Document(s) 18 of 24

Why should research be pro-poor?

Despite rapid economic growth, the magnitude of poverty in Asia remains daunting. In 1998, the population of the East Asia and Pacific region living on less than US$1 per day was 278 million or 15 per cent of the total. If we exclude China, this number dropped to 65 million or 11 per cent. For South Asia, the total was 522 million or 40 per cent of the total population. If we increase the threshold to $2 per day, the numbers become 892 million or 49 per cent of the total population in East Asia and the Pacific. And in South Asia, the total was 1,096 million or 84 per cent of the population (Smith and Jalal, 2000: 66). These figures pose an enormous challenge to the goal of poverty reduction. It is not surprising that the donor community and civil society clamour for more attention to poverty reduction.

The Millennium Development Goals mandate an eradication of extreme poverty and hunger. Goal No. 1 calls for halving the proportion of people who suffer from hunger between 1990 and 2015. Other goals endorse the need to promote gender equality and to empower women, and to integrate the principles of sustainable development into policies and programmes to reverse the loss of environmental resources (World Bank, 2001: 5). The emphasis of the Millennium Development Goals on poverty provides strong rationale for action which is informed by new knowledge from research. In fact, strategies and programmes for poverty reduction are better designed and implemented when based on pro-poor research. Consequently, there are three strategic roles for pro-poor research:

1. it ensures that the issues and concerns of the poor themselves are understood and given attention;
   
   
2. it promotes conditions in which the poor can use their own skills and talents to work their way out of poverty (IFAD, 2001); and
   
   
3. it empowers the poor to become agents of their own well-being (Moore, 1999).

An important first step for pro-poor research to meet the challenge of poverty reduction is to understand the constituents of poverty. Poverty must be dealt with not only in terms of changes in food supply and availability but also in terms of the complex social and economic factors that underpin it. Instead of proceeding from the conventional idea that poverty is about too little income, insufficient consumption or inadequate nutrition, it is necessary to accommodate more comprehensive notions of poverty such as vulnerability, insecurity and powerlessness. In addition, objective measures must be complemented with subjective ones that depart from the poor's descriptions of what poverty means to them.

From the poor's viewpoint, to be not poor means material well-being based on private consumption; access to common property resources; state-provided commodities and assets; physical well-being; a feeling of security and the ability to cope with emergencies; freedom of choice and action; autonomy and dignity. Corruption, conflict and violence, powerlessness and insecure livelihoods due to fewer economic opportunities are the poor's biggest problems. For the poor, security, dignity and autonomy are as important as income (Smith and Jalal, 2000: 67).

If poverty is multidimensional, it warrants no less than the comprehensive strategy proposed by the World Bank (2001: 6–7) to attack it. The strategy suggests three ways: promoting opportunity, facilitating empowerment and enhancing security. None is more important than the other because all three elements are complementary. Interventions as well as research that are pro-poor must not be judged only by their attention to income, but also by their focus on assets, rights and institutions, as these provide a structure for opportunities for the poor and reduce their vulnerability.

Over the last 10 years a consensus has developed that there is a need to move away from income or consumption alone as measures of poverty. Alongside these must be put broader measures that address qualitative measures such as rights, vulnerability, security and autonomy. In this regard, research into the qualitative elements of poverty by the World Bank's 'Voices of the Poor' study for the World Development Report 2000/1 leads the way. It is encouraging to see that many agencies are now more open to the qualitative assessment of poverty. Complementarities in qualitative and quantitative methods of poverty appraisal are being recognized and analysts and policy-makers are looking for a way forward in using the two approaches to effectively design poverty reduction strategies (Kanbur, 2001).

The failure of agricultural research establishments to sufficiently address issues that concern the poor are now well recognized. A significant part of past research has concentrated on high-input production systems in irrigated areas which have contributed to significant adverse environmental effects that further exacerbate the burden on the poor. But mountains, dry lands, forests and coastal areas have been largely ignored by agricultural research until only very recently, in spite of the fact that most residents of these areas are poor. Consequently, international agricultural research establishments have now become more focused on responding to the need for a pro-poor orientation.

Such organizations recognize the special role for research on NRM to complement their traditional emphasis on food crops, commodities and sectors. This change in mandates has propelled research with a concern for the poor and provided researchers with crucial support. The real challenge for agricultural research organizations is how to translate the poverty emphasis into research processes and products. Strategies for enquiry and dissemination must change. For example, centralized research and development and extension that worked during the green revolution will not apply, because unlike research on commodities, research in resource management is extremely location-specific. While the green revolution was input-intensive, sustainable agriculture is highly knowledge-intensive. The extension system should be devolved away from the centre as much as possible, with the local governments in charge of the public system and the private sector (particularly NGOs) encouraged to participate (Smith and Jalal, 2000: 51).

However, what factors determine whether research in NRM is truly pro-poor? This is the question we will attempt to answer below.

How can research be pro-poor?

We first asked what it means to conduct NRM research that is pro-poor. One of the other contributors to this volume suggested to us that in order to be considered pro-poor, research would need to have had a 'demonstrable effect in bringing about empowerment of poor people, access to their rights and sustainable poverty reduction'.¹

The authors do not disagree with the idea that the impact of research is an appropriate indicator of its being pro-poor. However, we propose that projects also be assessed in terms of their approach to the conduct of research, an activity that systematically produces and applies knowledge. We depart here from the concept of knowledge as a product or thing which exists outside us, which we possess and which is stored in finished form. We believe that the production of knowledge is a social activity relying upon linguistic, conceptual, cultural and material resources and that 'science is not a thing but a social activity' (Sayer, 1984: 19–20). Precisely because the poor are those who are excluded from social processes – falling through the cracks so to speak – we argue that research processes that ensure the meaningful participation of the poor are as important as research outcomes and impacts that can benefit them.

Research that intends to be pro-poor adopts a pragmatic research framework and employs participatory strategies and methods of enquiry in order to generate transformative knowledge that can then lead to social change. Armed with this research framework and imbued with its values and attitudes, all of this volume's CBNRM projects highlighted and applied these principles to the specific concerns of the communities with whom they worked. There was a deliberate choice about a problem-focused and people-centred research framework. Together with a participatory and interdisciplinary perspective, the project teams made a commitment to a collaborative undertaking with local research groups in government agencies, universities and NGO offices. In the following discussion, we focus on how the research framework,² – the study's approach to its subject matter, strategies and methods of enquiry – was implemented in order to consider the poor and their concerns.

Issues of NRM are a subject of vital importance and interest to the poor. Poor communities are particularly vulnerable to failed environmental governance, because they heavily rely on natural resources for subsistence and income.³ In terms of site selection for their research, case study teams specifically targeted poor and marginalized communities. We can illustrate this deliberate emphasis in the following examples, although it applies to all the cases:

In the mid-1990s, 70 per cent of the population in Ratanakiri was indigenous. But ethnic Khmer who are Buddhist and mainly paddy (wet rice) farmers are the dominant group in Cambodia, and 98 per cent of the government staff in Ratanakiri belong to this group. The Khmer culture regards shifting cultivation and the culture of indigenous communities with suspicion. (See Chapter 3.)

Hong Ha and Huong Nguyen are two of the 16 poorest communes in the A Luoi district, and among the approximately 1,200 designated poorest communes in the country, according to national poverty criteria. (See Chapter 5.)

Guizhou province is situated in the eastern part of the Yunnan-Guizhou plateau in southwest China, with mountains and hills accounting for about 90 per cent of the land, and minority people taking up 40 per cent of the population. It is one of the poorest provinces in China. The major socioeconomic indicators such as income per head, grain production and areas of arable land, are all among the lowest in China. Of the 34 million people in the province, 30 per cent are living below the poverty line and constituting over 10 per cent of the poor in China. (Deshou et al., 1997: 85; see also Chapter 9.)

The promotion of a community-based approach to NRM in these cases not only responded to the degradation of forest, water and land resources. At the same time it was an attack on poverty, because degradation affects livelihoods and destabilizes the natural resource base upon which the poor rely. Investments in NRM, aquaculture and livestock, agroforestry and other labour-intensive agricultural activities may be expected to have positive impacts on productivity and poverty (Hazell, Jagger and Knox, 2000). But payoffs in both production increases as well as poverty reduction result only if there is emphasis on the needs and productive possibilities of smallholders and the landless (Smith and Jalal, 2000: 51). These authors also noted that CBNRM finds allies in the poor because they 'have strong incentives to preserve natural capital as their only significant source of capital' (p. 55). Just as the Mongolia team observed, there was a keener interest from the poor herders in adopting CBNRM practices because these enabled them 'to improve their livelihoods, to secure pasture, to participate in decision making, and to reduce the costs of herding animals through economies of scale' (see Chapter 6).

Pragmatic framework

Poverty and the complexity of agricultural systems pose special research challenges. Less favoured areas where many of the poor live are characterized by a degree of heterogeneity that partly explains the failure of universal approaches to doing research. Cultures are astoundingly diverse. As well, social relations are a product of historical patterns and while they may share some characteristics from place to place, they are unique to specific locations. Similarly, nature represents a set of diverse ecosystems that have evolved and have been affected in many different ways from one location to the next (Smith and Jalal, 2000: 73). Therefore, pro-poor NRM research must successfully grapple with the high variability of ecological, technological, economic, political and cultural facets of the systems in which the poor live. To achieve this, the research must have a pragmatic framework⁴ where knowledge claims arise from situation-based action and consequences. It must emphasize applications that work and solutions to problems. The most important factor is to identify the problem using pluralistic approaches (Creswell, 2003: 12).

Pro-poor NRM research is directed at the enrichment of people's practical knowledge, which constitutes the bulk of our social life and accounts for the everydayness of life. Such practical knowledge requires the notion of a knowledgeable human agent (including the poor). As Beck and Nesmith (2001: 120) argue, the prevailing bias against poor people's knowledge and agency in development studies and practice needs to be removed. The study of people doing the most ordinary things must rest on the presumption that most of what people do, they do intentionally. In addition, they probably know the reasons for their behaviour (Mendoza, 1999). Hence it will come as no surprise that participatory methods ought to be adopted.

The CBNRM cases presented in this volume faced the challenge of variability and complexity of local ecosystems because they concentrated on site-specific work. This enabled research teams to capture the wisdom of local knowledge of forests, water, fisheries and coasts. They faced the issues of resource and land tenure, including how local institutions framed people's NRM systems. This meant that even though research on the sustainable use of natural resources began with a strong focus on issues of the environment, eventually equal emphasis in the research analysis was placed on property rights and local institutional issues. In particular, there was an unwavering attention to security of resource access. The Ratanakiri project team (Chapter 3) facilitated the endorsement by the provincial governor of community maps and NRM plans for Somthom commune, which supported the villagers' customary land and forest rights. Rural transformation hinges upon the primary importance of pragmatic access to resources for the powerless and the recognition that access is the poor's most important resource.

Participatory strategies of enquiry

Participatory methods criticize conventional social science enquiry as reducing the subjects of research to objects of research. Practitioners of these methods argue that the subjects who will be affected by research should also be responsible for its design. Participatory methods uphold a worldview that sees human beings as co-creating their reality through participation: experience, imagination and intuition, reflection and action. The knowledge and experience of people, including those who are often marginalized or oppressed, is directly honoured and valued. PAR values the processes of genuine collaboration (Reason, 1998).

PAR goes by many names: action research (AR), practitioner research, participative enquiry, participatory learning and action (PLA), participatory research and action. It is widely recognized as a powerful way of facilitating changes in complex situations (Laws, 2003). The core process enables participants to share their perception of a problem, to find common ground and then to engage a variety of people in identifying and testing out some possible solution. This is a process of shared learning for everyone concerned.

What can the poor do through participatory research? They can set the research agenda instead of simply carrying out an agenda already designed for them. They can assist in the conduct of trials and experiments. However, they can do more than this. In the Guizhou Academy of Agricultural Sciences (see Chapter 9), the research team started to work with farmers. Together, they tested technological options on farmers' fields for fruit trees, mushrooms, and intercropping of maize and wheat, as well as rapeseed and maize. With exchange visits among farms, this project enabled farmers to share their experiences with one another. In another of the project sites, the village operates and maintains a drinking-water system through a set of rules the villagers drew up themselves. Community members also constructed the water system with the cooperation of male and female farmers who designed it and mobilized labour, materials and funds. This accomplishment is better appreciated if it is also noted that government technicians had previously told the villagers that the site did not have any adequate water source. It is worth emphasizing that in order to ensure that research results benefit the poor, researchers must ensure that the poor participate in the research processes.

In the Hue project, farmers in the rice production group developed their own criteria to monitor and evaluate results. They made decisions on which crop varieties to use, the amount of fertilizer to apply and which cultivation techniques are best (see Chapter 5). Through a special skills-building activity initiated in 1999, PM&E was introduced into the Chinese CBNRM projects, providing farmers and other learners with a powerful, systematic tool for directing further interventions. This experience is documented in Vernooy, Sun Qiu and Xu Jianchu (2003) and served as input towards the adoption of the methodology by other projects (Gonsalves and Mendoza, 2003).

During the PAR process, researchers typically worked with disadvantaged groups, serving as partners and facilitators, so that people had the opportunity to act effectively in their own interests. They were enabled to define their problems, define potential remedies and then take the lead in designing the research that was to help them realize their aims (Babbie, 2004: 296). The China team who worked with the villagers of Chaoshan successfully modified their agenda in order to adapt to the capacities of the poor. The team facilitated village participation in a government-supported biogas scheme, by negotiating more flexible qualification criteria and adapting the design to local conditions (see Chapter 9). In another case, the Hue project team adjusted their communications strategy in recognition of the difficulties of engaging the poorest members of the community and women. They helped form problem-oriented farmer groups and separate women's groups which could focus on innovations to suit their own needs. As a consequence, the research team was able to reach the very poor, including more women (see Chapter 5).

Even with participatory research methods, dominant members of the community still tend to monopolize the process. Therefore, researchers must devise innovative approaches for reaching and working with the poorest people if their specific needs are to be addressed. The Mongolia team used a wealth-ranking exercise to identify where the poorer herders were located among the project sites (see Chapter 6). Similarly, the Hue research team (see Chapter 5) also used a wealth-ranking analysis to successfully identify groups of poor farmers and, subsequently, to identify their concerns together. In the lagoon project (see Chapter 4), the team worked separately with each group of competing interests. They ensured that the mobile-gear fishers were able to raise their concerns about exclusion from the lagoon by the expansion of net-enclosures. The joint committee that negotiated and monitored community management of the lagoon consisted of representatives from different fisher groups, government offices and researchers.

Participatory methods enable flexible manoeuvring by local research teams while they explore, define and confront the challenges of context-specific environmental problems. An example is the experience of the PMMR team in Cambodia (see Chapter 8). The project team worked with a community that could be judged as culpable in the degradation of mangroves because many villagers deliberately migrated to the area in order to produce charcoal. Dissuaded by a significant government crackdown in 1999, the villagers switched to fishing. Through project activities the formation of village management committees was facilitated. When the resource management plan was drafted by these bodies, it included the community's affirmation of their own responsibility for stopping charcoal production and dynamite fishing.

NRM issues often involve multiple stakeholders, which include external groups, government agencies and different interest groups in the community. It can be difficult for researchers to nurture participation by the less influential groups such as the poor. Therefore, researchers must establish partnerships with local organizations and support groups (such as NGOs or farmers' networks), which are specifically linked with the poor, so that their interests and concerns can be effectively heard. This multiple-stakeholder approach enables dialogue among different groups within a community, between communities and between communities and external agencies. All the projects in this volume, in different degrees, faithfully carry out this crucial spirit of inclusion, dialogue and consensus-building. On the whole, teams indicate that this approach has many advantages. However, it is important that researchers periodically assess whether the poor's perspectives are being heard in such multi-stakeholder settings.

The adoption of participatory research methods by the project teams also enabled the introduction of a participatory mind-set among researchers and the personnel of partner agencies of the projects. It provided models of new practice which expanded the possibilities for both practitioners and communities. Nonetheless, engaging in PAR is insufficient to ensure that the poor benefit from CBNRM activities, because communities are differentiated and stratified. It is often easy for practitioners of PAR methods to forget that not all members of a group have equal access to the resources, rights and benefits of community life. Social and gender analysis provides occasions for critical reflection on PAR methods. Such analyses alert researchers to the fact that the poor and marginalized members of a group may not be involved and benefited (see Chapter 15).

Many CBNRM site-specific projects evolve over periods of at least three years if not longer, as in the China, Cambodia and Vietnam cases. This is to be expected because problem-solving for NRM issues in collaboration with poor communities requires time for participation, learning and action. Some may view this process as ineffective and costly. In addition to the arguments above, it is useful to emphasize the value of PAR in nurturing social capital in communities. The effectiveness of development efforts, particularly in rural communities, is closely linked to social capital that draws upon the mutual trust and understanding built and shared among individuals and households. This enables cooperation, reduces transaction costs and makes it possible to optimize solutions to many problems. Because of rapid change brought about by migration, urbanization and modern means of communication and transportation, social capital has declined in many communities. This is generally true, notwithstanding the rhetorical invocations of solidarity expressed by local people when questioned about the social relations in their communities. This weakening of social capital has important implications for the implementation of development strategies: efforts to decentralize resource management by returning control over assets to local communities are but one example. Indeed, decentralization may not be effective if social capital has dissipated (IFAD, 2004). In this light, PAR projects add value. They may contribute to a reawakening of a people's sense of community as they once again talk to each other, work together and act together to solve problems that beset them (see Chapter 16 for a more detailed discussion of community action and social capital).

Transformative knowledge

Undertaking research is not only a means of knowledge production. It is also a tool for the education and development of consciousness as well as mobilization for action. Vernooy and McDougall (2003: 116) define transformative learning as an approach whereby learners build a more integrated and inclusive perspective of the world together. Through the learning process, they jointly transform some part of their worldview. For advocates of PAR, access to information signifies access to power. It has been said that power over knowledge is held by members of the dominant class, sex, ethnicity or nation. When the poor and disadvantaged see themselves as researchers, as learners, they regain power over knowledge (Babbie, 2004).

The ability of poor households to utilize natural resources and enhance their livelihood strategies is influenced by scientific discoveries and technologies. Therefore, it is imperative that these are managed and directed with them in ongoing partnerships with research and scientific organizations. Otherwise, the benefits from innovative technologies leak to other groups and are not captured by poor and peripheral communities. The lessons from the green revolution of the early 1970s for increasing agricultural productivity are not to be forgotten. Not only was higher grain production not tackling poverty as quickly as presumed; the strategy was also not addressing the production needs of a vast number of the poorest farmers. On the whole, the research system was unable to address the needs of client groups (Hall et al., 2002). Therefore, the imperative is that agricultural researchers must work with communities on their problems. As Hall et al. propose, agricultural science must operate in a developmental framework rather than a scientific one if it is to make a contribution to poverty reduction.

These cases put a premium on learning by doing and accepted that the learning process in PAR was time-consuming and largely iterative (see Chapter 2 for an explanation, and Chapters 4, 6, 10, 11 and 13 for illustrations of this approach). They understood that such knowledge requires continuous learning and can lead to adaptive management. The CBNRM projects are a critical input into a broader advocacy effort towards the effective implementation of sustainable development projects by national, regional and local bureaucracies. These agencies need to learn to work with dedicated researchers and develop partnerships at the local, regional, national and international levels. CBNRM contributes to the overhauling of the perspectives of bureaucracies with reference to their role in working for the benefit of the poor. For example, in Chapter 10, the Bhutan team states:

As the two case examples in water management and forestry illustrate, the RNRRC [Renewable Natural Resource Research Centre] reoriented its research agenda to reflect community priorities, rather than the interests of the researchers ... The project's integrated approach also altered the research planning process at the centre. Staff from all sectors and sub-sectors (crops, livestock, forest, Integrated Pest Management (IPM), socio-economics and water) now discuss their plans together and explore opportunities for synergy. Hence, a more integrated planning and implementation of research occurs. More emphasis is placed on participatory technology development, participatory plant breeding and variety selection, and the need to build on farmers' knowledge and practices.

Other than enabling the poor to capture benefits from the results of the research, the empowerment of the poor also comes from their ability to exercise choices. Establishing linkages with policy-makers, policy formulation processes and networking with academics are main components of an empowering process resulting from pro-poor research. Once they find their voice, the poor enter into dialogues with the powerful. By organizing themselves in collective action, they can gain more control over their resources so that they continue to provide the foundation of their livelihoods and contribute to the formation of private and social assets. This empowerment of the poor to take active part in the decisions that affect them is what a participatory approach catalyses.

It is not unexpected that these projects provided a most appropriate policy complement to the decentralization initiatives that are rapidly taking place in Asia (see Chapter 17). The strategy to focus research effort on the transitional economies of Asia has provided an excellent opportunity to contribute to the democratization efforts in these countries. The research projects provided governments with occasion to learn how to be more responsive and accountable to their own people. CBNRM not only changes the power relations in local communities, it also transforms such relations between local groups and external agencies. For example, villagers of Somthom commune in Ratanakiri prohibited the military from entering their forests and even confiscated their chain saws to prevent their illegal logging (see Chapter 3). In Koh Sralao, the village management committee stopped a boat carrying mangrove logs cut without their permission. Although the boat-owner was related to the provincial police commander, the village management committee was able to negotiate with the former to pay a fine and sign a written commitment to stop his illegal activities in the area (see Chapter 8).

Transforming the lives of the poor requires that they be targeted. Pro-poor research must purposively identify and locate poor households in a community, as was done in the Hue project (Chapter 5) and the Mongolia project (Chapter 6). Local conditions contribute to poverty, yet they can also contribute to solutions. Long-lasting and workable strategies for poverty reduction are often those which are constructed by the poor themselves and managed by them rather than externally introduced. That being said, measures for poverty eradication cannot be left entirely to the poor, because their condition has excluded them from the normal operations of social and economic processes. They require the assistance and support of government agencies from the village, commune and district levels, as in Cambodia's community forestry project (see Chapter 11), or from commune to provincial governments as in the Ratanakiri project (see Chapter 3). Through networks, the poor may be provided with support by civil society groups at the local, national and international levels, as illustrated by the use of networking in Koh Kong, Cambodia (Chapter 8).

CBNRM is not just about technical improvements to resource productivity but also about governance and livelihoods. Natural resource governance and community-based approaches in particular, are all about process. And as a process, CBNRM changes power relations by strengthening capacities at the level of local communities. Choosing the participatory approach as its core strategy has concretized what a pro-poor policy or project requires. To choose to be participatory is not simply choosing one research method over another. It is declaring that this is the way to do development research, that is, research that is truly people-driven and democratic. The sciences, natural as well as social, must be at the service of communities and peoples. Development research is an enabling process for communities to find their way, their solutions, to live their lives, to make their own mistakes. But always it ought to be their project, with researchers acting as facilitators, midwives to the birth of empowering processes.

Strengthening pro-poor CBNRM research: methodological issues

We have shown above how the CBNRM research teams executed the PAR framework to promote opportunity, facilitate empowerment and enhance the security of the poor communities with whom they worked. Now we turn to a discussion of selected methodological issues that can enhance the ability of CBNRM research to be pro-poor. There is no hierarchy of importance implied in the order of presentation.

Utilizing local knowledge through site-specific studies

Research on marginal agricultural lands has been criticized for omitting important gaps (Jodha, 1991). Research work is not sufficiently oriented to local situations, it does not consider indigenous systems as sources of learning and it ignores locally specific research results. The CBNRM research teams have responded to these inadequacies by stressing the strategic importance of site-specific and context-driven research in NRM. The projects cover a diversity of sites from community forests, mangroves, rangelands, to freshwater and coastal fisheries, to upland watersheds. They engage farmers, fishers, livestock raisers and forest dwellers. They contribute to a wide range of learning opportunities grounded in local problems and solutions.

There are other uses of site-based action research. First, it demonstrates that alternatives exist to top-down or blueprint approaches to NRM. For example, the Ratanakiri project developed participatory land-use planning techniques which are now being scaled up across Cambodia with SIDA and UNDP support, as discussed in Chapter 3. Second, site-based findings provide a more credible basis for local communities to influence national policy-makers. The action research sites serve as opportunities for meaningful dialogue by villages with local officials towards establishing rules of NRM governance (see Cambodia community forestry, Chapter 11, or the Guizhou case, Chapter 9). Third, the local communities in the villages in Tam Giang lagoon and the uplands of Hue in Vietnam (Chapters 4 and 5, respectively), or in Koh Kong and Kampot, Cambodia (Chapters 8 and 11) have demonstrated the capacity of communities to manage natural resources effectively. More importantly, these cases also exemplify how communities can contribute to successful efforts at reversing environmental degradation (Gonsalves and Mendoza, 2003).

In addition, site-based projects act as focal points for multilevel networking, as illustrated by the PMMR team (see Chapter 8). They also serve as nodes that enable a range of actors to address changing sociopolitical contexts, as demonstrated by the Mongolia team (see Chapter 6), and as ground from which springs community-derived experiences influencing policy review and reformulation. Projects in Bhutan (Chapter 10), Cambodia (Chapters 8 and 11) and the Cordillera (Chapter 12) illustrate this. To ensure that the concerns of poor communities are neither neglected nor bypassed, pro-poor research in NRM must continue to be oriented to long-term, site-specific studies.

Addressing complex interactions in physical landscapes

Pro poor NRM research must address issues arising from the complex interaction between and across ecosystems. Within watersheds there are strong linkages between forests, uplands and the lowland areas. When watersheds degrade, crop and livestock production are at risk. Degradation at the landscape and ecosystem level is often linked with farm-level degradation. Furthermore, FAO data suggest that the biggest cause of deforestation is agriculture. We also know from the International Union for the Conservation of Nature (IUCN, now known as the World Conservation Union) that most protected areas in the world have people farming in them. Agriculture can also pollute collectively managed bodies of water such as lakes or lagoons. But farmers can also recapture some of these agricultural nutrient flows. In communities around the Tam Giang lagoon in Vietnam (Chapter 4), for instance, the fertilization of otherwise sandy soils is augmented by sea grass (macrophyte) found in the lagoon. There are also linkages between forests and agricultural lands. In Cambodia, forests are maintained by local communities partly because they believe it has improved yields in adjacent rice paddies (see Chapter 11).

In the Lingmutey Chu watershed, if food production has to be improved in the lower part of the watershed, the upper communities need to be assured that there is enough water early in the season which can be released to lowland communities without adversely affecting production in the upper parts. Such interconnectedness can only be handled by an integrated approach to resource management, one that includes crops, livestock, forests and water. Thus, the Bhutan team (Chapter 10) departed from their traditional research focus on specific agricultural commodities and farming systems so that they could adopt an integrated NRM approach to adequately handle the linkages among forests, water management and agriculture. Several of the CBNRM teams, for instance in Hue (see Chapter 5) and GAAS in Guizhou, China, (see Chapter 9), have had a strong historical engagement with traditional agricultural research, but have now shifted their research approach. Like the Bhutan team, they have adopted the more holistic and integrative approach of CBNRM.

Sustaining livelihoods

The poor rely on a variety of livelihoods which depend in turn on a wide range of natural resources. As the sustainable livelihoods framework proposes, a form of livelihood analysis is needed so as to assess the causes of poverty. During this analysis process, attention must be addressed to the poor's access to resources (be these financial, natural, human or social), as well as to livelihood opportunities. Relationships between and interactions with relevant factors at the micro, intermediate and macro levels need to be understood.. This form of livelihood analysis is used by an increasing number of organizations, including UNDP, the International Fund for Agricultural Development (IFAD) and the Department for International Development (DFID) of the UK.

The framework is primarily a tool to ensure that the diversity of economic activities in which poor people engage is fully understood. The process starts by analysing people's livelihoods, including their dynamic nature (how they change) as well as the multiple strategies they use to secure their livelihoods (Adato and Meinzen-Dick, 2002). Livelihoods provide the foci that enable an assessment of development interventions and can guide prioritization decisions. A key feature of sustainable livelihoods framework is that it recognizes the people themselves as actors with assets and capabilities who are capable of rational action in pursuit of their own livelihood goals. As discussed above, this resembles the premise of the research framework of the CBNRM teams. Now that the goals of agricultural research are broadening to include resource management and poverty reduction, such a framework – complex though it is – is an important tool for researchers in order to help improve the relevance of agriculture and NRM research for poor people, especially those living in less favoured areas.

The poor rely more on common resources than the non-poor (Beck and Nesmith, 2001). Higher productivity gains lead to commercialization, which may provide incentives to privatize common property resources and exclude the poor. Thus, pro-poor NRM research must concern itself with the changes in the allocation of resource rights over common resources. Researchers must pay attention to the processes that establish resource management structures in order to ensure that their design does not neglect the interests of the poor. Just as CBNRM research directs its attention towards the sustainable use of collectively managed resources, the same must also be done for privately managed ones. It is appropriate for the initial emphasis on resource management to shift gradually to livelihoods after new assets are built from the rehabilitation of highly degraded resources.

Engendering NRM research

Integral to the concern of pro-poor research is the concern for gender equality and the empowerment of women. Judging from some of the cases in this volume, capturing gender issues in NRM requires more attention. Procedures often equated with engendering research, such as enabling one or more women to attend meetings, participate in projects, or become part of decision-making bodies can address representation but do not by themselves solve inequities in NRM.

One must ask, what about a woman's right to own land, to inherit land, to sell and trade the output of her field, to negotiate her labour input in her plots as opposed to other family plots, to benefit from animals that she raises, or to choose which animals to raise? Are inequities present in these areas?

If CBNRM projects are to strengthen their pro-poor orientation and enhance their ability to deal with the internal inequity of poor communities, they need a better focus on such issues. Enhancing the opportunity of poor communities rather than external agencies or non-poor communities is an excellent starting point for pro-poor research. However, it cannot be its destination.

Of course, differences in the sociocultural context among countries need to be considered. Gender issues will vary among East Asian communities like China, South Asian communities like Bhutan and Southeast Asian communities like Cambodia and Vietnam. CBNRM teams need to ensure that their projects consider and respond to gender issues that surface in NRM practices. Even though the China and Hue teams began with some success to confront gender issues, more is required.

Engendering research concerns itself with changing perspectives. Gender relations are socially constructed, as are gender inequities. Like any social phenomenon, they are intrinsically meaningful and concept dependent (Sayer, 1984:31). The researcher and the 'researched' hold ideas, beliefs and concepts about gender relations and what constitutes gender inequities. Unlike non-social objects, social phenomena hold meanings which are intrinsic and are not merely externally applied descriptions. In studying gender, status, politics or ethnicity, the social scientist must interpret what these actually mean to those who are affected or involved. This is not the case when studying atoms, cells, rock formations or black holes. Therefore, the distinction between social science, the humanities and everyday social knowledge on one hand, and (natural) scientific knowledge on the other, is fundamental. Ideas and meanings are not only in society but also about society. Therefore, researchers may adapt gender-sensitive procedures but may still fail to account for gender relations and, in fact, remain insensitive to gender inequities.

Two possible research 'blinders' can exist. One occurs when researchers from a different cultural background become conscious of gender inequities and seek to impose these concepts of gender relations on another community. What they see are inequities they have become sensitized to from their own social and cultural world; ones which may be completely misplaced for the new situation. The other blinder occurs with someone born and raised in the same community, but who is unable to escape his/her own gender ideology. This happens to many researchers in CBNRM projects. Engendering research will have to confront these blinders and enable researchers to change their perspectives. In this regard, more capacity-building initiatives in gender and social analysis for the research teams are welcome.

Embedding social scientific analysis in multidisciplinary NRM research

Project interventions often result in inequalities and unanticipated negative consequences. There is a growing awareness that the manner in which projects can reduce these inequalities and adverse impacts depends upon the integration of strong social criteria in the project design. This requires the contribution of the social sciences in research design and implementation andnot merely in evaluation. NRM research today not only must deal with issues of degradation and management of natural resources. It also has to take account of livelihoods, asset-building and gender inequities. For example, gender equity means giving women access and control over fundamental assets. This is increasingly perceived both as an objective of, and as an effective instrument for, poverty reduction. These considerations justify the strategic if not central role of social scientists working together with natural scientists in agricultural research teams of the future. The research teams profiled in these cases are already multidisciplinary teams. However, some of the cases point to ways in which the social science capabilities of the research teams continue to need strengthening, despite the efforts already made in this regard (also Gonsalves and Mendoza, 2003).

In research into biophysical conditions related to social processes, current research has done good work on farming systems. Among the identified inadequacies are the disregard for indigenous systems as sources of learning and the disregard of traditional systems and their adaptation to limitations as well as opportunities in marginal lands. Jodha (1991) traces these problems to the subsidiary role of social sciences in agricultural research. This leads researchers to give inadequate attention to social relations and rural power structures. Therefore, appropriate methods must be derived and developed to foster opportunities for multidisciplinary work in NRM, first among natural scientists and then with social scientists. Specifically, social scientists must be brought into the beginning phases of NRM projects so that they can work with natural scientists in research design, technology development and the evaluation process. Logistical implications and transaction costs of such a strategy should be viewed as essential and strategic investments that need to be made. Invariably, a menu of options should be developed for each agro-ecological niche.

Social scientific inputs are required not only at the level of projects but also at the level of organizations. Analytical skills in the social sciences must be introduced, strengthened or rebuilt. Organizational environments need to be re-crafted to nurture a change in attitudes and behaviour in the agricultural research community. Without a substantially enhanced social scientific perspective in research institutions, the goal of reaching larger numbers of poor could well be missed. It is fortunate for the CBNRM projects in this book that there was an early recognition of the necessity of balancing social and natural sciences as inputs in a research framework. Through this process, an interdisciplinary perspective is brought to bear on the problems at the community and landscape levels. Many of the CBNRM teams from Bhutan (Chapter 10), Cambodia (Chapter 8) and Vietnam (Chapters 4 and 5) have already internalized substantial sensitivity and analytical capacity in social sciences. However, more work must be done in this respect as researchers with natural scientific training learn and master social scientific skills, and as both natural and social scientists develop more integrative ways of working together. These challenges are difficult, even for experienced researchers.

Highlighting the need for conflict resolution and facilitation skills

Whether in watersheds, forests or mangrove coastal areas, CBNRM projects witness growing competition and conflicts over the control of natural resources. When assets are depleting, conflicts arise. As a result of the regeneration and restoration of resources, new assets are created and conflicts arise here as well. Conflicts also arise when existing resources are made available to local communities as a result of the decentralization of the management of natural resources. Conflicts occur all too often between upstream and downstream communities over water (Chapter 10), between aquaculture and fisheries (Chapter 4), between governments' objectives for protected areas and indigenous peoples' livelihoods (Chapter 12), and between local forest users and agencies of the state (Chapter 13).

Skills are needed to anticipate and resolve conflicts. CBNRM researchers and practitioners are often called on to contribute to the settlement of conflicts. However, they are not always properly trained for the task of conflict resolution in NRM. It is suggested that more explicit attention be given to this concern. The Tam Giang lagoon case relates how difficult it was for the researchers to address these issues of conflict (Chapter 4). Researchers are aware of the conflicts; however, they lack the knowledge, skill and experiences required to build awareness of the conflicts and change people's attitudes and behaviours. The IIRR case (Chapter 13) demonstrates how the recognition of fundamental conflict reoriented the entire research effort, leading to creative interventions directed specifically at improving conflict management.

The cases have amply demonstrated how important facilitation skills are for CBNRM researchers. The teams dealt with conflicts in many interesting ways. For instance, conflict resolution processes helped build trust, brought parties together and, in the case of Bhutan, employed role-playing exercises to build mutual understanding. Researchers helped build external policy support that resolved conflicts over water in Bhutan and reduced conflicts over pasture in Mongolia. Failures and successes were both evident while resolving conflicts over the use of the lagoon in Vietnam; intercommunity conflicts in resource claims in the Cordillera; and conflicts caused by the CBNRM interventions themselves in Laos.

Facilitation skills are also needed to discern to which issues the participatory methods apply. Some public issues require the reconciliation of stakeholders' priorities with agency priorities so that policy formulation can be tackled through participatory methods. This is because such methods are more suitable to the investigation of issues which are already shared by the community members. However, these methods may not be as successful in the investigation of issues that divide community members, such as inequalities in the family over control of land and other resources, or violence against women or children. A high level of community participation is likely to make it more difficult, rather than less, for those with less power in the community to raise issues which may be seen as private (Laws, 2003: 65).

Reforming research organizations

Research organizations are transformed by implementing CBNRM projects. Researchers engaged in PAR undertaken within a broad context need mandates from their own organizations that will not only allow – but also actually encourage – the forging of partnerships with a range of stakeholders. In addition to the reinforcement of teamwork, this requires the adoption of an interdisciplinary orientation as boundaries of various disciplines become diffused, and as research projects opt to address problems within a more complex and perplexing institutional and socio-economic environment. Institutional mandates need broadening in order to address a wider range of issues through new mechanisms and modalities of research. Reward systems that feature incentives and compensations in the organization, as well as methods of evaluation that favour a concern for the poor, are also important.

The new orientation has important implications in terms of organizational structure and environment, processes and staffing, not to mention the attitude and behavioural changes in research administrators and scientists. Some of the difficulties that the researchers experienced in these cases include the following.

• University instructors and professors in Hue were challenged not only to prepare scientific studies and technical analysis, but also to organize information-sharing and multistakeholder negotiations for lagoon planning (Chapter 4).
  
  
• The Bhutan research team struggled to introduce social science perspectives to their natural science backgrounds, and to integrate across sectoral studies and issues (Chapter 10).
  
  
• The GAAS researchers in China were not accustomed to public speaking or advocacy, yet had to develop diplomatic and negotiating skills when dealing with government officials. These were areas in which they felt they needed more training (Chapter 9).

Agricultural research organizations were originally designed to deal exclusively with the enhancement of productivity and commodity-oriented scientific research. Simply revising their mandates, refocusing conceptual frameworks and adding a new interest in poverty might not suffice. These CBNRM cases show how donors, international and regional research establishments can play critical roles in testing, developing and scaling up institutional approaches that have a clear focus on pro-poor issues. Local research organizations and researchers require assistance in institutionalizing CBNRM research. This is not limited to methods and tools, but includes strategies of scaling up research outputs, and improving the impact of research results on policy and uptake among local policy-makers in addition to interorganizational networking. The expanding skill set demanded for effective pro-poor research requires ongoing attention to research capacity-building. This means building on work which has started in some of the cases, such as the development of communications tools, gender and social analysis, as well as PM&E.

Conclusion

These CBNRM research cases demonstrate that research can respond to the issues and concerns of the poor, even as they seek solutions to the problems caused by the continuing degradation of natural resources. The teams profiled in these cases adopted and executed PAR as their core strategy, choosing project sites in marginal areas and equipping themselves with a problem-oriented and people-centred perspective. Through PAR, the project teams facilitated the ability of poor communities to act effectively in their own interests, whether this was defining a research agenda, undertaking trials and experiments or evaluating research results. Poor communities were assisted to contribute to changes in policies and procedures of natural resource governance at the local as well as national levels. They have been enabled to enter into dialogue with government officials and other user groups in multistakeholder negotiations over appropriate arrangements for the management and use of resources on which their survival depends.

To be more effectively pro-poor, research in CBNRM must sustain its site-specific work, and continue to apply holistic and integrative perspectives that enlighten our understanding of the complex linkages among ecosystems and social relations. Pro-poor research in NRM cannot be limited to confronting issues of external equity that arise from the relationship of poor communities in marginal and less favoured areas to external agencies and groups. It must also consciously seek the poor and disadvantaged groups in the community and ensure that their specific needs are brought forward and addressed. The concern for the sustainable management of resources must include giving importance to the poor's livelihoods; ensuring the poor's access to forests, waters, fisheries and pastures; and protecting the village's resource rights, both individually and collectively.

Working for the interest of poor communities as well as with the poorer groups within them demands that individual researchers and research organizations acquire new attitudes and skills. This includes the recognition that people are subjects – not objects – of research and that the poor are their own main resources in their quest to overcome poverty. Researchers and research organizations must become facilitators, mediators and partners in an empowering process of learning and transformation.

Acknowledgements

We gratefully acknowledge the comments and suggestions of Stephen Tyler, Tony Beck, Liz Fajber, John Graham, Peter Vandergeest and Hein Mallee. All interpretations and errors are ours.

Document(s) 18 of 24