The failure of both formal research and indigenous knowledge systems to fully address the problems facing the sector necessitates a search for alternatives. The evolution of more collaborative approaches to formal research over recent years implies that they offer benefits over non-participatory formal research, and also that the methods used by the formal and participatory approaches are not incompatible.
Whilst the rhetoric of participation may be more developed than the reality, this "rhetoric opens doors,
makes spaces, and provides points of leverage" (Chambers, 1998: 111). Approaches and methods for
allowing and enabling more balanced participation in research have evolved in a variety of sectors
providing a valuable springboard which the fisheries and aquatic resources sector can profitably use. In
addition the emphasis on participation in the policies and approaches in many of the donor agencies
means that the resources to achieve greater participation in research are, at least theoretically, available.
The potential then is considerable. As Chambers (1998:113) says: "Potentials are not just for local level
participation, but for changes at three levels: policy, institutional and personal". This potential must be
converted into some practical reality of "how to do it" and it must balance the issues, which participation
in research creates, with the benefits.
6.1 The Potential Benefits of More Balanced Participation
The moves to encourage and support a greater involvement of fishers in the professional research participants' research activities (Type C and D) have had positive benefits for both groups of participants.
The shift in emphasis in development towards people-centred approaches means that the knowledge and perceptions of those people must, at the very least, be considered in that process. Participation by fishers can also reduce the cost of collecting the data and may increase its accuracy. Where future action resulting from the research will include those people or influence their lives then such research will be much more effective if it takes into account the social, cultural and technical knowledge system which structures their lives. Opportunities for fishers to validate the research results will reduce the likelihood of inappropriate research results entering the policy process and generating poor development results at a later stage. It could also lead to more acceptable uses of the research findings in terms of the institutional, management and legislative changes which policy-makers adopt as a result of the research. Without such validity, acceptance and compliance can become critical issues. In fact there are many complaints of the under-utilisation of potentially useful research (Lammerink and Wolffers, 1994).Empowerment of fishers through this type of collaboration can occur but the research remains outside of their control and, because empowerment is not the motive, it occurs rather slowly.
Research where fishers encourage and support the involvement of professional research participants in their research agenda (Type F and G) also has similar potential functional benefits and possibly more empowerment benefits.
A more balanced relationship is one which starts from a collegial position (Type E) with the two groups working in a partnership of greater equality. From the perspective of this more balanced approach to participation in research, the functional benefits are that it is likely under many circumstances to also increase the efficiency and effectiveness of research. The efficiency; from governments' perspectives, may be increased by collaborative and collegial relationships which tap into the vast store of indigenous knowledge and accessing this relatively quickly and cheaply. From the fishers' perspective, a more balanced relationship may lead to access to more reliable and faster methods of research, and a wider perspective of problems and possible solutions.
In terms of effectiveness it is likely that research results generated through a more balanced relationship will be more valid in terms of the local socio-economic and institutional situation. Policies and their implementation resulting from such research are thus likely to be more specific to, and more acceptable to, the communities concerned. This in turn is likely to lead to faster and wider uptake and impact. This is particularly important for marginalised groups such as women and the poor who may not have the capacity to respond quickly to opportunities created through conventional research processes and thus loose out to more advantaged stakeholders. This will be particularly important in areas such as resource management where the distribution of benefits of management measures needs to be considered.
The combination of formal scientific research with community-based research approaches may provide additional benefits: the reliability of more formal scientific methods may complement, and be complemented by, the greater socio-cultural validity of participatory research methods. This in turn should produce results which are more acceptable to policy-makers and are more workable in the social and economic contexts of the communities concerned. Scientific research and fishers knowledge system working side by side on similar problems can also feed into each other and learn from each other.
Collegial research is also likely to raise the level of awareness of fishers and to .provide them with the knowledge, structures and processes which may allow them to take more control over their lives. It may also empower professional research participants by making their research results more effective in informing policy and thus make research funding seem a more appropriate way to spend scarce government money.
The incorporation of greater participation into fisheries research is not without its problems. Some of these are outlined below.
6.2.1 The Characteristics of the Sector
Much of the participatory research that has been carried out to date has been in the agriculture sector on farms where the resource is static, visible, owned and largely controllable. Capture fisheries, however, deal with mobile resources, which are generally invisible, often from a common pool and largely beyond the control of the researcher. Thus the opportunities to transfer research methods across from agriculture are fewer. As Kurien (1998: 32) says: “ Care needs to be taken to ensure that the data and information collection in fisheries is not modelled after the methods and formats used for the agriculture sector”. In aquaculture and fish processing the operating environment shares similar qualities with agriculture and the methods are more easily transferred. It is in these areas that participatory experimental research has been most common. This is not to say that participatory research in capture fisheries cannot be advanced. However, it does mean that greater emphasis needs to be placed on evolving and testing suitable methods.
6.2.2 Institutional resistance
As discussed above there are major philosophical divides between different schools of research. These not only concern differences of world views but also of what constitutes research and science. Without changes to these attitudes and a genuine desire to work together and share ideas and experiences, collaboration will not be possible.
If fisheries is to become more people-centred then its research will need to involve a greater focus on social as well as natural science methods. This will inevitably involve more inter-disciplinary and multidisciplinary approaches which may need to originate from a diversity of institutions. There is likely to be resistance to both a refocusing of methods and a more collaborative interdisciplinary approach. There will also be genuine difficulties of planning and co-ordinating research across sectoral and institutional boundaries. However, there is already some successful experience of this in fisheries through approaches to coastal area and watershed management.
In some situations this resistance will result in researchers adopting a veneer of participation to satisfy the demands of donors. If these constraints to wider participation in fisheries are to be overcome there is a need to create a more enabling environment in research institutions both in developing and developed countries.
There may be some resistance to participatory research approaches because of perceptions of incompatibility with other development tools such as the logical framework approach and ZOPP. Chambers (1998) refers to these as having a more top-down orientation. However, this is more a reflection of how such methods are currently (in many cases, badly) used. Whilst such methods have structure they need not be any less flexible or empowering than PRA. For example, PRA methods may be used to generate the knowledge which feeds into the logical framework approach.
The cost of participatory research is a concern that is often given for avoiding greater degrees of participation in research. It is by no means clear that in any situation greater degrees of involvement will increase costs. The costs will depend on a diversity of factors which include the form which greater participation takes, the institutional changes required, the need for investment in skills and the time taken to get results. Against any increase in cost must be balanced the potential increases in benefits that will accrue.
The greater involvement of fishers requires changes in the balance of control. This is not always considered to be in the best interests of those currently in control especially where fishers are considered of low social or political status. This resistance may act within the government administration or at a more political level. Greater involvement is not always easy to achieve as fishers generally work at distance from the point where policies are made. In countries where more authoritarian government systems exist, or where local elites fear change which they do not control, participatory research may be very difficult to implement (Lammerink and Wolffers 1994).
At the political level there may also be resistance to the time scales involved in adopting more participatory approaches. Such approaches are initially likely to take longer times to provide tangible benefits but those benefits will be cumulative as they feed into a more effective policy process. Unfortunately governments work in relatively short time frames which may demand faster results than can be provided.
Knowledge and access to information influence power. To professional researchers knowledge may mean publications and the associated status and he/she may be reluctant to share that knowledge through participatory research. Some researchers may feel that participatory research is less reliable or valid than more traditional approaches and feel that their reputations may suffer from becoming involved in such activities. In addition the research scientist may feel threatened by the increasing involvement of villagers in the research process particularly if the villagers progressively identify research questions which are outside the skills of the scientist or her/his institute.
To the community-based research participants, knowledge may define access to fishing resources or to markets. Sharing this knowledge may affect their livelihoods or social status. In some situations villagers may feel that the results of the research may be used by the authorities for purposes which are to their disadvantage (e.g. to assess taxes or to remove subsidies). In addition the basis of much indigenous knowledge is verbal. The very act of committing this knowledge to paper can disempower the community-based researcher (T. Sarch pers. com. 1999).
Politicians or managers who commission research may also wish to restrict distribution of the knowledge generated for a variety of reasons. Some of these reasons may be in the interests of particular stakeholder groups or in the interest of the sustainable use of the resource. Others may be less positive.
Freire (1972) noted a condition of oppressed people which has often proved a major obstacle to the active involvement of many rural stakeholders in research and development. This he referred to as “self depreciation” reflecting a view of themselves as being useless, ignorant and incapable of meaningfully contributing. To actively and positively engage people in research may require a long process of awareness raising and capacity building. There are also community-level institutional barriers to allowing different stakeholder groups from becoming involved (see section 2.2.2). The poor may be also be excluded from the process by social pressure or because they do not have the time or resources to engage (I. Harkes pers. com., 1999).
6.2.6 Limitations of the Approaches and Methods
Different participatory research approaches and methods have evolved to address different elements of the research and development process. In many sectors the recent focus has been more on development than on research and the methods reflect this. A suite of methods have been developed and many of these have originated or been refined through PRA. These have been largely aimed at the diagnostic activities (Farrington, 1998) concerned mainly with generating ecological and economic information but have been less effective for understanding social relationships (Mosse, 1998). Other social research methods such as participant observation may be more productive for such information. In some situations there may be no participatory substitute for the use of more conventional survey equipment such as complex oceanographic equipment or satellite imagery.
Within a particular sector such as fisheries there will be distinct research processes for which participatory methods have not yet been developed, adapted or tested. Work will need to be done to develop and test such methods.
6.2.7 Research Validity and Reliability
A major criticism of participatory research is that it lacks the objectivity of formal scientific research (Stringer, 1996). However, Rahman (1991:15) counters this by saying "the scientific character or objectivity of knowledge rests on its social verifiability, and this depends on consensus as to the method of verification. There exist different epistemological schools (paradigms) with different respective verification systems, and all scientific knowledge in this sense is relative to the paradigm to which it belongs and, specifically, to the verification system to which it is submitted". This again relates back to the concerns about how objective the world can be seen to be. Arnst furthers this by saying (1996:113) "The largely unquestioned assumption that scientific knowledge is more valid or valuable then other knowledge is erroneous. Traditional or indigenous knowledge is simply different knowledge formulated in response to differing environments, conditions and cultures" (italics in the original). The debate around the philosophical approaches to research continues but participatory approaches are now well accepted within at least some areas of applied social research (see Holland and Blackburn, 1998, for a discussion of this). Certainly the validity and reliability of participatory research has been a significant question in the minds of many researchers (Stringer, 1996; Pretty et al, 1995).
An important issue in maintaining the quality of research concerns who should be involved in research. As Lammerink and Wolffers ask: "Are only data that are collected by people who have been trained in a university valuable?" (1994:81).
Clearly the issue is not who or where they are trained but that the quality of the data collection is high and that it is shown to be high. Pretty et al. (1995) tackle the question of trustworthiness of research findings using participatory approaches and develop a framework for judging it (see box 7).
In this report the processes that generate indigenous knowledge have been broadly referred to as research to allow the opportunity for their ability to generate valid and reliable data to be assessed. In the framework in box 6, indigenous knowledge is usually generated by Type I participation.
However, one difficulty that has been raised by some supporters of participatory research is that training community-based researchers reduces their individualistic approach to research and makes them more like the professional scientist. As a result some of the benefits of collaboration may be lost.
Box 7. A Framework For Judging Trustworthiness How far the following criteria were used to generate information through participatory methods the trustworthiness of the information.
From Pretty et al., 1995 |
There is an element of the research community which views fishers in a universally benevolent way considering their knowledge to be almost infallible This is clearly not the case. As mentioned above, indigenous knowledge is not always the best knowledge. It can often be unreliable, dated, limited, and biased. Such complaints can also be levelled at some scientific research. Research of any type needs to be considered carefully to ensure validity and reliability, and new approaches need to be developed to deal with the interface between indigenous knowledge and scientific knowledge and to address the need to generalise research findings for policy purposes.
There are many reasons why indigenous knowledge may be inaccurate. Some of these may be because the fishers wish to represent the information so. In others it may be a genuine mistake of interpretation.
There are also concerns that the knowledge systems of fishers and those of professional research participants will be incompatible. That the differences in worldview will be too large to overcome. There is clearly a lot of work to be done in understanding these differences if collegial approaches are to form a major part of future research.
Science is dependent on the honesty of the researchers involved (Kesteven, 1973). A high degree of responsibility is placed on the scientist to ensure that he/she reports his/her findings accurately (Resnik, 1998). This forms part of a wider code of scientific ethic that is largely unwritten (Knael, 1994). Such a code is. often difficult for even those formally trained in the scientific method to grasp and incorporate fully into their work. When research moves out of the controlled environment of the research station into the village, concerns may be raised about the degree to which the code is adhered to.
6.3 Towards Greater Participation in the Research Process
If we are to adopt a more balanced approach to participation in research in fisheries, throughout the research cycle, and which empowers as well as improves research efficiency then we cannot do so merely by selecting and using a set of participatory research methods from the development tool box. Likewise we cannot merely transfer approaches from other sectors such as agriculture. From the discussion so far it can be seen that participation is not a "quick-fix" solution. More collaborative approaches are, as the name suggests, approaches. These involve changes in attitude, structures, processes, skills, knowledge as well as the methods used. These changes need to occur at all levels from international agencies through different levels of government, to research institutes, extension agencies and to the fishers themselves. They also need to occur at all stage of the research cycle as suggested in Box 8.
Although these changes are significant and substantial it is not necessary for them all to occur before better levels of participation can be achieved. Change can occur incrementally and even relatively simple processes such as relaying research findings back to fishers and talking with fishers sympathetically can provide a valuable starting point (R. Johannes pers.com. 1999).
What these changes are and how they might be achieved for each stage of the research is discussed below.
There is a need, amongst all participants, to raise awareness about each other's knowledge systems, knowledge bases, knowledge generating approaches and methods, knowledge validating systems and knowledge communication systems. There is also a need to raise awareness about the potential role of each participant group in each other's knowledge systems and of the potential benefits of collaboration. Within the research process the valuation of different knowledge systems will have to change so that there is greater parity in the way these systems are viewed and treated.
There is a perception amongst some professional research participants that villagers are unable to understand the research process and therefore unable to meaningfully contribute to it. In reality this attitude is more likely to reflect a difference of perceptions between the professional and village participants about what is knowledge and how it is generated. An improved interface between the knowledge systems of the professional research participants and the community-based research participants is crucial to the success of participatory research. This interface should allow the two systems to communicate and for knowledge and understanding to be transferred. Without a mutual respect for, and appreciation of, each others system the partnership is unlikely to be a fruitful one. The professional research participants has a particular role to play in establishing this interface. They are more likely to have a greater capacity to understand such interfaces because of the diversity of social and cultural experiences to which they have been exposed. It is therefore incumbent on them to ensure that an adequate interface is established which caters for the capacities of the community-based research participants and allows an equitable exchange.
There is potential for linkages with other approaches here, such as PRA and PAR, to enable the awareness raising process and to achieve a shift in the power relationships between the different participants.
6.3.2 Change of institutional structures and processes
The scientific research process will need to take a much more interdisciplinary/multidisciplinary approach to research which addresses the social science issues on a similar level to the technical and natural science ones. It will also need to view its area of research much more holistically reflecting the structure of fishers' lives. There will also be a need for research institutes and agencies to adapt to more collaborative approaches with other institutions to allow interdisciplinary teams to be established and for joint research programmes to be developed. In many research projects, there will be a need for different types of agencies, such as research institutes, government extension staff, NGOs, projects and community-based organisations and associations, to work together. This in turn will require these different organisations to adapt their planning and working conditions to accommodate those of other organisations.
It will also be necessary for researchers to undergo retraining to be able to work with other disciplines, to adopt other approaches and to use new research methods. The ways in which research is assessed and graded for funding and promotion purposes will also need to change to ensure that collaborative, interdisciplinary and interagency work is fully appreciated within the different organisational systems.
6.3.3 Defining the research need
If research is to be more collaborative it must to be driven by the needs of all those involved. To do so it must take into account the needs of the fishers and their capacity to both participate in the research and to respond to the likely outputs of the research. It also needs to involve the diversity of stakeholders in the fishing communities and not treat fishers as a homogeneous group. The research also needs to take into account the wider needs of society as reflected in the demands of policy at the national, local and international levels.
It must also take account of the needs and aspirations of the professional researchers. They too have wishes which have to do with their preferred research areas, their career structures and their social lives. These elements are often forgotten in the process to achieve participation.
The resultant researchable constraint which the .research will address should be clearly defined and agreed by all concerned.
6.3.4 Defining the methodology
It has been said that: “It should be obvious that the actual choice of the type, level and intensity of farmer participation in agricultural research is determined, in practice, by many factors” (Okali et.al., 1994: 23). This also applies to the fisheries and aquatic resources sector and to be over-prescriptive would be to constrain what is an area of considerable potential. The approach adopted must be influenced mainly by the desired outcome of the research process itself. It will also be influenced by the capacity and willingness of the community-based research participants to participate. The. specific biophysical setting of the research will also be important (Farrington, 1998).
Professional research participants in the natural sciences, and those in the social sciences have traditionally had different approaches to research methodologies. In its most practical form these methodologies often differ in the research methods used: qualitative and quantitative. More fundamentally there are philosophical differences about reality and objectivity. There is a need to resolve these issues and find common ground to move forward. Linkages with other approaches, such as PRA and PAR, will be important for exploring these different systems. Some researchers are advocating an amalgam of approaches not dissimilar to the convergence of participatory approaches discussed in section 1.1.3.
There is also a need for professional researchers to work with community-based researchers to identify an effective interface between formal science methodologies and those used by the fishers. There will be a need to ensure that the data generated complies with the validity and reliability criteria of the systems concerned.
There will be a need to find some common ground on the motives for the research. NGOs may be more concerned with the empowerment aspects of the research, whilst government institutions may be driven by the hope of generating data to contribute to policy in more efficient and effective ways. Fishers may have mixed wishes. They may primarily wish to make enough income to survive, secondly to inform a change of policy for medium-term improvement of their position and thirdly to strengthen their own position within the system.
Issues of the ethics of the research will need to be carefully considered and discussed. Different participants may approach the research from different ethical frameworks and these need to be understood and accommodated in the research process. The confidentiality of data is an important issue. Data provided during the course of the research by one group of fishers or individuals may be permitted to be accessed by outsiders but not by other community members. In some cases information on the location of fish stocks may be treasured secrets, information on wealth may likewise be considered as confidential.
Taking a process approach to research will allow a greater degree of flexibility, relationships to develop, awareness to grow, confidence in each others' abilities to be confirmed, and for later stages, such as assigning roles and identifying methods, to evolve gradually. If participatory research is to involve the poor and vulnerable it must adapt to their specific needs. For example the small-scale experimentation processes proposed by Bunch (1989) for agriculture could be adapted to suit marginalised groups in the fisheries sector. Instead of large-scale trials, which of necessity are high risk, poor fishermen and women should be encouraged to experiment with part of their time, catch, product etc.
6.3.5 Assigning roles and responsibilities
Assigning roles and responsibilities will be a difficult task which will need to take into account skills, attitudes, access to resources, and the priorities of the different participants. It will also need to be done with due consideration of the tine frame involved in the research and the time frames of the different participants. Fishers may work at different rates than professional research participants, they may have more free time at different times of the year than the formal researchers and they may like to participate at times of the day when formal researchers may not.
6.3.6 Identification of methods
The methods chosen for the research should be mainly driven by the requirements of the researchable constraint that the research is to address and by the desired research output. It will also be influenced by the skills of the researchers.
There are many participatory methods which allow collaboration of different types (see Annex A). These have mainly been developed in other sectors to allow villagers to participate in professional research programmes in a relationship of Type C and D. In many situations these will need to be adapted to address the kind of situations which confront fishers. There is a need to adapt more formal scientific research methods to allow the participation of fishers in them. In other cases new methods will need to be developed (for example DFID is funding research into more participatory methods for assessing fish loss in the processing chain). Some success has already been achieved in areas of data collection and recall surveys (S. Garcia pers. com., 1999).
In many situations both groups of researchers will need to be trained in new research methods which accommodate greater participation. This should be done with due consideration for the fact that most participatory research methods have been developed within the world view of the professional research participants and will not necessarily translate well for use by community-based research participants. Fishers also have methods for collecting data. These will need to be understood by professional research participants and, where appropriate, incorporated into the research methodology.
It is likely that a multiplicity of research methods will be used in order to allow different participants to triangulate the research results thus allowing validity and reliability tests to be applied across knowledge systems. This may increase workloads in the initial research relationships but will reduce as confidence in each other's approaches and methods increases.
The research implementation will require co-ordination to allow the different elements to work well. There will be a need for mechanisms to facilitate that co-ordination without a transfer of power which creates an unbalanced situation. Professional research participants are, as the name suggests professionals whose income and status derives from the research they do. Community-based research participants, however, derive their income and status from other activities which research is a small part of. They may also be much more risk sensitive and risk averse than professional research participants. It will be necessary for the implementation process to accommodate these different perspectives. It is not always desirable (from either the researchers' or the community's perspectives) to maximise the level of collaboration, some situations of research work may demand a low level of participation (I. Harkes, pers.com., 1999) especially where the potential participants have a vested interest in seeing a specific outcome from the research.
6.3.8 Analysis of research results
The analysis of data will be an important collaborative process and one where many of the benefits of collaboration will become most apparent. There are, however, major obstacles to harmonising analysis methods from formal and community-based approaches. Many of the approaches to data analysis from formal science are very systematic and involve precise mathematical or procedural stages. Communitybased approaches are likely to be more intuitive or iterative, taking place over a longer period of time.
Finding a common interface between the approaches to analysing, interpreting and explaining what the research has generated will be a major challenge. It may be necessary for different groups to use different methods of analysis but there should be much dialogue to allow a progressive evolution of some consensus as to what the research has generated and what it means. This process will require a major shift in the attitude which many researchers currently have concerning the ability and knowledge of fishers and may require some adjustment of thinking to accept that other ways of looking at the world exist.
6.3.9 Dissemination of the research results
As mentioned above, the data acquired through collaborative research is likely to make use of data which is confidential and to give rise to descriptions and explanations which may have important policy implications. The needs of all the stakeholders must be considered carefully to assess what information should be made available, to which audience, and in what format. This should also be considered in the light of the changes of behaviour which the research aims to achieve in different audiences.
Collaboration in this process can have real benefits in deciding the structure and format of dissemination media. Literacy levels, attitudes to colour, pictures and symbols all need to be considered carefully within cultural contexts.
6.3.10 Application of the knowledge in the policy process
The collaborative generation of new knowledge by itself is important, but many of the real benefits of this process come when that knowledge enters the policy system. If collaboration stops following the dissemination of research results then these benefits may be lost. At this level collaboration can enable the realities and priorities of poor and marginalised people to be expressed and communicated to policymakers (Chambers, 1998).
Fishers have an important role to play in the policy process particularly in contributing to the way the new knowledge is valued in the policy process, and is used in formulating policies and the plans for policy implementation.
Participatory research at this stage links into other approaches such as PIP and becomes an important tool in changing the overall policy framework.
6.3.11 Implementation of policy
Likewise the fishers need to be more actively involved in the implementation of policy and in deciding how the new knowledge is transformed into change which will affect their lives. Fishers often have important adaptive and coping strategies to deal with the problems which they face on a day to day basis. It is important that the policy implementation meshes well with these strategies. The development of local level implementation strategies again links the participatory research process to other approaches such as PRA and PAR.
6.3.12 Evaluation of policy implementation
It is important that the collaborative approaches extend into the assessment of the effectiveness and impact of those policies on the people concerned. Linkages with approaches such as PAME will be important for this.
6.3.13 Research agenda setting and prioritisation
The results of collaborative research need to feed into the research agenda setting processes. As Chambers (1995:42) says: "...determination of priorities in ...fisheries and other natural resources research should be much more by and through the analysis and experience of local people, weighted to give voice to women, weak and poor people". The priorities of the different stakeholders need to be considered. There should also be a involvement of the fishers in deciding what the research agenda should be. Without this level of collaboration then community-based research participants would have to participate in someone else's research programme. This is unlikely to lead to a sustainable collaboration.
The issues addressed by participatory research are likely to be very relevant at the local level and, due to the social and economic complexities of communities, to become less relevant as the results move out of their specific context. To maximise the effectiveness of fishers' involvement in research agenda setting it may be particularly appropriate to focus their inputs primarily into local-level research planning processes. At higher levels of agenda setting the involvement of fisher representative groups may be more appropriate.
Attempts to increase participation in decision making have been studied in other sectors and can be considered under three broad categories (Prior et al 1995:138):
These categories need to be explored within the fisheries context to identify appropriate agenda setting and prioritisation mechanisms.
This brief review of the practice and potential of greater participation of all key stakeholders in fisheries research illustrates some important points:
These conclusions suggest that participatory research in fisheries has a very significant and positive contribution to make to the lives of fishers, to the research process itself, and to achievement of sustainable and equitable policy objectives. However, the process is at an early stage of its development and there is much to be done if it is to be mainstreamed as an effective approach.
To assist that process the following recommendations are made:
