protection. CBNRM emerged in part as a programme to defend community access to crucial resources over which they had weak or non-existent formal property rights. One strategy was to exploit differences within governments, specifically, to approach local authorities (or provincial governments) as a way of getting leverage on centralized government resource agencies.

Inherent in the exercise of establishing collective property rights is the allocation of responsibilities and benefits among community members. Most case studies did not describe these processes in detail, the exceptions being those cases that also paid more attention to inequality within communities, for example, Bhutan, China, Laos and coastal Vietnam. The way that this activity is organized is central to the way that CBNRM either reduces inequality or is captured by local elites for their own benefit.

Property rights require both legitimacy and enforcement. It is because of these requirements that government officials were brought into a project's activities. As the two case studies on community forestry in Cambodia make clear, government authorities are often reluctant to accept community claims to resources unless the community involved can demonstrate that the resources can be managed in ways that satisfy what they understand to be scientific criteria. For example, forestry management plans need to use zoning categories and regulations that are recognizable to professional foresters, or they will not be accepted as legitimate. This is especially clear in the cases of Bhutan, Cambodia and the Philippines.

Few of the case studies describe in much detail how property rights were enforced in communities. Nonetheless, as the case studies in China and coastal Vietnam and Cambodia show, the lasting impacts or failures of the CBNRM research projects often followed from the community's ability to enforce rules and regulations among its members. Enforcement became more difficult when the resources were of potential value to powerful actors outside the communities. In such cases, government recognition was crucial though not necessarily sufficient. For instance, a central goal of the two Cambodian community forestry cases was to give the communities a better chance of enlisting government support for their claims to resources.

Transforming power relations and some inequality

Beck and Fajber's chapter (Chapter 15) examines the question of CBNRM and equity in depth, but I include some brief comments here because of this subject's importance in the critique of community in CBNRM. Critics have argued not only that CBNRM often fails to recognize inequality and difference within a community, but also that interventions can exacerbate these inequalities. In practice, any project that changes people's livelihoods and reorganizes their access to resources must also have an impact on inequality. How has community collective action changed wealth, gender and other forms of inequality? Do the projects described here proactively seek to improve livelihood security for the most disadvantaged? Do they ensure that projects do not increase inequality? It is one thing to identify inequality and to devise projects that address the needs of the poor as well as the wealthy. It is another thing altogether to address and change the actual causes of the inequity.

In the agrarian studies approach, rural inequality was often identified with inequality in access to productive resources, especially land. If land was a key cause of inequality, the solution was either land reform or collectivization. With its broader attention to all natural resources, CBNRM has the advantage of potentially offering a more contextually sensitive assessment of the sources of inequality. These include not only a disparity in access to land, but also differential access to a variety of other natural resources such as fish, water and forests. Because of the emphasis on participatory approaches to identifying differences in wealth and livelihoods, CBNRM research can also potentially account for inequalities not directly linked to natural resource access. These include a variety of ways to access labour markets, migration and education.

At the same time, it could be argued that with its emphasis on the community, CBNRM often diverts attention from efforts to reduce intra-community inequalities in access to resources, or postpones them as problems to be addressed after community control has been secured (Chapter 15). Even projects that identify differences within communities might nevertheless take these as given and simply make sure that the specific needs of the different wealth groups are on their agenda, without specifically addressing the causes, or attempting to reduce inequalities. As described more fully by Beck and Fajber in Chapter 15, most of the case studies in this volume did recognize inequality. Some devoted considerable effort to working with the poorer strata in rural communities, and a few – such as those in Bhutan and coastal Vietnam – went beyond this and attempted to reduce local inequities in resource access.

It is important to recognize that attempts to reduce intra-community inequalities through CBNRM do have some significant limits. Even if the rural poor derive a large part of their livelihoods from common property resources, inequality is multidimensional. These CBNRM case studies do not make any attempt to address inequalities that are not based on access to common property. Other sources of inequality could include differential access to private land, labour markets, education and migration opportunities. These inequalities are often shaped by gender and ethnicity. None of the projects described in this volume suggests redistributing private property like land, or cattle as in the case of Mongolia. The focus is on the implications of these private asset allocations for common property resource management.

CBNRM can have a significant impact on inequalities between rural people and other actors such as logging firms and powerful government resource agencies. In fact, a significant goal of CBNRM is to strengthen communities' claims to resources. Those cases which gave less attention to intra-community inequalities and differences were those that directed their primary attention to relationships between rural people and powerful actors, including state resource agencies, logging enterprises or agribusinesses. Examples include the case studies on community forestry in Cambodia (Chapters 3 and 11) and the one on policy incentives for community forestry in the Philippines (Chapter 13). The explicit reduction of diverse positions for the term 'community perspectives' in Chapter 13 can be understood as a simplification useful for effective external policy advocacy. However, readers must be cautioned that these sorts of simplifications should not be extended into arenas of local CBNRM practice. There is plenty of evidence that intra-community difference can be crucial for CBNRM practice even in the Philippines (McDermott, 2001; Mendoza et al., Chapter 12).

Becoming a voice for local villagers

If we base our understanding of collective action on what Meinzen-Dick and Di Gregorio (2004) call 'voluntary action taken by a group to achieve common interests', then one of the most important actions undertaken by the communities described in this volume was to speak for their own communities to other actors, with respect to interests held by some or all community members. Many of the techniques of PAR are to create a voice for communities, or more often, specific groups in rural communities. The cases are particularly self-conscious about community voice in relation to government. As this is addressed more thoroughly in Chapter 17, I will restrict myself to a few brief comments on the implications for understanding the concept of community.

In some ways it is through the question of voice that we might achieve the best understanding of community. This is because a group of people become a community when they are able to speak with a collective voice. It is around questions of voice that the importance of participation emerges as crucial, and it was the research that mobilized the active participation of community members that enabled communities to speak collectively.

There are three, perhaps four, cases in this volume where research projects were organized primarily around helping communities to establish a collective and effective voice with respect to government policy and practice. The project described by O'Hara (Chapter 13) found innovative ways of enabling community members to express their views to government officials, CBNRM practitioners and academics. Members emphasized that the broader policy context served to disable community forestry management. O'Hara's chapter shows that communities are not necessarily bound by locality, but that when they engage in broader networks they can contribute to changing the views and actions of other actors, including policy-makers.

The two community forestry papers on Cambodia also focus on the ways in which communities can convince governments to recognize collective rights to access, manage and use community forests. In other projects, while the collective voice to government was not the central activity, it appeared nonetheless. For example, Le Van An (Chapter 5) describes how residents in both of the upland communes in which the teams worked in Central Vietnam had lost access to significant livelihood resources. The research project prepared villagers to go to government officials to ask for expanded resource rights.

My observations so far, however, also point to a need for care in understanding community voice. The voices created through these studies are not the authentic voices of a pre-existing community, previously silent but encouraged to speak out through the project. Instead, like the communities themselves, voice is a product of interaction between communities and other actors. The most obvious interaction is that between research team members and community members, but more generally voice needs to be seen in the context of networks of actors. The ubiquitous resource management plans are an example: they are never simply the representation of pre-existing community strategies and intentions, but always need to be framed in terms that make sense to government officials. The case study set in Laos (Chapter 7) illustrates this well: Tubtim writes that she thought that prohibitions on fishing linked to the widow spirit represented a form of management; however, government officials did not consider this to be management. The new exclusive management regime was understandable to government officials, and emerged through interactions between the research project, government officials, residents in neighbouring villages and various groups in the community. However, the villagers were uncomfortable with the scheme until the widow spirit endorsed it.

Tubtim's chapter also points to an important qualification regarding the assumption of common interests in the articulation of community voice. She avoids ascribing voice to the entire community. Instead, she shows that the project had different meanings and outcomes for different members in the community, and indicates that not everyone was equally involved with respect to decisions and voice. In particular, as Tubtim notes, the village committee was in an ambiguous position because it represented the state to the village, but also the village to the state. Not all groups in communities may be present in the same way when a community speaks collectively. Inevitably, voice is also an outcome of power relations in communities.

As I noted above, the chapters (Chapters 3, 11 and 13) that are primarily focused on how communities can represent their collective interests to other actors pay the least attention to power relations and inequality in the community. Indeed, having an active community voice tends to put aside questions of power and inequality in communities. One way that power relationships in the community could be emphasized more in this kind of action research might be to systematically explore the networks traced by informal communication practices in a community, and see how these can coalesce into a community voice.

Becoming vehicles for other types of collective action

The projects described in this volume relate innovations in collective resource management as part of a CBNRM research programme. However, one of the striking features is how many different kinds of activities were undertaken in the name of CBNRM. Once the work to create collective action has been accomplished to achieve one purpose, those collective practices can become vehicles for achieving other goals. Although most of the activities reported in these chapters involved direct improvements to livelihoods (for example, crop improvement and credit schemes), the kinds of activities varied widely. One suspects there may also be related activities that were not mentioned because they seemed irrelevant to a volume on CBNRM. Ykhanbai and Bulgan's chapter on Mongolia (Chapter 6) provides some examples, in part because the opportunities for spin-off activities created more support for the research project. They point to how women supported co-management in part because it filled an unmet need to be involved in community social activities and services. 'During community meetings, people could meet each other and chat, get community help when someone was sick or needed money, or learn the best practices of herding, farming and livelihood improvements from each other.'

All this points to the ways that collective action and institutions can have multiple goals and unpredictable results, and can be sustained for reasons that do not fit entirely into a CBNRM framework but are nevertheless an important support for CBNRM. Although these case studies do not provide a lot of evidence for this, it seems that the motivation for the hard work necessary to sustain these communities was not just the promise of better resource management.

The opportunities such activities provided for socializing and organizing other unrelated activities were also crucial. Finally, there are hints in many chapters that CBNRM communities can easily transform into other types of collective action and communities: marketing collectives, service provisions, political actions demanding citizenships rights and so on.

Evaluation and new directions

Over the past couple of decades much has changed in development research and development policy. In academic work, questions of agency, actor-oriented development theory and actor-network theory have helped broaden ideas of who or what can be an actor in development. In development politics, revolutionary movements have been displaced by a plethora of social movements and NGOs, and class analysis has been displaced by community-based development work. This does not mean abandoning the notion of class-based agents. But it does mean contextualizing class-based agents in the broader realm of networks of actors including corporations, NGOs, communities, farmers, wage workers, consumers, academics, development agencies and others. In the case of communities, an approach concerned with collective action and networks of actors points to the recognition that local communities will never be idealized expressions of some kind of collective local action. Nevertheless, communities can be important actors in many fields of development practice.

One of the weak points of the broader field of CBNRM, including both advocates of community rights and mainstream development institutions, has been its concept of community. Practitioners, researchers and policy-makers often assumed that rural communities were stable and relatively homogeneous. Some proponents also argued that most rural communities have over time developed collective institutions for managing resources, and that these livelihood-based institutions were threatened by state intrusions and market integration. An examination of the case studies in this volume suggests a more nuanced approach, which understands community rather differently. With respect to the case studies in this book I have argued as follows.

Communities are best understood as being made through project activities. The projects remade existing communities or made new ones based partly on models for doing CBNRM that circulated in the IDRC and other networks involved with doing CBNRM. Such models were adapted according to local priorities as identified through participatory research.

The CBNRM communities did not just exist ready to be mobilized, but were the contingent and often tenuous outcome of considerable effort and hard work.

CBNRM activities were often welcomed and were appropriate to diverse kinds of situations, not just the idealized stable community with strong preexisting collective institutions for managing resources. CBNRM activities are particularly important among migrant groups in coastal and upland areas, as they are often among the poorest and have the least secure resource rights.

More broadly speaking, the communities described in this volume are the result of trans-local networks communicating with local networks (or communities). The term 'trans-local' allows us to talk about institutions or networks stretching outside the localities in which these projects are working, but without necessarily falling into the idea that they are external to these localities. In particular, the communities mentioned in this book often sought to strengthen their relationships with states and markets, and were created partly through these interactions, not outside them.

Not everyone participated in the communities in the same way. Differences were based partially on relative wealth and poverty, gender, ethnicity, status (government official or villager) and so on. This is a way of emphasizing that the case studies recognize that communities are internally composed of diverse networks, and that this diversity affects community participation.

The communities in these cases almost always became important collective actors in local and trans-local networks. Because the CBNRM communities participated in and thus influenced these networks, they could affect macro- as well as micro-development practices.

Rethinking community based on these studies of CBNRM in practice is useful not just for academic purposes, but also for policy and practice. I conclude with a few comments on the implications of this analysis for further research and practice.

If we understand the community in CBNRM as the product of articulation between local and trans-local networks, and as a collective actor in non-local networks, it becomes apparent that CBNRM is much more than simply a way of increasing community participation in NRM at the local level. The most optimistic scenario is that CBNRM can challenge the premise of 'trusteeship' that arguably underlies much development practice (Cowen and Shenton, 1996). In other words, communities can become rights-bearing agents in the practice of development, rather than simply trustees of agencies that act on their behalf. In this sense, CBNRM can make a substantial contribution to rethinking development policies more broadly. Nonetheless, much caution is warranted in relation to the widespread application of CBNRM, as there are many potential pitfalls that could undermine both local rights and anti-poverty work. In particular, a rigid or formulaic CBNRM could undermine existing informal community practice, undermine the livelihoods of the poor and empower local elites.

These are some of the implications for the broader academic analysis of development policies. What of the more practical insights for CBNRM research, practice and policy? I offer a few suggestions here, suggesting key arguments from my analysis of the cases.

When practitioners and researchers abandon unrealistic images of stable rural communities, they pay more attention to people who have been mobile, displaced and who often have the weakest property rights. A specific implication regards collective rights to hold and manage resources that may be created for people who can frame their identities as indigenous (not an obvious category in most parts of Asia). Where appropriate, these sorts of rights might also be extended to peoples who cannot frame themselves as indigenous, including people who have been mobile or who are described as migrants.

Practitioners need to pay careful attention to the way that creating CBNRM communities also creates territorial and social exclusion. Who is being excluded? How is it justified? For example, do programmes allocate resource rights to indigenous peoples at the expense of displaced or more mobile communities?

CBNRM research and practice need to come to terms with the way in which rural communities are integrated into markets. Research might explore ways that more of the benefits of market integration can be captured by the poor, as well as find ways of protecting collective rights and resource quality in the face of marketization.

Policy-makers need to recognize diversity and multiple forms of community in administrative villages or local government units. Enabling legislation for CBNRM must accommodate flexible definitions of community, to allow for adaptation to local circumstances.

Practitioners might want to acknowledge the limits on community-based collective action as a strategy to reduce inequality, although CBNRM is well placed to target and work with the poorest groups in rural areas. These kinds of efforts can be important for reducing poverty, but are not well suited to reducing inequality based on differential access to private property, labour markets, education and so on. Moreover, CBNRM is not the only kind of collective action that rural people can engage in, and other kinds of collective mobilizations (from social movements to electoral politics) might be more appropriate for addressing these kinds of inequalities. However, CBNRM is well suited to protecting important resource rights for entire communities that are threatened by powerful external actors such as state resource agencies or private enterprise.

CBNRM practice needs to develop more systematic methods for understanding and working with power in rural areas. For example, tracing informal communication networks might reveal power relations in the same way that wealth-ranking exercises trace economic inequities. This becomes important especially when we think of CBNRM as a means for creating collective actors who can speak for rural people. In turn, this suggests that we need to understand how the resultant collective voice is the product of power relations both locally and trans-locally.

Understanding that CBNRM communities are the product of hard work, and not naturally existing before a project's interventions, directs our attention more to the demands made on rural people to contribute their effort to this objective. The nature of these demands can filter the choice of participants. Women and the poor may not find it as easy to participate, which can lead to elite domination by simple selection.

If CBNRM communities are the unstable products of ongoing effort and work, then sustainability becomes an important issue. How can the effort made to create these communities be sustained through institutionalization? How can institutionalization happen in a way that maintains the flexibility necessary for institutions to continue to address new problems while they emerge?

Document(s) 20 of 24