Document(s) 5 of 15

Introduction

Despite notable productivity-related successes, traditional, scientist-led, technical research in natural resource management (NRM) has come under criticism from farmers, donors, and even scientists themselves (GFAR, 1999, 2000) for concerns about weak relevance and adoption. In response to this critique, participatory research approaches – with more user-oriented, flexible methods and a different set of assumptions about research – have emerged. Yet as these participatory approaches have gained momentum, criticisms of their application have also emerged, especially around ‘scientific rigour’, generalizability and naiveté about power relations (Cooke and Kothari, 2001). Furthermore, the relationship between traditional and participatory research is often confusing, the lines between them blurred. Despite increasing interdisciplinarity in enquiry and innovation processes, a multitude of tensions and even scepticism surrounds the two approaches. These tensions have created the risk of NRM researchers aligning themselves either with participatory or with traditional research, and missing opportunities to gain from the strengths of the other. In this chapter, we seek to address this risk by exploring some key dimensions (see below, ‘Traditional and participatory research: key dimensions of difference’) and the strengths and weaknesses (‘Putting it together: reflections on navigating the research spectrum’) of both approaches, as well as the related concept of diversity analysis (‘Diversity analysis in NRM research’). Through this exploration, we underscore the complexity and dynamism (‘The challenge: complexity, diversity and dynamism in human and natural landscapes’) inherent in the human and natural systems that

NRM research addresses. We argue that the desired improvements in NRM demand that research institutions assess, more explicitly and thoughtfully than ever before, the multiple facets of traditional and participatory research approaches, and consciously craft appropriate and innovative combinations of approaches for each research initiative.

The challenge: complexity, diversity and dynamism in human and natural landscapes

NRM research faces multiple challenges if it is to contribute to environmental sustainability, improved livelihoods and equitable social development. Many of these can be traced to three factors that underpin the resilience of human and natural systems: complexity, diversity and dynamism. We will briefly explore here how these affect both human and biophysical aspects of NRM systems.

NRM takes place in complex human landscapes. Multiple stakeholders such as local people, various levels of government, non-governmental organizations (NGOs), and private sector actors have different perspectives, interests, entitlements, knowledge, capabilities, values and power (see Chapter 3 in this volume). This is true at all scales of NRM, and in all ‘sectors’, including forestry, fisheries and agriculture. Within stakeholder groups, tremendous diversity also exists – a reality that dispels commonly held assumptions of homogeneous, consensual ‘communities’ and the existence of stable, universally valued ‘environments’ (Leach et al, 1997, in Anderson, 2001). Within a single community forest user group, for example, there are overlapping categories of human diversity, such as gender, age, ethnicity and caste, religion, wealth and proximity to resources. These ‘internal differences’ underpin critical issues of equity, power and access to resources and decision-making. This human landscape is also dynamic in nature. This is especially true in today’s global economy: there are no closed social systems (Anderson, 2001); governments are decentralizing; roles and rights are changing rapidly; rural people are often relocating in search of viable livelihoods or to escape environmental or political hardship; and households rely on a constantly changing mix of livelihood activities and strategies.

One of the implications of this human system of complexity, diversity and dynamism is that individuals and institutions face constant changes in terms of risks, opportunities and decisions. The majority of decisions in NRM affect a number of different stakeholder groups, and may affect them differently. Especially where resources are scarce, or have a high value, or where differences in power exist between and within stakeholder groups, NRM becomes an on-going process of negotiation and conflict management. This varies in nature and by degree; from the forging of agreements, through ‘hidden’ undercurrents, to explicit violence – (such as in illegal logging conflicts in Indonesia). Anderson (2001) insightfully notes that the role of human diversity, or pluralism, is ‘somewhat paradoxical since it provides some of the force that can break down or inhibit cooperation and collaboration, while it also provides basic forces for essential elements of robustness and adaptability’.

Source: Adapted from Anderson (2001), and drawing on concepts from Denslow (1980), Grubb (1977), Picket and White (1985), Maturana and Varela (1980), Capra (1997) and Holling (1995)

Figure 2.1 Key characteristics of natural systems

One human complexity issue of particular interest in the context of this chapter is the relationship between local people and NRM researchers. Traditional resource managers, such as farmers, may have an extensive understanding of local systems and an interest in applied learning that might lead to increased livelihood security and benefits. In contrast, the aims of most scientists are to understand systems or their individual components, monitor changes, determine responses to management and predict trends and impacts over various time periods. These different knowledge bases and interests reflect Anderson’s (2001) reference to the potential for ‘clash’ or, we hope, complementarity.

As with human systems, natural systems also need to be understood in terms of complexity and dynamism. The Cartesian view of the world provided a relatively simple way of understanding ‘nature’: that is by dissecting it into smaller and smaller pieces (Capra, 1997). This ‘building block’ worldview

evolved to recognize how ‘blocks’ are organized into systems. More recent, and still emerging, perspectives recognize that natural systems can more accurately be understood as self-organizing processes driven by the ‘messy’ principles of dynamism and complexity, as well as unpredictability and openness. Figure 2.1 highlights these principles.

To make matters more challenging still, human and natural systems are obviously interlinked. If the management of these systems is to be effective, it must parallel their interactive, dynamic and adaptive nature (Anderson, 2001; Costanza et al, 1997, 2000). If research is to successfully contribute to NRM, then the research itself must also embody these qualities. The question then becomes how to translate this into practice. The history of formal NRM research reflects a culture of research endeavours set up along political, sectoral and disciplinary lines, with specialists operating independently on a narrow set of issues, and agendas dominated by short-term, problem-solving concerns (Shreier, pers comm). If complex and diverse human and natural systems, and the research systems that are applied to them, can be brought into harmony, the current trend of diminishing returns and decreasing relevance from research might be reversed. It is for this reason that we now turn to an exploration of participatory and traditional research, and their joint potential to contribute to positive social and environmental change.

Traditional and participatory research: key dimensions of difference

Traditional (or conventional) research and participatory research do not exist as neatly definable and independent concepts. These terms refer to collections of approaches and experiences, which theorists bundle together out of convenience and necessity, as a way of making sense of experience. We draw from several sources in this chapter to outline those dimensions that we find useful (Biggs, 1989; Probst et al, 2000; Pretty, 1994; Milne et al, 2001; Lilja and Ashby, 2000; Johnson et al, 2000).

Who owns and controls the research?

NRM research generally refers to enquiry in which there are both local and external actors involved in some way in an innovation process (Probst et al, 2000). One (some would argue the) fundamental dimension of difference between traditional and participatory research is the issue of ‘who controls and makes decisions’ about this process (Lilja and Ashby, 2000). Biggs (1989) (see Box 2.1) offers a well-known four-tier framework for understanding this range of control. Other typologies of participation that researchers are likely to find useful include those of Arnstein (1969), Pretty (1994) and Ingles et al (1999).

The related question of ownership also needs to be considered when defining participation. Who is participating in whose process? Scientists might invite farmers to participate in formal research processes using different types of

participation or, on the other hand, the scientists themselves might participate to varying degrees in a locally owned innovation process (Probst et al, 2000).

What do these types of participation look like in practice? Contractual participation is well illustrated by on-farm research where a farmer provides the land, or socioeconomic surveys in which local people respond to researchers’ questions. In this chapter, the term ‘traditional research’ predominantly involves this type of participation, although it can also include consultative participation, or may involve no participation (that is, pure on-station testing of crops, or pure biophysical assessment of water or forest properties).

The other types of participation are effectively illustrated by the case studies found in the annex to this volume. For example, the McDougall et al case study illustrates research that combines consultative participation (in the background studies, which gave direction to the subsequent action research phase) with collaborative and collegiate participation (in the participatory action research (PAR) phase, in which local people, researchers and other partners confronted local issues of forest management decision-making, equity and income generation). While the background studies in this case were basically ‘researcher owned’, the PAR phase was ‘jointly owned’ by the forest user group (FUG) members and researchers, with the FUG members ultimately taking over and continuing to integrate the institutional innovations into their on-going management processes.

Where does research end and implementation begin?

The second fundamental difference that we highlight here is the significant difference in the links between the research and implementation (that is, application or adoption) phases of development. Traditional research collects results – typically for several seasons – before data are analysed, put into reports

and then ‘released’. These are (ideally) taken up by (separate) extension services and translated into extension messages, which are then disseminated. In participatory research, particularly participatory action research, the implementation of research findings – and the related technical and social changes in the rural areas – is integral to, rather than separate from, the research process (Pound, pers comm).

Other dimensions of difference

In the above sections we located ‘traditional’ research (if it involves local people) towards the ‘contractual participation’ end of the spectrum, and associated ‘participatory’ research with consultative, collaborative and collegiate participation. We also noted that traditional research operates with research and implementation as discrete phases, while participatory research tends to integrate or iterate between the two. But the terms ‘traditional’ and ‘participatory’ research also embody other dimensions. In Table 2.1 we illustrate these differences with a simplified view of the ‘extremes’ of a multi-dimensional spectrum.

Table 2.1 shows how the ends of the spectrum reflect different assumptions and foci (as discussed in the section entitled ‘The challenge: complexity, diversity and dynamism in human and natural landscapes’). We can see a difference in the complexity and ‘activeness’ of the research and in researcher and farmer roles: from single to multiple perspectives and types of knowledge; from neutral or passive roles to active and engaged; and from single level/linear to multiple levels/directions of dissemination. Another point of interest in this matrix, and in the Biggs typology, is the greater degree of overlap in methods compared to other dimensions of difference. Many people assume that any research that uses some participatory methods cannot be ‘traditional’; and that ‘participatory’ research cannot apply ‘traditional’ scientific tools. Methods, however, are less important in distinguishing these research approaches than the other dimensions or the degree of control over decision-making. One of the reasons for this is the increasing frequency with which traditional types of research (for example the ‘transfer of technology’ type, see‘NRM research in practice: four examples of research “types” ’ below) use participatory methods of accessing information as a means of increasing the accuracy of information or its legitimacy in the community.

NRM research in practice: four examples of research ‘types’

If we pull together the types of participation, the question of ‘whose research’, and some of the other key dimensions of difference outlined above, we see emerging patterns of NRM research. In Figure 2.2 we illustrate four ‘prototypical’ approaches to innovation development (adapted from Probst et al, 2000) in relation to control over research and form of participation. These are not the only possible ‘types’ nor are they mutually exclusive or fixed (as represented by the arrows); we present this typology because it is relatively simple, yet informative.

As Mcallister and Vernooy (1999) state so simply, ‘there is no right or wrong amount of participation’, nor is there any single ‘best type’, nor ‘best place’ on

Table 2.1 Comparison of traditional and participatory research on several key dimensions

Sources: Costanza et al, 2000; Biggs, 1989; Pretty, 1994; Lilja and Ashby, 2001; DFID, 1998; Chambers, 1994; Milne et al, 2001; Cooperrider and Srivastva, 1987; von Glasersfeld, 2001; Guendel et al, 2001; Allen, 2001

Source: Adapted from Probst et al, 2000

Figure 2.2 Examples of four ‘prototypical’ approaches to innovation development

the research spectrum. The challenge is for researchers to consciously navigate the research spectrum in order to maximize the effectiveness and positive contribution of their research to NRM and development. Before we go on to discuss this in greater depth, including implications for future research, we turn to another fundamental aspect of NRM research: gender and diversity.

Diversity analysis in NRM research

While it is an integral element of traditional and participatory research, diversity analysis is so significant and complex that it merits separate consideration in this chapter. It is obvious how diversity (see Box 2.2) affects some research

dimensions, such as methods. Yet diversity also interacts in subtle ways with other research dimensions; it influences, for example, the issues of rigour, validity and objectivity. These, in turn, affect the confidence and credibility that can be accorded to research results, and the domains into which they can be disseminated.

What is diversity analysis?

Diversity analysis is more than analysing data by gender or ethnic group. It is an approach in which key elements of human difference, such as gender, wealth, caste, age and ethnicity, become analytical variables throughout the research or programme, from design to implementation, analysis and evaluation. It involves exploring a range of questions and issues spanning both the structure (roles) and dynamics (relations) of human systems. Some of the points for exploration include:

• What are the roles and responsibilities of the different groups relating to natural resource management?
• How and why are these roles, relationships, patterns and differences changing over time?
• What are the differences in how resources are valued?
• What are the differences in the criteria for decision-making about resources and why do they exist?
• Who controls access to resources? Who makes decisions about them and why?
• Who benefits from each activity or enterprise? Who bears any associated costs?

• What are the relationships among the groups? What are the power dynamics?
• How do relationships, power and roles influence the decision-making of the group regarding resources and ultimate outcomes?
• What options exist for increasing equitable access to decision-making and natural resource benefits, especially for marginalized stakeholders?

Many practitioners and theorists increasingly emphasize that the focus of the analysis should be on relations rather than roles (McDougall, 2001), except at a very descriptive level. This is because a focus on roles offers a ‘static’ perspective on issues that are based on power relations, and thus inherently dynamic (Young, 1988). In Box 2.3 we offer a modified version of a typology of gender analysis developed by the System-wide Program on Participatory Research and Gender Analysis (PRGA) of the Consultative Group on International Agricultural Research (CGIAR). The typology ranges from an analysis aimed at a description of roles to a more dynamic approach that seeks to address inequities.

As with the participation typologies (see the section entitled ‘Who owns and controls the research?’), there is no single ‘correct’ type of diversity analysis; instead the researcher must identify the key dimensions of diversity or difference that merit inclusion, and navigate the spectrum of participation to find the appropriate level for the given objectives and context.² The different outcomes of navigating this spectrum are well illustrated by the range of diversity approaches used in the case studies (see Table 2.2), summarized in this volume.

In its shift from descriptive to more action-oriented enquiry, this diversity analysis typology (see Box 2.3) can be seen to roughly parallel the spectrum of traditional to participatory research. The very ‘ends’ of the traditional and participatory research spectrum are fairly clearly linked (for example,

Table 2.2 Types and significance of diversity approaches from case studies

participatory action research and transformation-oriented diversity analysis); the vast middle areas, however, call for considerable attention and flexibility on the appropriate type of diversity approach (eg, consultative research might take descriptive or design-oriented approaches). The potential ‘clash’ that can be identified between diversity approaches and some traditional research occurs where the research assumes that local stakeholder views and needs are homogeneous. It may be that these views and needs are, indeed, homogeneous in a certain case. In cases where there is significant heterogeneity, however, the ‘costs’ of overlooking the diversity may be very high, including limitations to the relevance, effectiveness or adoption of the research, or even the marginalization of certain groups. This potential clash can be avoided by research teams seeking out and examining the validity of underlying assumptions in this area, on a case-by-case basis.

Besides the analytical aspects, there is one further point related to diversity that we suggest as necessary for all parts of the research approach spectrum whenever any local people are involved: design and implementation of research that is sensitive to local stakeholders including, and especially, women and marginalized groups. This refers to the simple, yet still sometimes overlooked aspects of accommodating workloads, cultural and other factors that may create difficulties or discomfort for local people in research settings.

Diversity analysis, research and NRM

Experience has disproved the projections of development pundits who believed in the ‘trickle down theory’. The assumption that targeting development interventions at male heads of households would equally benefit other household members, particularly women, has not been validated in practice. (Sarin, 1997)

Natural resource managers’ perceived opportunities and constraints in decision-making, and their resulting actions and behaviours, are not only determined by the natural system in which they exist; they are significantly determined by their different (diversity) identities, including their interests, roles, knowledge and vulnerability and power (Schmink, 1999). In this way, diversity analysis in NRM contributes to ‘a more accurate and complete picture of a complex social landscape’ (McDougall, 2001). Research that better reflects the experiences of diverse (especially non-dominant) groups is more likely to lead to NRM policies or programmes that take into account those different experiences and aim for more sustainable and equitable impacts. Furthermore, a diversity approach tends to bring to light the ‘invisible’ poor stakeholders, and elucidate the fact that relative well-being is neither neutral nor random. In this way, more accurate and complete assessments also lead to more effective and efficient impacts of research, policy and development programmes (McDougall, 2001; Wilde and Vainio-Mattila, 1995).

While diversity analysis (and sensitivity) is necessary across the research spectrum, diversity analysis and participatory research are mutually reinforcing. In collaborative or collegiate research there is shared responsibility for, and

ownership of, the outcomes resulting from choices and decisions made jointly. Ideally, by taking part in research, learning, and negotiation, all those who participate become collectively responsible for decisions and for the resulting outcomes. They become the owners of their choices, and the research and consequent NRM process may be ‘democratized’. As diversity analysis addresses the key issue of who should participate in decision-making about NRM, it can enable multiple stakeholders to critically assess how different users affect, and are affected by, the status quo and innovations in technology, institutional arrangements, management practices and information flows. Used effectively, participatory research grounded in diversity analysis can draw out and build on the range of perceptions, interests, relations and power, to form the basis for a consensus-building process for increasingly equitable and sustainable NRM. At the same time, these changes are not without risks. These processes can shed light on and activate dormant or latent conflicts. Pound (pers comm) suggests, as well, that enhanced equity in NRM institutions can also reduce motivation and contribute to the loss of local natural resource enterprise viability.

A final and important question is raised here about the nature of diversity analysis: because it is rooted in a movement for social transformation, and is essentially ‘political’ in nature, is diversity therefore also transformative about other relations? Specifically, does it redefine concepts of ‘subject–researcher’ relations? Does it challenge traditional notions of ‘objectivity’? Certainly diversity analysis at least demands a certain degree of critical self-reflection by researchers in terms of their roles and relationships with the ‘participants’ or ‘subjects’ of the research. This, and the potential reflection on researchers’ own identities, influences and biases, pose a (welcome or sometimes unwelcome) challenge for research teams. (McDougall, 2001)

Putting it together: reflections on navigating the research spectrum

The foregoing sections explored the challenging context of NRM research, some dimensions of traditional and participatory research, and the indispensable concept of diversity analysis. The goal of this section is to contribute to critical reflection on research design by highlighting several key factors that relate the research approaches to the desired NRM impacts. The ‘litmus test’ of an NRM research approach is, of course, the extent to which it contributes to creating greater choice of livelihood and environmental options, benefits and security, while maintaining the quality of the natural resource base. If either traditional or participatory research alone had excelled at this challenge, then the choice of research approach would be relatively simple. In fact, both traditional and participatory research approaches have produced mixed and uneven contributions towards these goals. While there have been

advances in NRM and human well-being that relate to traditional research, for example, its record has been inconsistent and limited, and where it has contributed, it has tended to do so only for the middle-income to wealthy farmers (Conway, 1997). Traditional research also faces some valid questions in terms of its impact on long-term food security, because of its potential narrowing of the agricultural genetic base in farming systems. The field of biotechnology is also faced with questions about its potential environmental and human health threats. While participatory research, especially participatory action research, has made some significant progress in increasing well-being in some communities and management systems at a neighbourhood level, these impacts have been predominantly localized (Guendel et al, 2001). Furthermore, participatory research experiences have also abounded in which the quality of participation was questionable and/or the attention to gender and diversity was not sufficient to counter a bias towards the local elite (Cooke and Kothari, 2001). This being the case, in this section we consider several other factors that can help in the navigation of the research spectrum, including the linking of research approaches to objectives, research questions, knowledge and communication.

Research objectives and approaches

In this section we highlight three ways in which the level and kind of participation needs to be adjusted to the research objectives, defined collectively by researchers and other stakeholders. The first, and perhaps most ubiquitous, issue is the matching of type and degree of participation to the nature and quality of information required for each research case. Questions such as the following may help guide research teams through these decisions:

• What kind of qualitative versus quantitative information is needed for the desired analysis?
• Do some stakeholders (for example, credit organizations or variety release boards) require quantitative information from the experimentation process?
• To what extent does our research question require participation, and diversity analysis, to achieve the appropriate level of accuracy (and specificity) of information?
• What are the tradeoffs between local accuracy and the generalizability of findings, and how can they be addressed?
• To what extent can the ‘hypotheses’ be tested ‘in action’, in complex social or natural environments, and to what extent do they require a more controlled environment?

Research with a local-scale agenda, may do well to have a highly participatory and qualitative research process (possibly with a transformation-oriented diversity approach). The degree to which the research questions and objectives demand the ability to extrapolate the findings beyond the site will influence the design in terms of mechanisms for ‘scaling up’, such as comparative cases with common variables, or tools such as geographical information systems (GIS, see

Chapter 4; van de Fliert and Braun, 2002; Guendel et al, 2001 and the McDougall et al case study in this volume). On the other hand, research into a limited number of components of a specific biophysical process may be well suited to more ‘station-based’, scientist-led, disciplinary research, which provides an opportunity for ‘controlled’ experiments where the effect of individual variables can be isolated and compared. This ties into an important role of traditional biophysical research, the development of general theories for the structure and dynamics of natural and managed ecosystems, such as agroforestry and fishery systems (Schreier, pers comm; Settle, 1997).

The farmer field school experience (see Nelson case study, this volume; Braun et al, 2000) illustrates how traditional research can contribute to locally specific research needs, and how traditional and participatory approaches feed into one another. Farmer field schools emphasize experimentation aimed at discovering how the local agroecosystem operates as the foundation for decision-making. For example, in some contexts, experiments have helped farmer field school participants to realize that a reduced use of pesticides permits the development of larger populations of beneficial species capable of controlling pests. Farmers can translate this knowledge into relatively simple decision rules governing their pesticide use. In other contexts, for example where there are recently introduced species, local understanding of ecosystem components and inter-relationships is not sufficiently developed to permit the development of good field school learning exercises. In these cases, farmers cannot easily formulate simple decision rules (Settle, 1997), and thus there is a need for a more explicit link with formal research (as well as development of farmer capacity for controlled experimentation and relatively rigorous data collection (Loevinsohn et al, 1998; Whitten, 1996; Braun et al, 2000).

The second issue relating objectives to participation is that any research initiative involving local people must define its social goals and its level of obligation to improve the situation of local people. De facto responses to this question range from no responsibility, through return of research findings in an appropriate manner, to engaging in transformative activities at the sites. In fact, in many cases, this decision is made implicitly (by the research team), rather than explicitly, both at the research project and institutional level. This is indicative of the dimension of difference between traditional and participatory research regarding the degree to which the research merges into dissemination and application of the results (see the section entitled ‘Where does research end and implementation begin?’ above). Our own case experiences suggest that negotiating this decision in a more explicit way can help to strengthen the clarity of the design and objectives (and expectations) for researchers and local people.

The third, and related, issue in the consideration of objectives and corresponding research approaches is that participatory processes can catalyse institutional learning in the research team (as opposed to the acquisition of knowledge by scientists on NRM issues). They not only potentially change the kind and source of the information gathered but, more fundamentally, require that the research teams loosen their control over the ‘lens’ through which different NRM options or scenarios are viewed. Research teams that put a premium on their own learning as the basis for adaptive approaches to research may therefore

seek participatory approaches to their initiatives (and likely to their own internal processes as well – see Chapter 6).

Strengths and weaknesses of research approaches

Besides the implications of research objectives, several other factors may play a definitive role in developing an appropriate research approach for a given initiative. Table 2.3 and the following four sub-sections explore some differences, strengths and risks of traditional and participatory research relative to four underlying conditions for successful NRM research:

1. asking the ‘right’ research questions – those that are critical and relevant to stakeholders;
2. integration of local and scientific knowledge;
3. effective communication of research findings to those who need them;
4. assuring that ‘costs’ of inputs to research are acceptable to local stakeholders and researchers. (Pretty and Hine, 2001; Chapter 6, this volume).

Asking the ‘right’ research questions

No matter how good the research, if the question it is seeking to address is not the right one, then its impact is limited. Gladwin et al (2002), for example, suggest that the seemingly crucial question of ‘how much nutrient a farmer should put on her soil given the desired output’ is not the ‘right’ question for most poor African farmers. Although these farmers desire higher outputs, and being able to predict the impact of nutrient levels would be potentially useful, their resource constraints do not permit them the luxury of responding effectively to that information. The type of questions they need answers to are more like: ‘How much nutrient can we afford to put on, and how much yield will that give and how will we make up for the gaps? (Gladwin et al, 2002)

From Table 2.3, we can see that participatory research (especially the critical objective setting and design phase) offers the advantage of acting as a check on the relevance of research questions to local people (the ultimate beneficiaries). Yet, great local relevance may pose certain risks as well. It may make the research questions (and outcomes) so specific that it limits their generalizability to other areas, and it may shorten the ‘lifetime’ of the relevance of research questions and their outcomes. Traditional research, on the other hand, typically addresses ‘slower moving’ issues, and thus research questions with longer time scales. Furthermore, while in theory the outputs of traditional research may be more readily transferred to greater geographical scales than locally oriented participatory research, in practice its results may be of limited ‘real world’ relevance (especially in the context of relatively more marginalized lands and farmers). Clearly some articulation between the two, as well as appropriate application of diversity approaches, would lend strength to NRM research in terms of an ability to balance issues of scale. The McDougall et al case study in this volume offers one example of an effort towards such articulation. This research project has a set of overarching strategic research questions (that is, conditions and strategies for, and outcomes of, adaptive and collaborative management of forests) and a set of basic variables, shared by sites in

ten countries. Researchers at each site use these questions and variables for guidance in their own research design, and respond to them as a part of both their initial context studies and their final analysis. This shared framework enables cross-site and cross-country comparative analysis. Meanwhile, at the site level, researchers and local stakeholders engage in action research that falls within this research framework ‘umbrella’, yet addresses a locally specific set of priority NRM issues. For example, they explore the possibilities for improving equity and accountability within a forest user group, and enhancing forest-related livelihood benefits, through joint self-monitoring and adjustments in decision-making processes.

Integrating local and scientific knowledge

The dangers of theorizing while safely ensconced in the ivory tower are not exaggerated. Yet . . . throwing out the methods of modern science along with quantification and statistics . . . is putting the researcher in more danger – the danger of being wrong with no way to show it. By contrast, the scientific method requires the researcher to model their interpretation of reality by generating a hypothesis about people’s behavior, then collect observational data to test the model, then revise the model based on the test results. This hypothesis-testing sequence is the basis of science. Without it, researchers have no way of giving themselves a reality check. (Gladwin et al, 2002)

The value of the knowledge generated is perhaps one of the most hotly contested of the debates in development science, because it includes questions of rigour and generalizability, as well as different knowledge worlds. As illustrated by many of the case studies in this book (Dey, Stroud, Vaughan, Snapp and Rohrbach, Heong, Nelson and Braun), research that intends to produce agricultural or NRM technologies or processes for ‘adoption’ and ‘adaptation’ by farmers needs to integrate the best of local and scientific knowledge worlds, or else risk failure.

We start by addressing the issues of knowledge worlds: what is the relevance of the differences in knowledge worlds in NRM research? Traditional researcher-led experiments draw on the scientific method to provide information on theoretical maximum effects under controlled conditions, where constraints are minimal. They also provide understanding of key processes. Local research by farmers and other resource users provides responses under realistic management conditions where a wide range of constraints may affect the outcome. In traditional researcher-led experiments a limited number of factors are tested and hence the extrapolation of the results to more complex settings is problematic. In farmer-led research, on the other hand, it is more difficult to assess the direct causes of diminished performance, because of the fact that it is addressing issues in a complex and dynamic system.

In terms of the ‘validity’ of that knowledge, Chambers (1994) points out that there is significant evidence of ‘local knowledge’ being more ‘accurate’ than

Table 2.3 Strengths, weaknesses and risks of participatory natural resource management research approaches (relative to traditional approaches)

Table 2.3 Strengths, weaknesses and risks of participatory natural resource management research approaches (relative to traditional approaches)(Continued)

scientific knowledge in some issues; thus the need for ‘science’ to be humble in its claim for validity. At the same time, the reverse may also be true. Local resource managers may notice the symptoms, but incorrectly assign the cause. For instance, herders in east Africa avoided grazing their cattle on some areas because of their fears of ‘poisonous grasses’ that made their cattle sick, when it was actually tsetse flies causing trypanosomiasis. In Asia, K L Heong points out that the very visible damage done to the early stages of rice by leaf folders resulted in many farmers spraying their crops in the belief that this would lead to higher yields. In fact, carefully controlled studies showed that the leaf folders did not lead to economic damage, and that spraying was unnecessary (Heong, pers comm).

The issue of the ‘quality of science’ or ‘rigour’ has also been at the heart of much debate on the knowledge derived from traditional and participatory science. The wealth of experience of ‘rigorous’ research that did not take into account local needs, interests, knowledge and preferences, and resulted in ‘scientifically’ valid, but unused, outputs helps to contextualize this issue. Clearly, classical ‘rigour’ is not in itself sufficient if the science cannot transfer into real world, complex systems, and ultimately to benefits. In these cases, traditional research can profit from combining with holistic complex local knowledge through the incorporation of participatory approaches. Equally, there are some valid critiques of potential shortcomings of participatory research in gathering and analysing data. Gladwin et al (2002) point out three of these, in relation to ‘rapid rural appraisal’ types of research (which are participatory – but tend to be on a very short timeline):

1. each member of the research team tends to work with, and then draw generalizations from, too small a sample size (and these generalizations are pooled with other generalizations);
2. time pressures contribute to ignoring variations in farmers’ decisions and practices, and focusing instead on similarities;
3. the hypotheses and generalizations generated may remain untested, mainly because much of the data may remain uncoded and unanalysed.

In these cases, the research risks its quality, and/or its generalizability.

Research teams can avoid these pitfalls of participatory research, by directly and creatively addressing them through careful design (while maintaining flexibility), detailed observation and record keeping, and explicit analysis with clear documentation. The design can enhance the quality of its local findings by drawing on the essence of the scientific method (ie, reality checking through testing ‘assumptions’), not in a rigid way, but innovatively and as appropriate (eg, the Braun case study in this volume; Gladwin et al, 2002). Several sources highlight ways to safeguard the ‘quality of science’ in participatory and ‘soft systems’ research (see Chapter 5, this volume; Pretty, 1994; Chambers, 1994; Dick, 1997; Checkland, 1981; Checkland and Scholes, 1990). We highlight some of these in Box 2.4.

Integrating local knowledge and ‘scientific’ knowledge is not a simple task. As traditional research seeks to increase its ultimate effectiveness through the

incorporation of local knowledge and preferences, it comes face to face with diversity of all kinds, and needs also to seek the appropriate place on the diversity spectrum. Approaches for ensuring ‘rigour’ employed in the research cases in this volume include: the use of common strategic questions and analytical variables across PAR case studies; participatory modelling; the use of forums for the exchange of experimental results; analysis at different temporal and spatial scales; and the use of GIS for extrapolation and the integration of qualitative and quantitative information (see the Vaughan, Heong, Nelson, Braun, Vernooy and Schreier case studies in this volume).

Effective communication and application of research findings

A participatory research approach can contribute to NRM research its experience in the establishment of local communication and research findings ‘uptake’ mechanisms. The on-going horizontal, multi-directional information sharing between external and local stakeholders, and within local groups, which is typical of participatory research, may help to correct distortions in information. This distortion correction function should not be underestimated; most NRM research is not only operating in a complex system, but at the intersection of multiple

complex systems, and the likelihood of misinterpretations by research is very high. Furthermore, this kind of communication is a critical part of generating the immediate application of research findings and thus research impact.

Traditional research offers powerful methods of communicating to audiences well beyond the research site, including international donor agencies, national policy-makers, and extension services. In this case, participatory research experiences can benefit from borrowing strategies and tools for scaling-up results, such as GIS and modelling. The attention to explicit documentation and analysis contributes, also, to the ability of participatory research to effectively use the dissemination channels of traditional research, including scientific journals and policy fora. In other words, it seems very likely that combining the communication (and related impact) strategies of participatory and traditional research, on a base of high quality research, will generate movement towards research that is both relevant locally, and effectively and broadly disseminated and applied.

Costs of inputs to research are acceptable to local stakeholders and researchers

Participatory research is generally viewed as having higher time and ‘effort’ costs than traditional research, in the sense that a research team can do surveys in dozens of villages in the time that it takes to do participatory action research in one village. Although this is true in relation to the intensity of effort involved, this perception is also somewhat flawed because it is comparing initiatives with different objectives and processes. As noted in the section ‘Where does research end and implementation begin?’, above, traditional research views the research findings and/or output as the product, whereas participatory research typically views greater livelihood security and positive environmental change as the product. In other words, the participatory research ‘timeline’ necessarily involves an ‘impact’ phase that merges with the research phase itself (Pound, pers comm).

A related point, distinguishing the ‘efforts’ involved in research, is the nature of local stakeholders’ decision-making regarding their involvement (and expectations) in research. Traditional research is extractive, and there is no obvious research findings-related benefit to communities; their time is generally either paid for (in cash or in kind), or donated (out of interest, hoped-for benefits, curiosity, a sense of obligation or a variety of other motivations). In well-conducted participatory research, local stakeholders should be clear from the start, and throughout, about the costs and benefits that might accrue to different groups; in other words, they make a conscious decision (and an on-going series of decisions) whether or not to invest their time and effort in a long-term process. An additional consideration here is that the impact in the early years often gives an advantage to the stronger members of a community. This may be due to it taking longer for the socially weaker, less secure members to join the processes of joint experimentation – for reasons that may include time and costs. This can also be because research benefits relating to long-term improvements in financial and natural capital can depend on certain levels of human and social capital, and these must be built up during the research process as well.

Participatory research can speed up the innovation process. For example, participatory plant breeding consistently leads to faster release and dissemination of locally accepted varieties (PRGA Program, 1999). The key is sharing the responsibility for selection with farmers early in the breeding process when the amount of genetic variability is at its maximum (Ceccarelli et al, 2000). In one successful example, Syrian farmers working with barley breeders began producing seed from a few of their own best selections after only two growing seasons (PRGA Program, 1999). On the other hand, as research becomes more participatory, issues of democracy, equity and voice become increasingly significant – all of these require careful process and cannot be rushed. In these cases, and especially where participatory research involves significant or difficult negotiations of interests (either between external and local stakeholders, or among local stakeholders), it will add time to the research process. This cost must be weighed against the potential gains of this type of research; the outcome of this weighing-up will be different for different issues and contexts. As a whole, it seems likely that the future of research may indeed require a shift in the time and effort allotted to NRM research activities. If our hypothesis is correct – that NRM research requires both traditional and participatory research – then these costs may increase slightly; if donors and institutions are committed to impact, then they may have to adjust the traditionally short (three-year) research activity time frames accordingly.

Another critical input to the research process, which differs by approach, are the skills of researchers (including local researchers). Participatory research approaches require new skills, such as facilitation, which have not been part of the training of traditional researchers. The most challenging component of any participatory research approach is the organization and maintenance of the stakeholder processes (Cooperrider and Dutton, 2001). Success, therefore, hinges on positive interactions and creating a spirit of collaboration between researchers, local peoples and other partners. Participatory approaches require that team members have exceptional ‘people skills’. This implies that they should be culturally sensitive, tolerant, diplomatic, motivated to collaborate and possess appropriate enquiry skills (eg, avoiding the use of leading questions in favour of open and probing questions). They must be able to engage in multi- or interdisciplinary analysis, and be flexible while seeking ways of maintaining scientific rigour. These skills are necessary so that stakeholders can reach agreement on common objectives, keep the interactions transparent, and maintain open and active trust and communication. While access to these skills may pose a challenge to research institutions in the short term, it seems likely that these costs will be reduced somewhat with the current trend of increasing numbers of interdisciplinarians in research, and a greater emphasis on building teamwork skills, and collaboration in and between research teams.

Participatory action research is like 19th century physics: the best researchers are the ones with the handicraft skills for building the (social) apparatus necessary to test their hypotheses. (Dean Holland, pers comm)

Conclusions

There will always be some cases where pure ‘traditional research’ or pure locally oriented ‘participatory research’ are the appropriate approaches. However, if the assessment above is valid – that appropriate combinations of traditional and participatory research make for more relevant research questions, better knowledge bases and communication – then we can extrapolate a lesson from it for current and future NRM research: NRM research is more likely to achieve livelihood and environmental benefits through thoughtful and appropriate combinations of traditional and participatory research approaches than it is through adopting either approach on its own. This assumes that such combined methodologies can maintain the strengths of both traditional and participatory research vis-à-vis impact, including:

1. maintaining the generation of local benefits in the research sites as participatory research often has;
2. exploring both locally and generally relevant research questions, including the desirability or adoptability of the technologies or processes generated;
3. applying diversity analysis (including gender) to increase inclusivity and relevance;
4. applying appropriate aspects of the scientific method and good documentation and analysis for the validation of results;
5. creating impact by influencing stakeholders (including policy-makers at various levels) through research.

We are at a global crossroads in terms of human and environmental development. Research in NRM needs to respond more effectively than it ever has before, if we are to successfully meet the local, regional and global challenges facing humanity. And yet, NRM research itself also appears to be at a crossroads, with some latent tensions surrounding traditional research on one side and participatory research on the other. This is further complicated by the increasing recognition of diversity as a critical, but as yet weakly implemented, factor in development and NRM. Are traditional research, participatory research and diversity analysis compatible? Our response is that although traditional and participatory approaches may have different philosophical roots and other differences, they are not only compatible but, in many cases, they need one another. Together they generate richer and deeper knowledge, and more effective and appropriate technology than either one alone. How should they be combined to achieve this? There is no prescription for developing research approaches, nor will there ever be. The challenge is for research teams to implement careful, early and on-going assessments of their NRM issues and multiple objectives – through the lenses of complexity, dynamism, gender and diversity – as the basis for the thoughtful and creative building of research approaches for each research initiative. Research teams can use these assessments to sieve through the plethora of research options and decide, with their partners, which aspects of each approach are of value in that context.

Combining aspects of the different approaches thoughtfully and appropriately may enable NRM researchers, development practitioners and policy-makers to access greater understanding of both the fundamental biophysical processes that underlie NRM systems, and the human needs and interests involved. This thoughtful approach to research methodologies may enable societies, both large and small, to begin to deal more effectively and equitably with the challenges and opportunities posed by the complexity and dynamism inherent in human and natural resource systems.

Notes

1. See the Sustainable Livelihoods website of the Department for International Development (DFID): www.livelihoods.org.
2. ‘Given objectives’ refers to those agreed upon by key stakeholders in the research. Depending on the circumstances, this may include diverse local people, researchers, extension agents, non-governmental organizations and others.

References

Allen, W J (2001) ‘Working Together for Environmental Management: The Role of Information Sharing and Collaborative Learning’, PhD Dissertation (Development Studies), Massey University, pp12–29, http://www.landcareresearch.co.nz/research/social/ar_working html

Anderson, J (2001) ‘On the edge of Chaos: Crafting Adaptive Collaborative Management for Biological Diversity Conservation in a Pluralistic World’, in Buck, L, Geisler, C, Schelhas, J and Wollenberg, E (eds) Biological Diversity: Balancing Interests through Adaptive Collaborative Management, CRC Press, Boca Raton, LA, pp171–186

Arnstein, S (1969) ‘A Ladder of Citizen Participation’, AIP Journal, July

Biggs, S D (1989) Resource-poor Farmer Participation in Research: A Synthesis of Experiences from Nine National Agricultural Research Systems, OFCOR Comparative Study Paper ISNAR, The Hague, pp3–37

Braun, A R, Thiele, G and Fernández, M (2000) ‘Farmer Field Schools and Local Agricultural Research Committees: Complementary Platforms for Integrated Decision-making in Sustainable Agriculture’, AGREN Network Paper, No 105, ODI, UK

Capra, F (1997) The Web of Life: A New Synthesis of Mind and Matter, Harper & Collins, London

Ceccarelli, S, Grando, S, Tutwiler, R, Baha, J, Martini, A M, Salahieh, H, Goodchild, A and Michael, M (2000) ‘A Methodological Study on Participatory Barley Breeding I Selection phase’, Euphytica, 111:91–104

Chambers, R (1994) ‘Participatory Rural Appraisal (PRA): Analysis of Experience’, World Development, 22(9):253–268

Checkland, P (1981) Systems thinking, Systems Practice, Wiley, Chichester

Checkland, P and Scholes, J (1990) Soft Systems Methodology in Action, Wiley, Chichester

Conway, G (1997) The Doubly Green Revolution Food for all in the 21st Century, Penguin Books, London

Cooke, B and Kothari, U (eds) (2001) Participation: The New Tyranny?, Zed Books, London

Cooperrider, D L and Srivastva, S (1987) ‘Appreciative Inquiry in Organisational Life’, Research in Organisational Change and Development, 1:129–169

Cooperrider, D L and Dutton, J (2001) ‘No Limits to Cooperation: An Introduction to the Organization Dimension of Global Change’, http://appreciativeinquiry. cwru.edu/research/bibPapers cfm

Costanza, R, Daly, H, Folke, C, Hawken, P, Holling, C S, McMichael, A J, Pimentel, D and Rapport, D (2000) ‘Managing Our Environmental Portfolio’, BioScience, 50(2):149–155

Costanza, R, d’Arge, R, de Groot, R, Farber, S, Grasso, M, Hannon, B, Limburg, K, Naeem, S, O’Neill, R V, Paruelo, J, Raskin, R G, Sutton, P and van den Belt, M (1997) ‘The Value of the World’s Ecosystem Services and Natural Capital’, Nature, 387:253

Denslow, J S (1980) ‘Patterns of plant species diversity during succession under different disturbance regimes’, Oecologia, 46:18-21

DFID (Department for International Development) (1998) Action Research for Community Forestry: Sharing Experiences from Nepal DFID, London

Dick, B (1997) ‘Rigour and relevance in action research’, http://www.scu.edu.au/schools/gcm/ar/arp/rigour html

Edmunds, D (1998) ‘Notes for CIFOR Gender Analysis Training in Yunnan’, Centre for International Forestry Research, Bogor, Indonesia

GFAR (Global Forum on Agricultural Research) (1999) ‘Activities of the NARS Secretariat: Implementing the Programme of Work Towards a Global Agricultural Research System’, Global Forum for Agricultural Research, FAO, Rome, Italy

GFAR (Global Forum on Agricultural Research) (2000) ‘GFAR-2000 Conference, Dresden Germany, 21–23 May 2000, Highlights and Follow-up Action’, Global Forum for Agricultural Research, FAO, Rome, Italy

Gladwin, C, Peterson, J, and Mwale, A (2002) ‘The Quality of Science in Participatory Research: A Case Study from Eastern Zambia’, World Development, 30(4):523–543

Grubb, P J (1977) ‘The Maintenance of Species-richness in Plant Communities: The Importance of the Regeneration Niche’ Biological Review, 52:107–145

Guendel, S, Hancock, J and Anderson, S (2001) Scaling-up Strategies for Research in Natural Resources Management, Natural Resources Institute, Chatham

Holling, C S (1995) ‘What Barriers? What Bridges?’ in Gunderson, L H, Holling, C S and Light, S S (eds) Barriers and Bridges to the Renewal of Ecosystems and Institutions, Colombia Press, New York

Ingles, A, Musch, A and Qwist-Hoffmann, H (1999) The Participatory Process for Supporting Collaborative Management of Natural Resources: An Overview FAO, Rome, Italy

Johnson, N, Lilja, N and Ashby, J (2000) ‘Using Participatory Research and Gender Analysis in Natural Resource Management: A Preliminary Analysis of the PRGA Inventory’, Working Document No. 10, CGIAR Systemwide Program on Participatory Research and Gender Analysis, Cali, Colombia

Lilja, N and Ashby, J (2000) ‘Types of Participatory Research Based on Locus of Decision-making’, PRGA Working Document No. 6, CGIAR Systemwide Program on Participatory Research and Gender Analysis, Cali, Colombia

Lilja, N and Ashby, J (2001) ‘Overview: Assessing the Impact of Using Participatory Research and Gender/Stakeholder Analysis’ in Lilja, N, Ashby, J and Sperling, L (eds) Assessing the Impact of Participatory Research and Gender Analysis, CGIAR Systemwide Program on Participatory Research and Gender Analysis, Cali, Colombia

Loevinsohn, M, Meijerink, G, and Salasya, B (1998) ‘Enhancing Capacity to Manage Resources: Assessing the Farmer Field School Approach’, Second Meeting of the Integrated Pest Management Network for the Caribbean, February 4–6, Kingston, Jamaica: CARDI and CTA

Maturana, H and Varela, F (1980) ‘Autopoiesis and Cognition’ in Autopoiesis: the Organisation of the Living, D Reidel, Dordrecht, Netherlands

McAllister, K and Vernooy, R (1999) Action and Reflection: A Guide for Monitoring and Evaluating Participatory Research, IDRC, Ottawa, Canada

McDougall, C (2001) ‘Gender and Diversity in Assessing Sustainable Forest Management and Human Well-being’ in People Managing Forests: The Links Between Human Well-being and Sustainability, Colfer, C J P and Byron, Y (eds) Resources for the Future and CIFOR, Washington, DC and Bogor, Indonesia

Milne, M, McDougall, C, Siagian, Y and Uprety, L (2001) ‘A Participatory Research and Gender Analysis Impact Assessment: The Local People, Devolution and Collaborative Management of Forests Research Programme’, CIFOR Report to the CGIAR Systemwide Program on Participatory Research and Gender Analysis, Center for International Forestry Research, Bogor, Indonesia

Pickett, S T A, and White, P S (1985) The Ecology of Natural Disturbance and Patch Dynamics, Academic Press, New York

Pretty, J (1994) ‘Alternative Systems of Inquiry for a Sustainable Agriculture’, IDS Bulletin 25:37–48

Pretty, J and Hine, R (2001) ‘Reducing Food Poverty with Sustainable Agriculture: A Summary of New Evidence; Final Report from the SAFE-World Research Project’, University of Essex, Colchester, UK

PRGA Program (1999) Crossing Perspectives: Farmers and Scientists in Participatory Plant Breeding, CGIAR Program on Participatory Research and Gender Analysis, Cali, Colombia

Probst, K, Hagmann, J, Becker, T and Fernandez, M (2000) ‘Developing a Framework for Participatory Research Approaches in Risk Prone Diverse Environments’, Proceedings, Deutscher Tropentag 2000, University of Hohenheim

Sarin, M (1997) ‘Meeting of the Group on Gender and Equity in Joint Forest Management’, 29 September to 1 October, Report on Proceedings, Society for the Promotion of Wastelands Development, New Delhi, India

Schmink, M (1999) ‘Conceptual Framework for Gender and Community Based Conservation Case Study Series on Gender, Community Participation, and Natural Resource Management’, No. 1, University of Florida, Center for Latin American Studies, Gainesville, Florida

Settle, W H (1997) ‘Science by Farmers and Science by Researchers: Challenges and
Opportunities’, Paper presented at the International Symposium on integrated pest
management in rice-based ecosystems, 20–24 October, Guangzhou, China

van de Fliert, E and Braun, A R (2002) ‘Conceptualizing Integrative, Farmer
Participatory Research for Sustainable Agriculture: From Opportunities to Impact’,
Agriculture and Human Values 19(1):25–38

von Glasersfeld, E (2001) ‘The Radical Constructivist View of Science’, Foundations of Science, special issue on Riegler, A (ed), ‘The Impact of Radical Constructivism on Science’, 6(1–3):31–43

Whitten, M (1996) ‘The Empowerment of Vegetable Farmers in the Philippines Using IPM as an Entry Point: A Four-way Partnership between Farmers, Researchers, Trainers and Industry’, Intercountry Programme for Vegetable IPM in Asia, Makati, Philippines, FAO

Wilde, V and Vainio-Mattila, A (1995) Gender Analysis and Forestry, Forests, Trees, and People Programme International Training Package, FAO, Rome, Italy

Young, K (1988) ‘Gender and Development: A Relational Approach’, in Gender Analysis for Management of Research in Agriculture and Natural Resources, International Service for National Agricultural Research, The Hague, Netherlands, pp101–109

Document(s) 5 of 15