Science in Society Archive

Sustainable Agriculture Pushing Back the Desert

Desertification - land degrading into desert - is often blamed on mismanagement and misuse of land. Local people are allegedly guilty of over-farming, over-grazing and allowing their populations to exceed the environment's capacity. Lim Li Ching contests this myth, describing how local farmers in arid Africa are using innovative means to farm productively without destroying the environment, and highlights some criteria for sustainable agriculture.

Debates concerning natural resources often pivot on a 'received wisdom' [1] about environmental change and people's role in this process. In the case of the environment, the received wisdom is that people invariably degrade natural resources. Outsiders, perceiving environmental change as degradation, blame local land-use practices.

The dominant idea about desertification has been that dryland environments are rapidly degraded by a combination of natural and human factors [2]. Desertification is defined as the degradation of drylands, involving loss of biological or economic productivity and complexity in croplands, pastures and woodlands.

From the 1930s, the blame was laid largely on the land use practices of farmers and herders, and on increasing populations. This was reinforced in the late 1970s and early 1980s, culminating in the UN Conference on Desertification in 1977. Some scientists were uncertain about the causes and extent of desertification, and expressed concern at the lack of long-term data. Despite that, the Conference ended by stating that desertification was threatening 19% of the earth's surface, and that this threat came from increased intensity of land use, overgrazing and inappropriate irrigation, exacerbated by drought.

Such claims were reiterated by the UN Environment Programme, which in 1984 professed that desertification threatens 35% of the earth's land surface and 20% of its population. UNEP was the driving force for the UN Convention to Combat Desertification (UNCCD), which entered into force in 1996. The rationale for the Convention is that "over 250 million people are directly affected by desertification, and some one thousand million (or one billion) are at risk…. Over the past two decades, the problem of land degradation in dryland regions has continued to worsen" [3].

Yet, evidence has been mounting that some of these assertions are unfounded. Most received wisdom on 'desertification' and 'land degradation' assumes an equilibrium environment with linear development. Thus, observations of expanding desert at certain periods and certain locations are extrapolated as ongoing, even accelerating, desertification.

For instance, work by Lamprey in northern Sudan, which estimated that the desert edge had shifted 90-100 km southwards in 17 years, is often cited. Whilst the period and location of the study was limited, it produced widespread understanding that "the whole southern Saharan edge" was expanding at a rate of 6 km a year [2]. These results were later disproved.

In the first instance, there are in effect three related but distinct phenomena - drought, dessication and dryland degradation - that have been conflated in the term 'desertification'. On a continuum, these have increasing time-scale effects and decreasing potential for recovery.

In particular, drought pulses are now seen as a key driving force of dynamic ecological systems - droughts lead to variability in ecosystem process and productivity, not its decline. And in many cases, what was assumed as dryland degradation is actually a result of drought, and can reverse quickly under normal rainfall. Additionally, data from dry years were often compared with that from wet years, ignoring longer-term climatic cycles. This led to an interpretation of a decline in productivity, rather than a variation in the response of natural vegetation or crops to soil moisture availability.

The new understanding is that arid and semi-arid areas are non-equilibrium environments, characterised by high levels of temporal and spatial variability and therefore, are unpredictable and uncertain. The critical indicator is a high coefficient of variation in rainfall (30% or more). Rainfall doesn't follow regular patterns, at least not in the short-term, and it affects variability of patterns and amount of vegetation.

This dynamic conception of drylands is underpinned by changes within ecological thinking - the 'new ecology' - that have suggested that nature is in a state of continuous change [4]. It contests conventional ecology, which depicts nature as tending toward stability, with notions of 'carrying capacity' and assumptions of a stable climax at equilibrium i.e. if the carrying capacity is exceeded, deterioration occurs. Conventional ecology overlooks climatic variables and change as a result of different combinations of factors, complexity and uncertainty.

The 'new ecology' also involves a conception of historical time, with emphasis on the irregular periodicity of environmental variations and ecological functioning. Initial conditions matter, as different environments change at different times. Land that appears 'degraded' may have already been that way long before farming or grazing. Researchers might erroneously associate degraded land with destructive human activity while knowing little of a place's environmental history.

Dynamic ecological systems mean that ecological drivers are external in dryland environments, and hence not necessarily subject to density-dependent events. Instead, human livelihood adaptations in these environments are very specialised; people are in reality raising meat and crops under ecologically sound conditions.

New research reveals that in many of the poorest African countries along the Sahara's edge, in Nigeria, Niger, Senegal, Burkina Faso and Kenya, integrated farming, mixed cropping and traditional soil and water conservation methods are increasing per capita food production several fold, keeping well ahead of population growth [5].

For example, the use of sheep manure for fertiliser has allowed increased yields for farmers in Kano, Nigeria. Additionally, planting leguminous crops increases nutrient levels in the soil by fixing nitrogen from the air. Integration of crops and livestock enhances nutrient cycling - legumes and manure return to the soil what crops take out. The Kano region is the most agriculturally productive part of the country, with increased yields of sorghum, millet, cowpeas and groundnuts.

A 4-year study in eastern Burkina Faso challenges the assumption that land is degrading largely due to human activities and the extrapolation that "soil fertility in Africa is at stake" [6]. It found that despite declining rainfall since the late 1950s and increasing populations, there is no evidence of land degradation connected to human activities nor a decline in food productivity. Conversely, yields of many crops have risen. The study found no proof of soil fertility decline over 30 years.

Farmers have not achieved environmental sustainability through a capital-intensive or high-tech path. In Burkina Faso, the increased yields of sorghum, millet and groundnuts is hardly attributable to increased external inputs, because these crops receive little fertiliser and are largely based on hand hoe cultivation.

Farmers have a rich repertoire of soil and water conservation technologies, such as crop sequencing, crop rotation, fallowing, weeding, selective clearing, intercropping, appropriate crop & landrace selection, adapted plant spacing, thinning, mulching, stubble grazing, weeding mounds, paddocking, household refuse application, manure application, crop residue application and compost pits. Mechanical practices include perennial grass strips, stone lines, wood barriers, earth barriers, brick barriers, stalk barriers, stone bunds, earth bunds and living hedges.

Perhaps more important than the practices is the selective way they are used, which vary with different field types, allowing optimal adjustment of limited labour and inputs to the requirements of different crops and soils. If land becomes limited, farmers do not need to invent new management systems; they apply these soil and water conservation practices more intensively. Farmers also apply land management practices only when and where needed. Using their knowledge of crops and soils, they treat only the parts of their field needing particular attention at any one time.

High local population densities, far from being a liability, are actually essential for providing the necessary labour to work the land, dig terraces and collect water in ponds for irrigation, and to control weeds, tend fields, feed animals and spread manure [5]. As population densities increase, farmers intensify their cooperation systems, grouping to tend each other's fields at busy periods, lending and borrowing land, livestock and equipment, and swapping seed varieties.

People thus invest heavily in creating and maintaining social networks [6]. These networks enhance ability to cultivate sustainably, allowing intensification by applying more soil and water conservation practices (see Box 1). Networks allow people to diversify their livelihoods and minimize risks, thus avoiding poverty traps.

Box 1. Social networks provide access to productive resources, reallocate slack resources

  • Land networks allow people to borrow land if their own land needs to be left fallow. This redistributes land between those who have use rights to more land than needed, and others who have not accumulated use rights to enough land to leave their own fallow. More land can be cultivated, while ensuring that cultivated land regenerates through fallow periods.
  • Labour networks permit farmers to complete tasks such as field preparation, weeding or harvesting quickly and on time. This frees a farmer's own time for applying labour-intensive soil and water conservation practices.
  • Women's natal networks often extend to different villages with differing agro-climatic characteristics, providing farmers with richer varieties of landraces to match variations in rainfall and changing soil qualities.
  • Cattle networks allow unused bush land to be grazed while land around villages is used for crop production. Farmers can own more cattle, while not overgrazing village territory. Networks with pastoralists include herding agreements whereby pastoralists graze cattle on farmers' fields to increase access to manure.
  • Technology networks allow intensified use of land-enhancing measures, letting farmers complete agricultural tasks on time by providing temporary access to technologies e.g. ploughs that increase moisture and nutrient availability by better incorporating plant residues into soil.
  • Cash networks

Furthermore, in Maradi district in southern Niger, where repeated droughts have wrought environmental damage, farmers have been fighting back, and have actually reversed desertification [5]. This is also true of Machakos (renamed Makueni) district in Kenya. In the 1930s, British colonial scientists condemned the eroding, bare hills of the drought-prone area to environmental oblivion; likewise the local Akamba people were doomed to a miserable poverty-rife existence. This narrative was consistently reproduced in the 1950s and 1970s. Yet, while there have been droughts, the hills are greener, less eroded and more productive today than before, despite a fivefold population increase. The Akamba had responded to the droughts by switching from herding cattle to settled farming, giving them incentive to work the land effectively.

"This is no high-tech breakthrough, nor a result of Western aid programmes" [5]. A major reason for the overestimation of land degradation is the underestimation of local farmers' abilities [6]. As shown, farmers have a large repertoire of technologies to draw from and have developed flexible, efficient and effective land management strategies while adapting their social institutions to cope with change and uncertainty.

This demonstrates the importance of relying and building on local people's knowledge and practices. Many external interventions have usurped and undermined local systems of decision-making and resource management. It's time to turn the received wisdom on its head, and learn from local communities, instead of blaming them wholesale for land degradation.

In light of all this evidence, a 'new realism' now exists about desertification, which gives climatic variation equal footing with human activities, as a cause [2]. The UNCCD now takes care to point out the reversibility of drought, the influence of climatic variation, and recognises that the causes of desertification are complex, as is the human-environment relationship [3]. Measures proposed to combat desertification are more modest, better linked to reality on the ground and admit uncertainty.

Policies need to appreciate the inherent uncertainty in science. 'Opportunistic management' [7], i.e. seizing opportunities to evade problems, working within complex systems, adapting to instability and exploiting environmental instability, is needed for dynamic ecosystems. The 'new ecology' calls for "flexible environmental management strategies that accommodate at once change, risk, complexity, and development based on local participation" [4].

Dynamic ecological theory does not replace conventional theory but is more appropriate in some contexts, such as in dryland ecosystems. Environmental complexity doesn't lend itself to simple, linear or reductionist technological fixes. Ecosystems are dynamic wholes and sustainable agriculture works in tandem with these (see Box 2), as local farmers in Africa are showing.

Box 2. Sustainable agriculture [8]

  • makes best use of nature's goods and services by integrating natural, regenerative processes e.g. nutrient cycling, nitrogen fixation, soil regeneration and natural enemies of pests
  • minimises non-renewable inputs (pesticides and fertilisers) that damage the environment or harm human health
  • relies on the knowledge and skills of farmers
  • promotes and protects social capital - people's capacities to work together to solve common management problems
  • depends on locally-adapted practices to innovate in the face of uncertainty
  • contributes to public goods, such as clean water, wildlife, carbon sequestration in soils, flood protection and landscape quality

Article first published 24/03/02


  1. Leach, M. and R. Mearns (1996) 'Environmental change and policy: challenging received wisdom in Africa', in Leach, M. and Mearns, R. (eds) The Lie of the Land: Challenging Received Wisdom on the African Environment, London: The International African Institute, 1-33.
  2. Swift, J. (1996) 'Desertification: Narratives, winners & losers', in Leach, M. and Mearns, R. (eds) The Lie of the Land: Challenging Received Wisdom on the African Environment, London: The International African Institute, 73-90.
  3. United Nations Convention to Combat Desertification,
  4. Zimmerer, K. (1994) 'Human geography and the "new ecology": the prospect and promise of integration', Annals of the Association of American Geographers 84(1): 108-125.
  5. Pearce, F. (2001) 'Desert harvest', New Scientist 27 October 2001, 44-47.
  6. Mazzucato, V. and Niemeijer, D. (2001) 'Overestimating land degradation, underestimating farmers in the Sahel', Drylands Programme Issue Paper No. 101, London: International Institute for Environment and Development,
  7. Behnke, R.H. and I. Scoones (1993) 'Rethinking range ecology: implications for rangeland management in Africa', in Behnke, R., Scoones, I. and Kerven, C. (eds) Range Ecology at Disequilibrium, London: Overseas Development Institute.
  8. Modified from Pretty, J. and R. Hine (2001) Reducing food poverty with sustainable agriculture: a summary of new evidence, Centre for Environment and Society, Essex University,

Got something to say about this page? Comment

Comment on this article

Comments may be published. All comments are moderated. Name and email details are required.

Email address:
Your comments:
Anti spam question:
How many legs on a tripod?