ON AGRICULTURAL SUSTAINABILITY AND ITS MEASUREMENT IN SMALL-SCALE FARMING IN SUB-SAHARAN AFRICA

We argue in this paper that there is a need to bring the current debat e on sustainability down to a more pragmatic level, to 'operationalise ' sustainability. Given the diversity of agricultural systems througho ut the world, the best that is possible may be to develop proximate de finitions that do generate testable theories, as we try to do in this paper by developing a pragmatic and measurable definition of the conce pt which is relevant to small-scale farming in sub-Saharan Africa. The definition we offer takes the community of the village agroecosystem as a relevant and embracing social and spatial unit of analysis. It is based on the anthropocentric value judgement that future generations must have the same range of options concerning the use of agroecosyste ms as the current generation. Given that village agroecosystems genera te various environmental amenities for the human communities, this val ue judgment means that a sustainable agroecosystem is one which has th e capacity to respond to exogenous change as well as to internal disru ptions by maintaining non-declining trends in its resources and amenit ies over a period of at least one decade. The approach we advocate has the dual dangers of over-simplification and the creation of tautologi es. It should always be borne in mind that the criteria of sustainabil ity may differ from ecosystem to ecosystem, zone to zone, nation to na tion. Each situation should be scrutinised for the essential criteria for assessment, which may be different, and greater or lesser in numbe r, than those we have proposed. As a generality we do maintain, howeve r, that for progress in sustainability assessment, this process of cho ice should be based on the principle of parsimony. Tautology, the erec tion of self-fulfilling predictions, is a more fundamental danger in t he approach we have taken. We have emphasised several times the necess ity of rigorous testing of hypotheses concerning the determinants of e ach of the components of sustainability that we have described. This i s the real challenge to the researcher, but insight into the design of appropriate experiments may be gained from preliminary application of pragmatic tests of relative sustainability of the type described here . We try to avoid ecological and economic dogma as far as possible in our discussion of sustainability, attempting to argue instead from wha t we actually observe in the systems we study. Whilst a great deal of ecological and economic theory may seem at first look to be applicable to the analysis of the sustainability of small-scale agricultural sys tems, we find that closer analysis reveals the inapplicability of much of what has been subsumed under the titles of agroecology or natural resource economics. Concepts such as stress and disturbance, stability and diversity, community structure, economic optimality, maximum soci al welfare, do not, on inspection, provide particularly strong predict ive power, particularly when applied to the dynamics of resource-poor, small-scale, community-structured, and to a non-negligible extent, no n-market-driven agricultural systems.