6.1 Methodology for the analysis of LKMS
6.2 Potential applications
6.4 Recommended follow-up activities
The preceding chapters have shown how LKMS was instrumental in the management of natural resources by local people, and how these techniques and systems have, or have not, changed in the past decades. Furthermore, their value for designing and implementing development assistance has been demonstrated through the experiences of several recent and not so recent projects. This chapter addresses the question of what needs to be done to effectively incorporate LKMS into the development process.
Consideration is first given to developing a methodology for collecting and analyzing information on LKMS, followed by a look at some potential applications of LKMS in the realm of social forestry and natural resource management. A third section discusses some of the prerequisites: what needs to be done on a policy and institution-building level before LKMS can effectively be incorporated? And finally, certain gaps in our knowledge of both the content and the value of LKMS are highlighted.
A main reason why LKMS is seen as a means to further popular participation in development activities is that in every step of the way - collection of data, analysis of its significance, incorporation of data into technical and managerial solutions, implementation of activities, etc. - the involvement of the local people is a necessary factor.
Some experiences in the past show how it is possible to be willing to incorporate LKMS into the development process, but still fail due to faulty data collection. For example, in 1921 the District Commissioner of northeastern Uganda wanted to draw county lines according to traditional territorial boundaries. However, by conducting his survey in the wet season, a lot of dry season pastures were declared unoccupied and given to tribes who had no traditional rights in the area. This led to increasing hostility in the area, and contributed to the destruction of traditional organization1.
Data collection on the LKMS of natural resource management cannot be concerned solely with physical matters, but must also look at the socio-economic and institutional aspects. The classic type of socio-economic survey, based on questionnaires administered to a random number of people, is not sufficient by itself and cannot show the logic behind production systems. An integrated survey method must find the specific aspects and constraints facing the decision-maker, and determine which information is actually used to make the decision, and which information is unused but may nevertheless be given to interviewer. Such a survey method requires active participation by the local people and daily observation of the production routine by the researcher².
Data collection on the physical aspects must go beyond land use, botanical, pedological, and other classical surveys, to how the people perceive their environment, and which aspects of the physical resources are filtered through into their decision-making process. For example, many ethnotaxonomies and other cognitive studies of ethnoecology tend to be too exhaustive; one needs to always ask what is the value of this information to the local people's actual behavior and strategies - is the information reflecting an ideal or a norm?³ Environmental perception surveys should consider, among other things, the perceived value of resources for their livestock and other production activities4, the choice of indicators for predicting environmental dynamics, their knowledge of the spatial and temporal variability of resources (and their standard deviations), etc. Environmental cognition surveys, including cognitive maps, are useful for obtaining these types of information, since they show how the people characterize their landscape, and which of its parts they consider important or prominent. Some researchers, mostly in the late 1960's, have already studied spatial cognition in the African context5. The method, as carried out among the Maasai6, obligates the researcher to accompany the herder into the field.
Other types of surveys on LKMS are aimed at collecting data on decision making processes and logic, fundamental paradigms of knowledge (e.g. how the physical world is ordered and related to the social world), means of coping with risk and uncertainty, manual and mental skills, etc.7.
The data collection methodology should be an iterative process that starts with a pilot survey to test the survey method, observes activities related to the survey in order to verify the survey method, modifies the survey method, and finally does the full survey. Four types of survey methods may need to be carried out for the same project: 1) specification surveys, where the highest level of detail is needed on some aspects; usually the most knowledgeable persons are interviewed; 2) formalization surveys, where one makes sure that all knowledge is covered; usually one will need to contact a wider range of people, and to randomize the sample; 3) generalization surveys, where non-specialized information is needed to make sure that data represents a wider population; and 4) verification surveys, where the data is re-collected or compared with existing information, etc.8 (see BOX 6.1).
C.G. Knight (1980)9 has provided a concise listing of survey methods used in the study of LKMS. Specification surveys include informal discussions with the most knowledgeable person(s), participant observations, informal/semi-formal interviews, questionnaires, and teacher-pupil relations where the informant is the teacher.
Formalization surveys include paired comparison (e.g. between ideal and norm, or between traditional and modern techniques, and then asking people to explain the differences10), tree drawing (flow-charts with yes/no dichotomies)11, slip sorting (sorting diagrams of plants, soils or other resources according to what they have in common), eliciting dictionary definitions, triad tests, gaming approaches, and group discussions. Triad tests are used to show and explain differences and similarities between objects, such as plant species, soil types, etc. These tests can become too exhausting if there are too many things to compare, therefore they are often combined with other methods12. For example, triad tests have been used to compare the taxonomy and classification of body parts among Maasai, Kikuyu and Americans, as a way to facilitate cross-cultural communication13. The gaming approach (i.e. lets imagine conversations) are useful in stimulating discussion on some central theme (e.g. droughts, bush fires, etc.). They are valuable because the respondents use their own opinions and logic rather than the interviewer's14. Group discussions are especially useful in the pastoral context. They increase our understanding of the dynamics of communal decision making, since it is often arrived at by a consensus within the group15.
Generalization surveys use formal questionnaires, ranking
tests, and sentence completion exercises. Verification surveys usually involve a
return to the same population to observe and compare behavior at a later date,
to translate and re-translate the information, to extrapolate from the data to
other situations and ask again (gaming approach), and to actually use the data
on a trial run (e.g. technical details of the production routine are actually
tried out in the field).
Apart from an appropriate methodology for collecting information on LKMS, we also need to develop a framework for analysis that highlights local ideas that are obstacles to those change that are necessary, and identifies local ideas that are realistic appraisals of, and realistic solutions to environmental problems17. Because of the high heterogeneity in LKMS (both within and between groups), and the fact that the social and physical environment is rapidly changing, the final test of the value of a local technique or system for development will only be when it is actually used on a trial basis in a pilot project. For example, the viability of a local technique or system, and its value for development, can be evaluated through a process whereby the LKMS and formal scientific outlooks and objectives are compared and combined through discussions between the local people and extension agents, during or after the data collection stage. In addition, these discussions should help identify whether a particular LKMS is still remembered and in use, by all or only a few people.
Only relatively few projects and programs have incorporated LKMS into their designs - some of these have been briefly described in the previous chapter. However, many field researchers and workers have proposed and recommended a diverse portfolio of potential applications for LKMS. These can be grouped under several project-types.
The first, and perhaps the most important type is the use of local knowledge and social structures in development planning. This can either be done through discussions with the target people during the project formulation stage, or through regional, national and even international workshops. In the case of the former, the short period allotted to project formulation by the donors must be extended to a flexible period that allows the local people to fit in the time required for discussions and decision-making with their production and social schedule.
Workshops have been used recently to help local people identify their development constraints and to communicate these to government and donor agencies. For example, a recent regional NGO workshop, entitled Survival in the Drylands held in Mali, was aimed at strengthening cooperation between NGOs in the Sahel and drawing up an action plan to combat land degradation and to assess the impact of local and national dryland policies18. Workshops have also been organized for local people without the existence of an umbrella NGO. For example, a UNEP workshop held in May 1986 brought together pastoralists from four different tribes in Kenya for discussions across ethnic lines as well as across local/scientific lines. A follow-up workshop and field visit was later organized, this time including two additional ethnic groups19.
A second type of applied LKMS is aimed at the education system. For example, inclusion of the history of the tribe into formal history lessons would help the student establish his people's role and relationships to the outside world. In addition, local concepts of measurement and mathematics (see section 18.104.22.168) can be used to teach modern math20.
Many authors have proposed the use of local social structures (see section 2.2.2) as the basis for new organizational institutions. In the case of pastoral associations (herders' associations, group ranches, etc.), most researchers suggest that the relevant socio-political level should be the herding unit rather than higher (larger) associations21, because the herding unit is usually a tighter cooperating unit, can be identified more precisely with a given land area, and is easier to manage because of its smaller size. However, the problem of higher fluidity remains: the members of a herding unit do not necessarily remain the same from year to year, whereas membership in higher social levels, such as a clan or tribe, is more constant.
Many technical details of LKMS appear to be still viable and can be used to develop locally appropriate solutions to environmental problems. For example, the knowledge of trees and browse, and the traditional protection of spontaneously regenerated seedlings (see section 2.2.3) can be used in reforestation projects, such as those proposed for the Pokot and Turkana22. The traditional range management system, such as rotation schedules and deferment periods (see section 2.2.2), can be modified with Western range techniques to develop a system more appropriate to current rangeland shortages and droughts. In addition, analysis of the traditional range system - for example, where it lies on the continuum between year-round, continuous grazing and rotation grazing - can direct the design of new systems and shorten the period needed for research and trials. The recruitment of herders as primary level veterinarians in various projects has also shown one venue for the use of local veterinary knowledge23.
Another area where LKMS can be useful is in environmental monitoring programs. One aim of most of these programs is to identify when and how the local people have to change their resource exploitation strategies to avoid or minimize environmental damage. Using local methods of monitoring (see section 22.214.171.124), alongside Western ones, would facilitate the communication of the results with the people, and help convince them of the need for change. In addition, local people could assist researchers in conducting the monitoring exercise. Local observers could record vegetation changes as part of an early warning systems for environmental degradation, locusts and pests, and farmers could report crop yields for census and monitor the results of on-farm experiments24. Herders could report on disease epidemics, help police the boundaries of group ranches against outsiders, and in general help ensure compliance with range management plans and controls. Although local observers need to be backstopped by extension agents, they could allow a saving on the project's manpower and logistical support especially in remote pastoral areas.
The local knowledge of environmental dynamics (see section 2.1.2), in conjunction with that of formal science, can be used as short cuts for resource inventories, such as the distribution of soil and vegetation types25. Local concepts of measurement can be used by local observers to conduct some forms of research for formal scientists, especially if the standard deviation of traditional units can be ascertained beforehand. Areas that are still being managed well by the traditional system, could be delimited as research stations in order to recognize the value of traditional knowledge, and prevent its untimely demise26. These research stations can then be used to experiment with modifying and improving the traditional system and extrapolating the lessons to other areas.
Paying lip service to the need to incorporate LKMS into development designs can be just as bad as paying lip service to popular participation. Too many projects have tacked on a research on LKMS phase as an after thought, resulting in volumes of interesting but too exhaustive and inappropriate research reports, which are then filed and not used by project designers and implementors. LKMS needs to be incorporated effectively into the development process, but to do so would require certain preliminary changes in the process.
For example, project planning should be accompanied with a flexible timetable for discussions with local people, pre-formulation action research designed to identify promising and viable LKMS, and a joint (local and outsider) project formulation process.
Another problem is that Western experts and Western-trained Africans are trained not to see LKMS as potentially useful. The general attitude is that formal science is superior to LKMS, thus there is very little political will to incorporate LKMS into the development process27. Many government officials, accustomed to a paternalistic one-way mode of communication, become impatient at the prospect of using popular participation and LKMS28. Extensive discussions at the level of policy-makers (government officials and donor representatives) and national and international scientists will need to highlight previous successes with LKMS. At the level of extension agents, similar discussions need to be accompanied with a change in extension training programs, such as incorporating LKMS subjects into term papers, theses, etc., and training on an appropriate methodology for data collection29. The local people too will need to change their attitude, especially those groups who have seen too many failed projects and have come to expect only top-down interaction with governments and donor agencies.
Certain government policies that affect arid lands and pastoralists in particular also need to be changed. The most notable are land tenure laws, the absence of land use planning and enforcement, economic incentives that pull young herders off the range, urban-oriented education systems, etc. In addition, strict environmental impact studies should be required of all projects by the Governments. These studies should make a point of going beyond physical aspects to impacts on the social and production systems, and in particular, their underlying LKMS30.
Apart from elucidating the many different types of LKMS that may be of eventual use to the development process, this paper has also served to point out some of the gaps in our understanding of the LKMS of natural resource management in arid and semi-arid Africa. Several follow-up actions can be suggested.
Certain topics within LKMS seem to be more neglected than others. For example, the descriptive knowledge of how and when natural resources are harvested, stored and processed is less well known than what the resources are used for. Furthermore, our understanding of how climatic changes and other ecological dynamics are predicted is far less than how the different resources are labelled and classified. In other words, more research is needed to highlight those aspects of LKMS directly useful for the management of natural resources.
Daily management practices, as a whole, also tend to be less well known than descriptive knowledge and organizational structures. Studies of range management practices need to go beyond simple descriptions of seasonal movements, to details on the frequency of movements, duration of sojourns, and other rotational practices. Traditional range improvement techniques also need to be identified in order to find a local base on which to build new improvement and revegetation techniques. The internal organization of herders and herding units is another neglected topic, without which it would be difficult if not impossible to revive traditional techniques or to introduce new ones. In particular, the role of women herders needs to be clarified.
Our knowledge of tree and shrub management among pastoralists is much weaker than among farmers in more humid areas. More details are needed on how they harvest browse, tannin and other products (cutting techniques, frequency of cuts and duration of rest, quantities harvested, etc.), and how they protect and regenerate trees and shrubs.
We are gradually becoming aware of the tremendous diversity in land and resource tenure types across Africa. There are many different types of communal resource tenure, including reserved areas and sacred groves, each of which have different implications for resource management. To illustrate, take the case of the Turkana in Kenya. It was only after about 20 years of detailed research, that the existence of private, household-owned trees and sections of riverbeds came to light. The fault was not in the research per se, but the fact that the earlier work had been done almost exclusively among those Turkana living in more humid areas in the south, while the private tree tenure was found among the northern Turkana living in more arid areas31. Thus, although quite a bit may appear to have been done on resource tenure so far, more studies need to be commissioned to unearth inter-tribal and intra-tribal variations.
Pure, exhaustive research on LKMS may be of academic interest, but would be a waste of development finances. As far as possible action research, ie. the identification of potentially useful LKMS and, when appropriate, the incorporation of the LKMS into trials and experiments, should be the preferred type of research. A compendium of potentially useful LKMS should be accompanied by an evaluation of their value for development. To do so would require a framework or guidelines for analyzing their viability and how they could be modified or revived. While the development of such a framework will be facilitated as more and more projects experiment with LKMS, a workshop bringing together development workers involved with LKMS will help to pave the way for it.
A few case studies can be suggested. Certain groups and countries of arid and semi-arid Africa appear to be under-represented in the English and French literature. More work needs to be focused on them, either in the form of field research, or the translation of literature written in other languages. Some examples are Angola and early Italian work done in Somalia, Ethiopia and Libya. Other groups and countries have particularly interesting forms of LKMS that have direct impact on development, or may be useful for extrapolation to other neighbouring groups. A few examples are, the intricate water management structure of the Borana of southern Ethiopia, the role of women herders among the Somali, the emergence of private range enclosures in Sudan and Somalia, the rotational range use of the Wodaabe Fulani, the current viability of the range controls devised by the Macina Fulani, differences in herding style between the Twareg and the Fulani, the range tenure arrangements of the Tonga of Zambia, and the Kgotla meetings of the people of Botswana. Finally, particular attention can be paid to those groups that have traditional range reserves and sacred groves, such as in Zambia, Kenya, Botswana, Mali, Burkina Faso and Morocco.
1. Baker 1975, p. 192.
2. Richards 1980, p. 185; Gladwin 1980, pp. 9-11; Warren & Meehan 1980, p. 327.
3. Johnson 1980, p. 49.
4. Stiles & Kassam 1986, p. 14.
5. Knight 1980, p. 222.
6. Western & Dunne 1979, p. 95.
7. Knight 1980, p. 221.
8. Knight 1980, p. 226.
9. Knight 1980, p. 226.
10. Johnson 1980, p. 63.
11. Gladwin 1980, p. 25.
12. Richards 1980, p. 189.
13. Burton and Kirk 1980, pp. 269-302.
14. Richards 1980, p. 186.
15. McDermott & Ngor 1983, p. 5.
16. Knight 1974a, p. 62; Richards 1980, p. 190.
17. Richards 1975, p. 109.
18. Survival in the Drylands: Regional NGO Workshop, Segou, Mali, September 14-17 1988, organized by International Institute for Environment and Development (IIED), London and UNEP.
19. UNEP 1986a, and UNEP 1986b.
20. Fink 1980, p. 261.
21. Bourgeot 1981, p. 178; Barrow 1988, p. 15.
22. Barrow 1988, p. 21.
23. McCorkle 1986, p. 137.
24. Warren & Meehan 1980, p. 327; Howes 1980, p. 339; Richards 1980, p. 193.
25. Aubert & Newsky 1949, p. 109; Howes 1980, p. 339.
26. Brownrigg 1985, p. 41.
27. Warren & Meehan 1980, p. 332; Howes 1980, p. 339; Brokensha et al 1980, p. 5.
28. Brokensha & Riley 1980b, p. 266.
29. Brokensha et al 1980, p. 7.
30. Brokensha & Riley 1980b, p. 268.
31. Dr. James Ellis, Natural Resources Ecology Laboratory, Colorado State University, 1989, pers. comm.