Les représentations généralement données des forêts tropicales ont trop long
Abstract-Tropical forests, myths and realitiesTropical forest and savannah Iandscapes are a particularly important theme in the environment 1 development paradigm, and countless projects are being conducted in the field. However, over the past few years, studies have been multiplying. that lead us to ask oneself whether our current approach of the dynamics of theses spaces and of their management by the local communities, does correspond to the actual devetopment of these systems and of the practices involved. Could it not be thal a subjective reality, our own reality, born of a specifie sociocultural, economical and historical context. has been constructed then perpetuated ? Does this view of the various environments, modes of exploitation and organisation of space and production on which our interventions are based match the actual tacts? Is it operative with regard to the development and durability of these environments? These few questions are addressed in this paper in the light of previous and sometimes fairly old studies more or Jess oblirated by various form of pensée unique, or recent works that radically question the postulates on witch the western views on the subject are based. The numerous interventions on these ecosystems are indeed based on a number of widely acknowledged and therefore not questioned ideas, whose origin often dates back to the beginnings of cotonization, sometimes even in the very first explorations. This set of postulates, a construction belonging to a given political, economie and social context, has prevented and still prevents one from making out the actual interactions at work between these environments and these societies. Il is really important to try and blow away this smoke screen because, after ali, if we are now trying to protect the biodiversity and much praised rich ness of these spaces, il is also because sorne societies have taken care to maintain or construct them despite the trials and tribulations of history. Local populations, even the most "primitives· ones, have sophisticated management systems that are able to slowty but effectively modify the natural environment ; these systems, including and especially slash and burn cultivation are numerous, originals, evolutives and capable of intensification. Fallow land is not simply left to pedologie potential recover, it is a productive land and, if necessary, may develop into various forrns of forest gardens or agroforests. Similary, the existence of dominant types of production systems should not hide the fact that the farmers manage multipurpose activities within the framework of multifonctionnal productions and plots. They do so empirically and with such flexibiliy 1 adaptability as demanded by their frailty in the face of an unforeseeable future. For having always viewed these environments and these societies throught stereotypes, maybe our views have remained frozen after ali? vie, jugés incapables de modifier les écosystèmes, paraissai...