Previous Page Table Of ContentsNext Page

Chapter 6

Farmer Experimentation: Building Knowledge

One of the premises of this case study is that experimentation is a knowledge-building process. It is not a process aimed at designing a specific agroforestry system. When an agroforestry system is created as the result of experimentation, it does not mean the process is completed. As stated in the introduction, agroforestry systems are dynamic and adapt to ecological, economic and social changes. This chapter looks at farmers' experimental or knowledge-building activities, in terms of how farmers themselves define knowledge.

Previous chapters described how the combined efforts of farmers and outsiders have led to enormous changes in the landscape and created complex agroforestry systems. Farmers have evolved from being primarily exploiters of natural vegetation to active tree planters and managers. Simultaneously, their knowledge has expanded from knowing only the uses of trees to knowing about tree planting and management practices.

The second part of this chapter looks at farmer-based innovations in agroforestry related to these changes. Farmers use experimental methods resembling Western methods for food crops (Richards, 1986 and Box, 1988). But is this also the case for trees, which can take years to mature? Tree cultivation is so interwoven with agricultural production and daily activities that during the guided farm tours and ranking game discussions, consultants could not recall if they had used a trial period before deciding to adopt a new species. This may have been because farmers had been obliged to adopt many agricultural technologies (i.e. anti-erosion ditches, the planting of Grevillea alongside these ditches, Eucalyptus woodlots and cassava cultivation).

Three processes can be identified through which farmers acquire agroforestry knowledge:

There are a number of inter-related activities associated with these processes including decision-making about species choice, farm and field location and planting methods; implementation of a species trial; and evaluation and determination of suitability, usefulness and benefits.

The biographical information about the 44 tree experts was gathered during repeat visits which offered the opportunity to learn how tree cultivation and knowledgebuilding activities fit into the farmers' lives and which experimental methods they had used when trying new species or tree management practices. However, the case studies did not yield sufficient information about the specifics of farmers' experimentation with trees, so it was decided to set up a species trial in collaboration with the ICRAF/ISAR programme in Rwanda as part of the field research. The ICRAF/ISAR programme provided seedlings or seeds of 15 tree species; some were already widely cultivated in the research areas, and some had not yet been tested on-farm and were unknown to farmers. ICRAF also provided short descriptions in Kinyarwanda of each species and consultants were asked to chose the species they would like to try on their farm on the basis of these descriptions. During the focus group meetings, several questions were asked dealing with consultants' choices, risk associated with testing, experimental methods they intended to use, and evaluation criteria.

What is knowledge and how is it obtained?

Discussions identified four factors that have to be present for a person to be recognized as knowledgeable: intelligence, experience, communication and putting information in practice. Being knowledgeable about something means knowing its use, holding its secrets and showing a sustained interest in it. Some considered knowledge to be "natural" (i.e. a gift of God)2, some stated that it comes from experience, while others felt it comes from being well-informed and having gone to school (although they considered knowledge learned in school "artificial knowledge"). But, whatever the source, a person must use the information to be recognized as knowledgeable. For example in the case of medicine, a person will be recognized as a healer if he or she is capable of curing diseases. Thus, the knowledge of a healer is revealed through his or her actions as a healer. Or, a person who has planted many species of trees and then takes good care of them will be recognized as being knowledgeable about trees.

Individual versus collective knowledge: People in the meetings emphasized the differences in their knowledge. In the words of one participant:

While some collective knowledge is necessary for any society to function, the person who provided the above information probably referred to agroforestry knowledge in particular. In addition to meetings, another participant stated, "In general, collective knowledge is acquired in the classroom. Beyond that, each of us possesses skills we can exploit in our own individual ways." Based on this premise, there does not appear to be any collective agroforestry or tree knowledge, because such knowledge is not acquired in school.

Why does agroforestry knowledge differ?

Many explanations were given for the different levels of agroforestry knowledge, many of which are beyond the control of the farmers.

    ..It starts at birth because God does not equip everyone with equal intellectual abilities ... There is a difference in the knowledge passed on by ancestors... Whether someone becomes a carpenter, blacksmith, healer of cow diseases or a farmer depends on education and training provided by parents ... To be recognized as highly knowledgeable or as a specialist depends on how the initial information is used... People differ in character, interests and taste ... Some are more serious in their tree planting activities than others ... Some have a feeling for experimentation and research and are constantly tinkering in search of new knowledge and technologies, others are satisfied with what they have and know ... Economic circumstances affect knowledge ...A large farm offers more opportunity to experiment and acquire knowledge.

There is also the problem that there is not a regular, organized way for individual exchanges of information to take place, as this conversation between a female healer and the researcher demonstrates.

    Q: Do you have a collective knowledge of trees, in other words, is your knowledge of trees similar to that of other people?

    A: No, it is different. N. treats ifumbi from which his mother suffers. He has his own knowledge about the treatment of this disease and I have mine. But, now we have exchanged our knowledge and we have harmonized our treatment of ifumbi.

    Q: But can you confirm to us: Do you know whether other healers treating ifumbi have the same knowledge that you have about ifumbi and its treatment?

    A: No! We do not know what they know, as they also do not know what we (i.e. N. and I) know.

Agroforestry knowledge and gender

During the ranking game, an equal number of men and women were asked to rank the other farmers on their colline. The resulting scores of couples were combined because many informants stated that if wives are knowledgeable about trees, their husbands must also be knowledgeable. This may have been because of the traditional belief that women cannot admit they have more knowledge than their husbands (Bucyobukiro, personal communication). In addition, since knowledge was defined primarily as having species diversity on the farm which could only be possible through the efforts of the husband, he must also be knowledgeable about trees. Ranking game results also indicated that the nature of trees and agroforestry knowledge differs between men and women. To confirm this and to determine whether farmers consider men or women more knowledgeable about trees, a deliberately-ambiguous question was asked during the community interviews: "If tree knowledge differs between men and women (men know how to plant and manage trees, women can identify species and know uses of trees), who has the most knowledge of trees? Who are the real experts?". The subsequent discussions on this question were the most animated and contentious in each of the thirteen community interviews. Table 10 provides an overview of the opinions expressed and the response frequency.

In spite of the fact that 286 men attended these interviews compared to 156 women, the majority said that women are more knowledgeable about trees. They also felt that men's and women's knowledge is not the same.

Men's knowledge relates to big trees (Eucalyptus, Grevillea, Markhamia, etc. ), tree planting and management, the introduction of new species and/or varieties of trees, and the decision-making aspects of species choice, place, planting times and harvesting (i.e. cutting of the whole tree). Women's knowledge is confined to species identification, naming of species, and the use and harvesting of tree products such as yeasts for brewing beer, seasoning, medicine, fuelwood and love potions. As knowledge of the use of trees, particularly for medicinal purposes, is highly valued and as women are especially skilled in human healing practices, women were considered more knowledgeable about trees than men. However, there were men with extensive knowledge of healing plants and practices and there were women who knew nothing about them. Knowledge depends on intelligence obtained at birth, education, training by parents and interest in discovering and learning new things. Since these factors vary with individuals, regardless of gender, there are knowledgeable and non knowledgeable persons in each group.

Is knowledge about trees essential for tree planting and agroforestry activities?

Consultants were equally divided on the question of whether prior knowledge about a new species or practice is necessary to practice agroforestry. For traditional (i.e., indigenous) tree species cultivated by ancestors and inherited with the land, no additional knowledge is necessary as people observe them from early childhood. For newly-introduced, exotic trees, people need instruction because "you cannot plant a tree


SOURCE: den Biggelaar 1994

which you have just seen for the first time". Thus, with new species, farmers need information about (1) the utility of a species, (2) method of propagation and where and how to plant it and (3) where to obtain the necessary seeds, seedlings or cuttings. Other farmers, especially neighbours, were cited most frequently as information sources:

Farmers also become interested just by hearing about the benefits of certain new technologies. However, according to this Rwandan proverb:

A second group of consultants maintained that farmers can cultivate a new species without prior knowledge of its name, planting and management methods or its utility. The necessity of prior knowledge depends on personal preferences, individual spirit of research, the amount of risk s/he is willing to take, imagination of possible future benefits plus the aspect and physical form of the tree. This group of consultants was willing to take risks and look for new technologies that might contribute to a better livelihood. They all said that it was possible to discover the benefits of a species during growth or after its harvest. If the tree did not meet expectations, it could be cut and replaced with another species. It was also a source of pride to leave something new for posterity or to provide a showcase for neighbours who could try the innovations on their farms. For the majority of the population, observing what neighbours or friends have done and asking for information in order to follow their examples remains a powerful motivation to adopt new technologies:

The importance of neighbours and friends as a source of knowledge and new ideas is also expressed in the following Rwandan proverb:

Taking risks and learning

These three consultants offered anecdotes to illustrate why it is a good habit to take a risk and try new species without any prior knowledge in order to learn something new.

Consultant 1: One day I went to Butare. I stopped at Rwabuye to rest a bit when I saw a plant that I did not know but fell in love with. I uprooted it and the next day I planted it on my farm. I learned later that it was a guava tree which was not known in my region when I planted it. Therefore, it is possible to plant a tree without knowing its name or its utility.

Consultant 2: Walking in a forest, I came upon gapeli (Physalis peruviana) plants. I observed them with a lot of curiosity. One day, I bought their seeds from a boy for 15 ERw. I planted them among my coffee trees. After a while, I devoted a whole field to the young seedlings.This field enabled me to deliver fruits to KONFIGI for a total sum of 150,000 F.Rw. When one discovers a tree, one observes it. If one finds the tree can provide a benefit, one takes this tree home. This is true of medicinal plants. Their importance and utility are learned from neighbours who have used them. One does everything possible to introduce these plants within one's own fields.

Consultant 3: What I can add to the above is that in most cases the neighbour provides the information concerning planting such trees and yield. The person who introduced gapeli spread this species among us and we have cultivated this plant in our region for more than 15 years.

Farmer experimental practices

In each of the three processes of agroforestry knowledge-building (knowledge-generation/ acquisition/adaptation), farmers need basic information about species choice, location for planting and method of planting (from seed, seedlings or cutting). How does the farmer determine if a new species is beneficial? How does s/he evaluate trials with new species?

Choice of species: A species is chosen because of its benefits and utility, because it has been seen on another farm, or it has been talked about, or because of tradition. In a few cases, consultants mentioned having no choice, because they were forced by the State to plant certain trees, for example Eucalyptus sp. during the Belgian colonial period, and Grevillea_robusta as part of many tree planting drives by various rural development projects. Mostly, though, farmers planted trees because of their utility, even before the arrival of the `white people' and the establishment of nurseries. For example, a father would show his sons how to construct a home compound using such trees as Ficus thonningii, Euphorbia tirucalli, Vernonia amygdalina, Tetradenia riparia , Erythrina abyssinica. These traditional species, still used because they grow quickly and form a solid fence, can also be used as fodder (Ficus, Vernonia), medicine (Vernonia , Erythrina , Euphorbia , Tetradenia ), or to protect the inhabitants and their possessions from unwanted visitors (Erythrina ). Copying what other farmers have done is the most common way consultants choose a species to try. As was explained in one meeting, "Rwandans are imitators." Observing how the species developed and whether it was damaging to the soil and/or to associated food crops also helps determine if and where a species fits, i.e., if it conforms with other elements of the farming system (especially competition with crops), where it grows best (in the woodlot, fields or home compound) or its possible function (boundary marker, fence, food production, medicine, anti-erosion, etc).

Other criteria mentioned were availability, ease of planting and whether or not the species demands a lot of manure or fertile soils. One female consultant stated that it was a matter of chance; she planted many trees of various kinds because only some would survive. While most of the consultants chose their species carefully, they concluded that it is always good to try one or two trees to determine the results under the conditions of one's own farm.

Species location within the farm: There are no rules to govern where a species can be planted. Location depends on what the species produces or provides, its competitive and/or allelopathic effects, and its social benefits. For example, many smaller farmers plant fruit trees (particularly avocado) on the home compound, in order to free space in the banana fields for crops. In addition, Eucalyptus is often found on the home compounds of small farmers who can no longer afford to maintain woodlots.

Species location in the fields

There are several rules of thumb that farmers follow when determining where a species should be located within a field.

    - Timber and fuelwood trees and those that are highly competitive and/or allelopathic will be planted in woodlots distant from crop fields (for example Eucalyptus, Cupressus, Pinus, Grevillea).

    - Ficus was traditionally planted on the home compound to provide fodder but, as animal ownership has declined, many are presently found in the middle of the fields (remnants of former home compounds) where, according to the consultants, they contribute to soil fertility through leaf litter.

    - Species that are good for holding the soil are planted along anti-erosion ditches or on terraces (Grevillea, Vernonia).

    - Fruit and medicinal trees are grown in the fields closest to the home mostly for social reasons (fruit trees because of the risk of theft and medicinal trees in case of necessity). The exact location also depends on the farm's size.

How to plant species: Discussions of planting methods included planting holes, use of manure, time of planting and type of propagation material necessary (seedlings, cuttings or seeds). Most consultants again emphasized they obtained much information about planting methods through observation of other farmers and by asking for additional instructions if necessary. The size of the plant hole depends on the species: timber trees need big holes, but small holes suffice for species used as vegetables or medicine. Consultants disagreed on the necessity of adding manure. Some maintained that all trees need manure, just like crops. Others felt this wastes scarce manure because they believe manure is only necessary for fruit trees

To produce high quality timber, it was considered best to plant seedlings. Cuttings grow faster and are best for making a fence around the compound. The consultants agreed it is important to plant all the desired species at the same time when building a home compound, so they develop together and make a strong, dense fence.

Consultants differed about the best time to plant trees. One group maintained that trees should be planted at the same time as other crops, i.e. at the beginning of the short rainy season in September. Another group said that the seedlings would dry out if planted that early and should be planted in late October-November, when rains are more plentiful, which is also the time to obtain seedlings from the nurseries. As each species has it own planting method, both groups could be right. Seedlings are generally planted in late October-November, but cuttings and seeds of some species are often planted earlier. They did agree that, in view of the uncertainties of rainfall, plant diseases, marauding goats, landslides or weeding accidents, it is often necessary to replant a species several times before it survives.

Are the decision criteria regarding species, location and planting methods the same for all farmers? A minority of the consultants responded that the criteria are the same for all farmers because they follow the example of their parents or imitate what neighbours have done. In some cases, extension workers, through the information they provide, push farmers to make uniform decisions. But, the majority of consultants said the criteria inevitably vary because farmers have different motivations and objectives.

Our criteria cannot be the same. Some people calculate the benefits they will obtain beforehand and plant few or many trees as a result. Other persons are more carefree.

Some can plant Vernonia amygdalina along the anti-erosion ditches. Others can plant it on their home compounds or even on the ditch, but in their own ways.

There are many variables to be taken into account in deciding about species, location and planting methods. The consultants said they consider such elements as utility, benefit, desired end product of the tree, farm size, soil conditions, slope, competition between trees and crops, allelopathy, preferences, motivation, interest or possibility of theft. And, because there are many species available, some of which provide the same or similar benefits, an almost unlimited number of variations in agroforestry systems is possible. This diversification of species within an increasingly-complex farming system shows a deliberate strategy used by the farmers to overcome ecologic and economic uncertainties. Thus, the tree experts who participated in the species trial preferred to try unknown species.

The tree experts' species trial: studying farmer experimentation in action

A minority of the 44 tree experts chose to try species that they knew. One choose Grevillea, and less than five chose to try Leucaena leucocephela, Calliandra calothyrsus or Sesbania sesban. The remaining tree experts all rejected the tree species they already cultivated and provided a number of reasons for why they were interested in the new and unknown species. These reasons speak for themselves:

From the above statements, several themes emerge, some of which were already mentioned in the discussion of the nature and distribution of knowledge. On one hand, these themes relate to individual characteristics of tree experts such as curiosity, desire for knowledge, an internal drive for discovery and the hope to leave a better world for children. On the other hand, consultants expressed strong feelings about the importance of experimentation with new technologies for development and progress. They felt that lack of experimentation would lead to stagnation and make an already precarious existence even worse, that constant adaptation increases subsistence security by providing a diversity of species for farmers to choose from, and that the benefits of these new technologies could exceed those of presently available species.

In a follow-up question, consultants were asked if it were possible for all farmers to plant and experiment with unknown species. The vast majority agreed that everybody needs trees and:

Others talked about the benefits of planting unknown species, even though they may not always be obvious at planting:

One consultant stated that most farmers would be able to plant unknown species if they received sufficient information about the potential benefits, planting methods and best location for planting beforehand. Some qualified their answers by saying that the possibility of testing any new species depends on the size of one's farm:

Thus, in principle it is possible for everyone to try new technologies, but in practice not everyone will do so because of a lack of interest or curiosity or because of insufficient resources, especially land. For farmers who have little land, the risk associated with experimentation may be too great because they need all that land to produce sufficient food for their families.

How do farmers deal with risk associated with introducing new, perhaps unknown, technologies? Consultants were well aware of the risks in trying new technologies, especially if they had not been able to observe them somewhere else. But, such risks are also associated with well-accepted and common agricultural practices. For a farmer, it is impossible to escape risk:

Part of dealing with risk is accepting that it is there. Through education and training from a young age, farmers are made aware of the risks associated with cultivating some tree species. Consultants want prior information so they can deal with possible negative effects of new species. In addition to the inherent risks (i.e. toxicity, thorns, prickles, etc.), there may be negative effects on the soil or competition with crops or other tree species. If negative effects appear, the consultants can simply remove the species from the field, either by uprooting it or by replanting far from any food crops, preferably in a woodlot. If the species still shows promise and possible benefits to the farmers, the information gained from the experiment will be passed on to neighbours, friends and future generations.

Farmers' reactions to a new, unknown species: a hypothetical scenario

As an additional test of how they react to new technologies and how they would approach the incorporation of such technologies into their farming systems, the consultants were presented a hypothetical scenario. First, they were shown a seedling of a species they did not know, Chamaecytisus palmensis. This species had been tested for several years on the ISAR stations of Rwerere (Ruhengeri Prefecture) and Gakuta (Kibuye Prefecture) but none of the farmers in southern Rwanda had seen or heard of it. The seedling was passed around to the people at the meetings who sniffed its leaves, touched it, observed its stature and speculated about what it could be and how it could possibly be used. The following dialogue illustrates the initial reactions of the farmer-consultants when asked whether or not they knew the species.

After this introduction, the name of the species was given (farmers who had tested it around Rwerere had tentatively baptized it `umuvumbuka') but no additional information about its potential benefits. Consultants were asked what they would do if they were given 500 seedlings of this species. This high number was chosen deliberately to provoke audience reactions. The tree experts at these meetings were interested in testing new species and would very likely have accepted any proposition involving only a small number of seedlings. But the idea of 500 seedlings took them aback.

A number of consultants emphasized that it is important to get some information about the species before any decision is made about planting 500 seedlings:

Although most consultants felt it would be difficult to plant 500 seedlings of a species without knowing their utility, they did not completely reject the offer. One consultant stated that he would accept all 500 seedlings, but instead of planting them all on his own farm he would distribute 2-3 seedlings to each farmer in the cellule. Most consultants would accept from 2-4 seedlings so they could test performance and find out more about the utility of the species. If the seedling showed positive aspects such as fast growth, potential for profit, non-competitive behaviour and provided useful products, the consultants would increase the number of trees. In one meeting, consultants accepted much higher numbers of seedlings (100 to 200) after one consultant speculated out loud that, since umuvumbuka resembled a Crotolaria species, he could plant all 500 seedlings because, just like Crotolaria, it probably would fertilize the soil. Following are comments regarding the consultants' plan for testing the umuvumbuka seedlings.

A farmer in Karama pointing his climbing bean stakes. A total of 16 species were observed in this pile: Eucalyptus and Grevillea were the most durable (lasting six seasons), while Pennisetum and Vernonia were the least durable (lasting only one season).

One starts with a small number to determine the way in which they grow and their utility.

Farmers' approach to testing and evaluating tree species experiments

As mentioned previously, the species trial began at the end of the field research period in Rwanda. The following discussion is, therefore, not based on observations of what consultants actually did with the species they were given, but on interactions. during the focus group meetings. Consultants used the Kinyarwanda word for experimentation, igerageza, when talking about this species trial, which is quite different from the connotation of researchers trained in the scientific tradition. It is derived from the verb kugerageza, meaning to submit to a test. For trees, there does not appear to be any specific research method and tree experiments are not done on a separate piece of land. In contrast, farmers do something resembling a scientific experiment when testing new food crop varieties. For example, to test new bean cultivars, women use small plots of land to plant beans first in pure and later in various mixes of cultivars repeated over several seasons on different locations and soils (Sperling, 1992; Sperling et al, 1993; Sperling, personal communication). As trees take several years to mature or to yield any usable parts, most farmers cannot afford to tie up even a small piece of land for several years just to test out one particular tree species.

When farmers test a new tree species, they observe and evaluate it over a certain number of years in order to determine whether it meets their goals and expectations. They also study whether these new species will provide more benefits than the trees they already have. Most consultants are concerned with an end product: timber, fuelwood, construction poles, stakes, etc. If the species produces the desired product within the predetermined time frame, the trial is considered a success and the farmer may decide to plant additional trees of the species:

Consultants less fixed on a particular goal stated that other products or benefits may beobtained from the species being tested that may not be known at the onset of the trial:

A consultant in Maraba told of planting one Trema orientalis to test it for timber production. He was not able to explain what had happened, but the tree turned out to be a very bushy shrub. He realized that this shrub would never yield any timber and the experiment as such had failed. He was, therefore, not interested in planting additional Trema trees but kept the one he had so he could at least harvest some fuelwood from it. These examples also explain why the few farmers who had planted Calliandra calothyrsus had done so primarily to produce timber, not fodder, stakes or green manure. As can be seen from these examples, fuelwood is often not a major objective for planting trees 3 Farmers experiment primarily to discover the utility of a new species and to determine whether the species provides more benefits than those they already cultivate:

How do consultants evaluate their experiments and determine whether the new species will be of interest to them ? Consultants evaluate new species by size, form, amount of growth per year, diameter, health and the effects on crops and soils:

Almost all consultants talked about distributing plant material of the new species they
would be testing to their friends and neighbours. Such extension appears to be an im
portant part of the evaluation and validation of the success of these new technologies.

This emphasis on extension of information and knowledge gained from consultants' experimentation should not be a surprise, since communication of knowledge is a major part of being recognized as an expert on trees and tree cultivation. In the consultants' opinions, agroforestry knowledge is largely individual; collective agroforestry knowledge is built by sharing technologies and knowledge gained through trials. However, there is no central depository for this knowledge and this knowledge is not recorded in any written form, so any collective knowledge remains stored in the wisdom of many individuals.

Previous PageTop Of PageNext Page