Cover Image
close this bookIndigenous Agroforestry in Latin America: a Blueprint for Sustainable Agriculture? (NRI, 1994, 24 p.)
View the document(introduction...)
View the documentForeword
View the documentAcknowledgements
View the documentGlossary
View the documentSummary
View the documentIntroduction
View the documentForest destruction and environmental degradation
Open this folder and view contentsSwidden agriculture
View the documentBenefits of swidden agriculture
View the documentRecent agroforestry research
View the documentBlueprint for colonist agriculture?
View the documentConclusions
View the documentReferences
View the documentFurther reading

Blueprint for colonist agriculture?

Much of the research completed to date on indigenous agroforestry practices has been of a descriptive nature. There have been few attempts to translate the results of an improved understanding of the complexity of these systems into practical advice for new colonist farmers in Latin America.

Warner (1991) argues that indigenous swidden management cannot serve as a model for agricultural development for future farming systems in the tropics. She states that the regeneration of the forest is crucial for the longterm productivity of swidden systems and many traditional farmers are no longer able to leave their fields fallow for a sufficient length of time to maintain productivity. Also, it is argued, that the technical knowledge of indigenous communities is too area-specific or too tied to cultural and religious systems to be readily transferable to other societies and groups. Wilken (1989) also argues along similar lines, suggesting that transfers of traditional technology require a high degree of social, economic and technical agreement between senders and receivers.

These arguments, however, appear to ignore the success of both riberecaboclo and Japanese (see Cases 6, 8, 10 and 11) adaptations of traditional agricultural techniques in the Amazon Basin, and the success and adaptability of Huastec agroforestry in Mexico (see Case 5). The Japanese of Tome Acu, although they had little cultural similarity and had access to capital and education, based many of their agricultural practices upon the techniques of the Amerindian, as it was essential to do so to survive during the early days of their colonization.


It is acknowledged that there are enormous cultural differences between the Amerindian and the recent colonists in Latin America. It will not be a simple task to transfer the more sustainable indigenous systems to the new colonists whose current practices are based upon the techniques used in their home states. It is admitted that many traditional Amerindian practices are too area-specific and tied to indigenous cultural practices to be wholly transferable. However, the principles of traditional practice have been adopted by the caboclos and ribereof the Amazon Basin, in part commercialized, and can serve as a basis for the development of agroforestry practices by the new colonists. The more important elements of traditional practice suitable for replication on colonist farms are listed below.

1. The integration of trees into the agricultural system, in particular in enriching fallows to increase their productivity.

2. The flexible use of micro-environments, for which different crops are suitable, and the maximum use of light and nutrients in multi-cropping practices.

3. The use of intensive mulching practices alongside extensive fallow 'slash and burn' practices.

4. The maintenance of agroforestry plots at different stages of succession, which reduces risk and maintains flexibility in the system.

5. The use of wild game and fish.

The adoption of agroforestry does not imply an end to the production of new colonist preferred annual crops such as rice, or of the rearing of domestic animals. Annual crops are produced at the beginning of the swidden cycle, and animals are integrated into indigenous practices.

Systems based on indigenous agroforestry may however, result in a reduction in the area per farm of annual crops as farmers become more dependent for subsistence and sale upon tree crops in the swidden succession. This would be a positive change resulting in less annual destruction of forest for the production of staple food crops.


There are several major constraints which may prevent the successful transfer of traditional agroforestry techniques to colonists in the Amazon Basin.

1. Those currently practicing swidden agriculture have low social status which is a barrier to the adoption of their techniques by settlers (Dejou, 1990). Many new colonists aspire to become 'modem' cattle ranchers despite the low productivity of ranching in the Amazon region, as they see cattle as an inflation-proof capital reserve.

2. Those who practice traditional agroforestry systems are increasingly migrating to centres of population to benefit from health and education services. Their children do not inherit their local technical knowledge. Even where such farming families reside in rural areas, their children attend boarding schools elsewhere and do not learn farming skills. There is, therefore, a danger that their knowledge may become increasingly diluted over the coming decades.

3. Recent colonization has resulted in little contact between new colonists and local traditional farmers, leading to little exchange of information and agricultural techniques (Parker, 1989). The latter inhabit, for the most part, lowland floodplain areas, whereas much of the colonization has followed the construction of roads in the dry lands.

4. It appears to be technically possible to produce a wide range of useful fruits and other products from traditional agroforestry in Latin America. Marketing of these products is seriously constrained, however, by the distance to market, crop perishability, a shortage of processing capability, a shortage of credit and a lack of marketing co-operatives. Where markets are not too distant, agroforestry and extraction appear to be reasonably successful, e.g. Iquitos in Peru and Belem in Brazil, although the greater the distance from the market, the less profitable they become (Padoch, 1987). The urban areas of Amazonia are growing, and half the population of the region is estimated to be living towns (Butler, 1989). Therefore, there appears to be a market for new colonist annual crops but not a well established market for traditional forest and agroforestry products.

5. Title to land, or at the very least usufruct rights to land over the long term are required for the successful adoption of agroforestry. Indeed, successional swidden management requires access to a particular piece of land for up to 30 years or more. Current new colonist agriculture can only succeed while there is further land at the frontier to colonize. If title could be guaranteed to settlers and no further rights allowed whilst in possession of a plot, then the colonist may be encouraged to invest over a longer period in the land. There is, therefore, a need for government policies to be put in place which help to curb land speculation, and / or grant title or clear usufruct rights in order to encourage the new colonists to adopt agroforestry practices.

For example, there is little evidence as yet that neighbouring Brazilian farmers are adopting the Japanese techniques (see Case 11) on a large scale. They do not have secure title to land (a prerequisite for obtaining credit) and cannot, therefore, invest labour and capital into a perennial crop-based system that requires several years before it begins to yield a return.

6. Little research has been completed on productivity and labour use in swidden agricultural systems. It is not clear how much land is required for household subsistence. Although high incomes from riberegroforestry in Peru have been reported (Padoch and de Jong, 1987), these are the result of the proximity to the market. There are few references to actual labour use in traditional systems of agroforestry. It is clear that in successional management systems labour requirements on a given plot decline over time. High inputs, however, are required at the creation of the swidden and during the production of annual staple crops. As tree crops begin to dominate, labour needs are much less; only occasional weeding and harvesting work is required. However, where there are dense stands of tree crops ready for harvesting, labour requirements will once again be substantial.


From the factors outlined above, further research is required to enable a suitable blueprint for sustainable agriculture in the tropical rainforests of Latin America to be drawn up. The main research areas are listed below.

1. There is a danger that the agricultural systems of the indigenous populations of these regions may disappear in the not too distant future as contact with outsiders increases. Although market-orientated systems derived from Amerindian techniques of swidden management are not subject to the same pressures and will probably survive, they are adaptations of traditional practices. It is therefore essential that further research is commissioned on indigenous management systems whilst this is possible.

2. Micro-economic research is needed to establish resource requirements and productivity, especially of land and labour in both Amerindian production systems, and those derived from them. Such research needs to identify the socio-economic and biological constraints to improved performance and replication.

3. Further investigations are needed to establish the market potential for the products of traditional agroforestry practices, and to assess the effects of market development; i.e. if markets were to develop would this result in large scale clearance of forest? There is little or no evidence to suggest that any Latin American farmers would not readily respond to market opportunities.

4. More research is needed in the fields of sustainable agroforestry both formal and of a more participatory nature. The Overseas Development Administration is already addressing this problem with its 'Floodplain Forest Ecology and Management Project, Belem' and the 'Tocantins Forest Management and Rural Development Project', although the latter does not intend explicitly to utilize indigenous technical knowledge as part of its objective to encourage sustainable farming systems. The need is so great, however, that there is potential for many further such projects throughout the Amazon region and more generally in Latin America.