Shorter version published as:

Batterbury, S.P.J. & A.Warren. 2001. Desertification. in N. Smelser & P. Baltes (eds.) International Encyclopædia of the Social and Behavioral Sciences. Elsevier Press. Pp. 3526-3529.



Dr Simon Batterbury, SAGES,  Univ. Melbourne

Professor Andrew Warren, Department of GeographyUniversity College London


Key words: desertification, Africa, dry lands, deserts, drought, land degradation, Sahel, Dust Bowl, UNCED, international agreements




Desertification is difficult to define in measurable terms, and many of the assertions made in its name have proved alarmist. In large measure, the problems arise because environmental damage is judged differently by different cultures at different times. For this reason, evidence for overgrazing, desert advance, climatic change, the fragility of dry ecosystems, and the effects of increasing population have all been elusive.  Nonetheless, the term has an interesting history, dating from fear about the spread of the desert in the US midwest and in West Africa, which dates from the early years of the 20th century.  The severe desiccation of the 1970s and 1980s revived the term, and international interest in it.  “Desertification” is best seen as a portmanteau word for serious environmental problems in dry countries, many of which are among the poorest on earth.  One problem is drought.  Although many ecological and social systems have evolved to deal with drought, societies in rapid transition are vulnerable.  Another problem is “desiccation” (the longer term drying out).  There have indeed been long droughts or dry periods in recent history and they have caused havoc in many societies.  Finally, there is “degradation”.  One form of degradation, overgrazing, is now thought to be far less of an issue than it was twenty years ago.  Soil erosion is another form of degradation, and although probably very serious in some of the drylands, not enough is known of its drivers.  For all the difficulties with the term, it has a future, if only in the United Nations Convention to Combat Desertification, to which many dry countries have signed up, and which will require planning into the early 2000s.



The importance of desertification for the peoples of the world’s drylands should not be underestimated,  despite considerable uncertainty about its meaning. Firstly, the term, for all its problems, envelops a set of processes, sometimes listed as drought, desiccation and degradation, that pose very real and severe challenges to drylands, and have strong knock-on effects on society.  Secondly, desertification language is deeply entrenched in policy circles and in the current UN Convention to Combat Desertification (CCD), and will inevitably continue to inform the financing of development interventions in drylands.


1. History and Definitions


The “Agenda 21” document of the 1992 United Nations Conference on Environment and Development (UNCED) defined desertification as “land degradation in arid, semi-arid and dry sub-humid areas resulting from various factors, including climatic variations and human activities”. The processes of degradation were said to be soil erosion, nutrient depletion, crust formation, salinization, reduction in pasture and agricultural productivity, loss of biodiversity, and reduction in vegetation cover in susceptible drylands. Of course, uncertainties abound: the circumstances fostering reversible or irreversible change; the resiliency of different semi-arid environments; the different roles of natural and anthropogenic driving forces; and the synergies among these elements.


Scientists were exercised as early as the 1920s with the ‘the advance and spread of the desert’ in Africa.  It was then that France began the first studies of the process in West Africa, finding evidence of mobile sand dunes, human fossil remains in now-dry areas, and declining annual rainfall. It was concluded that the Sahara had grown, and was still growing, owing to poor land management, which had worsened under the colonial regime.


These early judgements preceded what environmental historians are now discovering to have been a much more traumatic event, the American ‘Dust Bowl’ of the 1930s.  Sears’s Deserts on the March  (1935) evoked accusations of widespread anthropogenic degradation. He and others believed they also had evidence of widespread environmental degradation in the dry lands of Europe, whose cause, as in the Midwestern states, was also thought to be mismanagement.  With these narratives very much in mind, E.P. Stebbing, a forester, identified the causes of degradation in British West Africa to be shortened agricultural fallow periods, shifting agriculture, and overgrazing (Stebbing, 1935). An Anglo-French Forestry Commission [which went to look at forestry in the European colonies] toured the Niger-Nigeria border in 1936-7, and their conclusions were far more circumspect.  They saw degradation as place-specific and treatable. Yet a member of this Commission, the influential French botanist Auguste Aubréville, held to the ‘desert advance’ hypothesis, and first used the term ‘desertification’ in 1949.


In dryland Africa, severe drought and famine in the 1970s, following decades of good rainfall, again revived the desertification debate, with depressing reports of desert advance appearing in scientific publications and the media. This alarm stoked the United Nations Conference on Desertification (UNCOD), held in Nairobi in 1977. UNCOD, and the National Plans of Action agreed by each participant dryland country  at that meeting, still viewed people as the main agents of desertification.


2. Current research


Research into the links between climate, management and degradation has taught us a great deal since the 1970s.  Since then, many detailed studies of human-environment interactions at the local scale have been carried out in the drylands, spurred on by advances in climatology, soil science, botany, geo-informatics, agronomy, and the social sciences.


This research, conducted by numerous scholars from the natural and social sciences, offers a longer-term view of what happens as rainfall fluctuates, or when soils are lost and gained. More generally it has examined the complexity of the linkage between resources and people in the drylands, and the role of adaptive local resource management. Above all, it is now possible to challenge notions about an equilibrium state or “carrying capacity” in these environments and from there to challenge the desertification narrative with better, if still imperfect evidence.


Yet laying the desertification discourse to rest is difficult. At least six themes remain essential elements of the desertification debate.


Advance of desert. Large-scale Saharan expansion has not been proven within the period of the historical record.  Desert advance in the Sudan  has been asserted since the 1930s, but Helldén (1991) found no such evidence from satellite imagery, settlement histories, sand movements, and degradation conditions around boreholes. The current view is that the Sahara expands and contracts periodically as  rainfall varies. Desert advance (as in the severe drought of 1984), and localised degradation, is usually short-term (Tucker et al. 1991; Nicholson et al. 1998).


Resilience of dryland ecosystems and land-use systems. Degradation is, therefore, usually localised and ephemeral [short-lived], largely a function of the complexity of relationships that exist between humans and natural systems. Semi-arid ecosystems, far from being fragile, exist in a range of semi-permanent ‘states’ dictated by disturbance, drought, fire or insect attack; they are well adapted to these forces.  Similarly many indigenous management systems redistribute nutrients, plants, grazing pressure and water to create ‘enriched’ patches in the landscape.


Influence of grazing and livestock. The UN Atlas of Desertification (1992) asserted that 58% of soil erosion in dryland Africa was the result of overgrazing by livestock, lending support to Garret Hardin’s ‘tragedy of the commons’ hypothesis, that said that too many herders all using the same piece of rangeland would tend to degrade it through their own selfishness.  Such figures gave way to more conservative estimates in the latest version of the Atlas (Thomas and Middleton, 1996).  There is now greater confidence about the heterogeneity and patch dynamics of pastoralist grazing strategies so critical to many drylands. Less existing vegetation is not necessarily worse vegetation in terms of biodiversity. It is difficult to overgraze in a dynamic non-equilibrium system, dominated by annual grasses, where the external forces like drought are more powerful animal numbers as influences on rangeland quality.


Effects of increased population in rainfed dryland agricultural systems. No necessary Malthusian link exists between elevated population levels and resource degradation. [see earlier indv 103 lecture] .  Indeed, because there is often an incentive for rural people to invest in anti-degradation measures in these circumstances, more people (and labour) may mean less erosion, and the initiation of compensatory risk management strategies.


Soil erosion and fertility decline.  Although topsoil erosion does occur in drylands and soil is blown or washed away, these processes do not themselves necessarily result in degraded landscapes. Soils are often replacable or may be very deep, and can accumulate down-slope or down-wind where they may be more valuable to local people. Indigenous dryland soil conservation systems are impressive. Yet even after decades of research, soil erosion and the values placed upon it by society are very difficult to measure.


Climate Change.  There are two issues here. Charney & Stone (1975) suggested that devegetation might induce regional drought bare soils are cooler because they reflect more solar energy and cool surfaces discourage rainfall.  This hypothesis remains disputed.  Current research provide strong evidence that changes in surface characteristics, including vegetation, affect local and regional rainfall patterns significantly. Some climatologists however, find little evidence in this claim as registered in the coarser scale climate record (Williams & Balling, 1996). As well, potential global warming and the ENSO (El Niño-Southern Oscillation) will have differential, and as yet uncertain effects on certain drylands, with the majority of current models suggesting increased rainfall and temperature variability in Africa.


3. Desertification Policies


Research, policymaking and interventions are strongly linked. For example, when scientists pinpointed advancing deserts, the policy response, was ‘plant green belts!’ [around the Sahara]. Roosevelt commissioned a study of a massive green belt of trees between Texas and North Dakota after the Dust Bowl. Stebbing recommended much the same for Africa, and UNCOD proposed a circum-Sahara green belt.  The Algerian government actually began planting trees, employing conscripts to do so.  Where scientists believed that herdsmen or farmers were irrationally causing degradation, policy-makers and government officials prohibited goats, the destocking of herds, the prohibition of tree cutting or grass burning; or on a much more massive scale, they enforced soil conservation programmes.


Anti-desertification measures have been international in scope. A suite of international institutions materialised after the great Sahelian drought of the late 1960s and early 1970s. First, the United Nations Sudano-Sahelian Office (UNSO) was charged with combatting drought, later taking on desertification issues. It was later joined by the UN Environment Programme and the FAO in these roles. Scientific panels and advisors produced a “veritable sandstorm of literature” (Mortimore 1998:23) on the complexities of drylands. In general, the prognosis was gloomy: 10% of the earth’s surface was human-made desert; and the process would continue.  Desertification became enshrined in international policy.


African states absorbed some of UNCOD’s recommendations from the late 1970s, although they were held back by poor funding. Projects were initiated, most designed to arrest soil erosion. Some were implemented without adequate local consultation or participation. Many, like the GERES project in Burkina Faso where bulldozers were used to create earthen bunds across many slopes to slow water runoff, actually increased erosion rates.  Fuelwood supply projects [to plant loads of trees as plantations] and the drastic destocking of indigenous pastures [reducing livestock numbers] failed for the lack of local complicity.


Unforeseen ramifications followed in the post-UNCOD, post-Sahelian drought era. In Niger, West Africa,  proposals for green belts resorted to authoritarian measures for implementation, for in local eyes, as in the eyes of scientists today, they were deeply irrational. In similar vein the military rulers of Mali banned the burning of forests and grasslands to release nutrients to the soil, this was  counter to local know-how about the enriching powers of fire on pastures. In many countries, the cutting of fuelwood was unnecessarily proscribed. Most disturbingly, the characterisation of African pastures as "overgrazed" came to infest writing about them.  Attempts were thus made to reduce the numbers of livestock; to privatise grazing lands; and even to replace herding the commons with systems based on European and North American models (Warren 1996). The desertification threat was used to justify authoritarian control over natural resources and land-use practices.


A broadened, perhaps softened conception of desertification suggests a new range of policies that, given sensitive application, may prove more successful because they are more trusting of dryland populations and more aware of ecological diversity. New initiatives are based on a better understanding of the three main issues in dryland management: drought, desiccation and degradation.  Drought is common currency among dryland folk.  It may not be welcome, but people clearly have a portfolio of coping strategies to deal with it, and reflexive management that constantly adapts to an ever-changing situation. These indigenous systems are now being supported, rather than replaced or obstructed. Desiccation is known to have been frequent in the history of human occupation of drylands.  Large-scale shifts in livelihood practices occur under progressive aridity, but Sahelian drought illustrates that relief, restocking and resettlement are necessary when this happens.  Degradation requires a range of responses.  Rich repertoires of conserving indigenous techniques exist, some – like permeable contour bunds - already improved through hybrid experimentation between farmers and scientists. Drylands and dryland peoples and their economies are heterogeneous.  Policies increasingly aim to protect this diversity, rather than erase it. Greater clarity of land tenure security is needed, and rangeland areas require special forms of collective management . In brief, livelihood security is essential to good conservation.


4. The Convention to Combat Desertification


UNCED, the Rio Conference,  held in Rio in 1992, breathed new life into the unpopular, almost forgotten ‘desertification’ concept.  Fortunately, Chapter 12 of UNCED’s Agenda 21 document is far less alarmist in its tone than UNCOD’s recommendations of fifteen years earlier, adopting some of the findings and views of the new research reported above. For example, climate change is now seen as a major contributor to desertification, and drought preparedness, education, and strong local participation are now stressed.


The Convention to Combat Desertification (CCD) came into force on 26 December 1996 (see for full details and current negotiations). Although there was scepticism about this new international effort, non-government organizations and other actors from developing countries had a relatively strong involvement in drawing up the Convention. The various CCD components are legally binding on the countries that sign up to it, and include detailed plans for Africa, Asia, Latin America, the Caribbean, and the northern Mediterranean. Almost 160 countries had ratified the Convention by September 1999. The focus has moved to implementation, despite significant questions about who pays, coordinates, and monitors the activities. By 2000 National Action Plans were drawn up by most dryland nations, but progress is hampered since there is no international CCD implementation fund available, so many of them are really just paper documents.


5. Conclusion


In the absence of well adapted technologies, life in drylands is harsh.  That degradation of these lands follows from social and climatic processes is not in dispute. But whether degradation amounting to desertification of the magnitude assumed in the early debates, or even in the recent Convention [CCD], has been strongly challenged by the research and other communities. Nonetheless, desertification remains a powerful narrative that has captured the imagination of certain policy makers, governments, and scientists, and it has persisted with peaks and troughs of interest since the 1930s. Its power is problematic, since “the search for accuracy appears to be vulnerable to generalisation, over-simplification and distortion for political or funding purposes” (Mortimore1998:33). Desertification has certainly served specific constituencies quite well: -

  • national governments seeking to extend their control,
  • international bureaucracies seeking legitimacy,
  • and scientists seeking research funding.


Tackling the various processes that the term desertification subsumes is a major challenge, given that dryland populations suffer the impacts of droughts and desiccation, and need to exploit land, water, and timber resources in ways which often result in degradation. As an explanatory term, however, desertification has been used too broadly and too loosely, and this has been counterproductive.



Key Reading


Charney, J.G & Stone, P.H. 1975. “Drought in the Sahara: a bio-geophysical feedback mechanism”. Science, 187: 434-445.


Helldén, U  1991 Desertification: time for an assessment. Ambio 20(8) 372-383.


Middleton, N.J. and D.S.G. Thomas (eds) 1997. World Atlas of Desertification 2nd edition. London: Edward Arnold.


Mortimore, M. 1998. Roots in the African Dust: Sustaining the Sub-Saharan drylands. Cambridge University Press.


Nicholson, S.E, C.J Tucker, and M.B Ba. 1998 “Desertification, Drought and Surface Vegetation: an example from the West African Sahel.” Bulletin of the American Meteorological Society 79 (5): 815-829.


Stebbing, E.P. 1935. The Encroaching Sahara: the threat to the West African colonies. The Geographical Journal 85: 506-524.


Swift, J. 1996. “Desertification.” In The Lie of the Land: challenging received wisdom in African environmental change and policy, ed. M Leach and R Mearns. 73-90. Oxford: James Currey.


Tucker, C.J, Dregne, H.E, Newcomb WW. 1991. Expansion and Contraction of the Sahara Desert from 1980 to 1990. Science 253: 299-301.


Warren, A. 1996. “Desertification.” In The Physical Geography of Africa, ed. W.M Adams, A. Goudie, and A.R. Orme. 342-355. Oxford University Press.


Williams M.A.J. &  R. C. Balling, Jr., 1996. Interactions of Desertification and Climate. London: Edward Arnold.