Previous Page Table Of ContentsNext Page

Chapter 4 - Choosing the right tree for the job

This chapter will attempt to draw together the various indications emerging from the analyses in the preceding chapters and suggest a simple algorithm for the selection of trees and tree planting practices to fit the diagnosed needs and opportunities of specific clients. The hope is that the adoption of this or some equivalent approach to designed tree planting interventions might replace the haphazard style that has characterized the majority of tree planting projects thus far, supporting a new standard of project relevance to people's needs.

A three-pronged strategy

Before presenting the methodological tool and the supporting resource materials, however it may be useful to indicate the limitations of the suggested methodology and its place within a comprehensive decision making strategy. The systematic approach to tree planting interventions presented here does not and cannot represent the whole of the strategy required for successful tree planting projects; it represents only the middle part of the following three-pronged strategy:

If there is one lesson to be drawn from the past decade of experience in tree planting projects it is that there is no substitute for local experience. The indigenous knowledge and decision making strategies of local people represent ages of accumulated local experience which are, in the first instance, the best source of project guidance. The most important strategy for successful tree planting projects, therefore, is to acknowledge the decision making sovereignty of the local people (they will do what they want in any case) and assist them to make informed choices. Failure to implement this common-sense strategy is the most destructive and at the same time the most easily avoided of mistakes!

The intended beneficiaries are not only the ones best equipped to sort out what works best for them, they are also the only really important judges of project success or failure. The judgments they express, not through speech but through their adoption behaviour, are the sole lasting evidence of project impact (or lack of it). To "give them their heads" is not simply the best way of avoiding project mistakes, it is also the easiest of strategies to implement. All tree planting projects have nurseries or seed supply operations. Simply asking people what they want in the first place, and subsequently monitoring what they actually take away with them from the nurseries or seed stores over time, is the best source of feedback information on "what works" from the adoption standpoint (see the box on the next page for an indicative summary of what has worked in one region of the world).

Characteristics of successful tree planting programmes in Central America

Having said that letting people make their own choices is a necessary condition for project success, it must also be said that this will not always be a sufficient condition. Locally familiar uses of trees and the simpler tree arrangements and management practices will not suffer much from the lack of external inputs, but a wide range of potentially useful--but less familiar or more complex--agroforestry practices may go untried without the catalytic activism of outsiders.

Recognition of the value of indigenous knowledge and dedication to participatory principles must not prevent us from facing facts. Many rural communities are in serious trouble due to unsustainable land use practices. While new land use systems must be based on and integrated with local expertise and traditional land use practices, without infusions of new ideas and plant materials, many local communities are doomed. The archaeological record is littered with extinct cultures that failed to meet the adaptive challenges posed by changing environments. To minimize the hazards faced by rural communities in the name of sensitivity to local traditions is a cruel and misguided form of dilettantism.

Participation is a two-way street. If tree planting projects are to have any justification, they must be prepared to play an active catalytic role. They must be active not only in facilitating the application and further development of local expertise, but also in providing relevant new models for the local people to experiment with, and then doing whatever is necessary to bring the innovations to trial. Local people are, ultimately, the best judges of what will work for them, but it is not fair to expect them to render their best judgment until they've had a chance to give an innovation a fair trial.

The old paradigm in the application of scientific forestry and agriculture to problems of rural development is based on the implicit assumption that "the experts know best." Lately this has come under heavy criticism and is tending be replaced by the opposite assumption, i.e., that "the local people know best." Eventually, if full effectiveness is to be achieved, we must acknowledge that the most realistic and genuinely helpful stance in the long run is that both outside experts and the local people each have unique areas of expertise which together provide a better foundation for development than either in isolation (Raintree and Hoskins 1988).

Project personnel and local people must learn from their shared experience and set their standards high enough to address the long-term needs of the community, not just those that can be met by easily adoptable practices. All project participants must treat all initial choices of trees and tree planting practices as working hypotheses.

The following methodological suggestions and supportive resource materials are intended as a contribution toward what is needed to equip the "new forester" to play an open-minded but not empty-headed role in catalyzing community-based activities.

The range of tree planting options

We have seen in the preceding chapters that an appropriate tree planting practice involves more than just choosing a tree with the right attributes. This is because the attributes that determine the appropriateness of a particular tree are themselves conditioned by the effects of a whole set of other decisions about the tree growing technology: the management system under which the trees are grown (i.e., the pruning, lopping, coppicing, pollarding or harvesting regime), the spatial arrangement in which the trees are planted (i.e., the pattern and density of planting, either singly or in combination with other trees and crops), and the location within the landscape where the trees are planted. All of these decisions, in turn, depend on the specific function the tree is intended to perform for a particular user within a particular socioeconomic context and development strategy.

To avoid unintended consequences and to sharpen the focus of particular tree planting interventions we ought to deal with these factors in a conscious and systematic way. We can do this by making each of them the focus of a deliberate design decision, in accordance with the decision framework presented in Table 7. These decisions are indexed to supportive reference materials contained in the appendices, indicating the wide range of options which exist at each of the decision points.

The suggested decision criteria, case histories and reference materials are placed in the Appendix in order to preserve the logical simplicity of the decision algorithm. Also, by separating them from the logical framework, we can indicate the flexibility and judgment that is required to deal intelligently with the decisions that are implicit in any act of tree planting. The extensive collection of reference materials is intended to open-up thinking about alternatives in the context of a creative, interactive and participatory approach to the design of tree-based interventions. No claim is made that they represent a definitive catalogue of options. While the material presented in these appendices represents a distillation of considerable multidisciplinary experience and judgment, and may therefore prove useful as a source of heuristic guidelines for tree-planting projects, under no circumstances should they be treated as recipes for unthinking implementation.

A simplified matching algorithm

We started out by positing that there exists a systematic relationship between the biophysical attributes of trees and the socioeconomic attributes of tree users, reflecting what we have called the "socioeconomic attributes of trees."

We have seen that it is not until the tree growing practice or technology is specified that trees begin to exhibit definite socioeconomic attributes. Without the specification of which technology they will be used in, the attributes of trees remain indefinite because the siting, arrangement and management of the trees can modify or override any of their inherent biophysical attributes. Likewise, without specifying which user the intervention is intended for, the attributes have no socioeconomic frame of reference. There is still a great deal to learn about the socioeconomic attributes of particular trees but we are now in a position to suggest a practical expedient.

If we group the main decision factors together under the heading of the tree growing technology and if we treat technology as an intervening variable between tree users and tree attributes, then the task of making rational decisions about species choice for different users becomes a relatively simple two step process: 1) first match the technologies to the users and then 2) match the trees to the technologies.

Table 7. A decision algorithm for matching technologies to users and trees to technologies.

(To use this simple design tool, first work through the questions to arrive at a design for each client group within your area and then go back and review the interactions between groups, adjusting the designs to reduce conflicts and accentuate complementarities.)


Choice of client and strategy

See Appendix


Who is the intended beneficiary or client?


What is the appropriate development strategy

C.2, C.3

given the socioeconomic context of this client?



Choice of technology to match the client

B.7, C.2, C.3

What specific function, is it intended to serve?


At what location within the landscape?


In what planting arrangement?


Under what management?


With what inPut requirements?


With what output expectations?



Choice of tree to match the technology


Ideotype specifications .


Choice of tree to meet the specifications.


Matching technologies to users

Appendix B.7 presents an overview of the technology repertoire, followed in appendices C.2 and C.3 by some mutually reinforcing perspectives on "what fits where." Any and all of the suggested ways of matching technologies to users can be used to identify "candidate technologies" for further consideration. Ultimately, of course, validation and adaptation of the technology will be carried out by the intended users.

Matching trees to technologies

Once a particular tree growing practice or technology has been identified as relevant to a particular user, the next step in the selection algorithm is to find trees with attributes that match the requirements of the selected technologies. This is facilitated by first writing a set of "tree specifications" to serve as a search frame for identification of appropriate species, provenances and cultivars. Appendix D contains a lengthy summary of indicative tree specifications for a selected set of tree-planting technologies that were derived from D&D exercises undertaken by ICRAF scientists in collaboration with national scientists in a wide range of land use systems in Africa, Asia and Latin America. They represent preliminary results of a larger on-going multidisciplinary work in progress and should not be treated as more than a collection of indicative examples of the kind of considerations involved in writing tree specifications for situation-specific variants of the selected technologies.

Following the derivation of tree specifications, the next step in the decision algorithm is to select trees which match the specifications. Given the very inadequate state of current knowledge on the design-related attributes of most trees, no attempt has been made to develop yet another list of tree species for tree planting projects.

While it is certainly important to provide reliable lists and sourcebooks on tree species for technologies within the expanded repertoire of the new forester, any attempt to arrive at an authoritative compendium at the present time is severely constrained by the lack of reliable knowledge. To put it bluntly, most of the people who know trees know them only from a limited disciplinary point of view (i.e., as foresters, or horticulturalists, or range management specialists, etc.). Very few have given any systematic attention whatsoever to the attributes of trees within an open-ended agroforestry design framework. An exception to this generalization is the recent book by Rocheleau, Weber and Field-Juma (1989) which contains a design-oriented catalogue of agroforestry species for dryland farming conditions in Africa. Hopefully resource books of this type will become available for other regions in due course.

For the purposes of the present study, rather than attempt to develop yet another species list, it seems preferable to keep the emphasis on the logic of the discovery procedure and encourage tree planters to use this logic to make their own selections for local trial. In the final analysis there is no substitute for in-depth, site-specific trial and evaluation.

Previous PageTop Of PageNext Page