At an abstract level, the definition of desertification offers a number of problems for clear and objective understanding. Firstly, rather than a unitary process in itself, desertification is an amalgam of drought, desiccation and degradation which are three interrelated but discrete phenomena (Warren, 1996). Each of these has different causes and feedback mechanisms but has a similar outcome, and therefore desertification creates operational difficulties for assessment and subsequent intervention (ibid., 1996). Pinpointing the exact cause of similar short-term physical manifestation is important to provide an effective solution (Warren & Khogali, 1992). The inclusion of a subjective causal element such as human pressure in the definition also leads to problems delimiting the identification of desertification from its measurement (Mortimore, 1998). For example, in its attempts to map desertification hazards in 1977, UNEP added the parameter of high human and animal pressure to a map of aridity (ibid, 1989). However the duality of human action and natural changes is not necessarily so distinct and defined, and this obscures the mutual constitution of the two (Thomas, 1997). Secondly, and possibly as a result of this complexity of process, a definition of desertification has been problematic (ibid., 1997). As noted earlier, there are more than 100 published definitions of ‘desertification’, which variously include or exclude the range of ecological processes within its boundaries (Glanz & Orlovsky, 1985; Swift, 1996). Mainguet (1991) has argued that the term has been rendered obsolete by the extent of the confusion over its meaning. At a purely conceptual level therefore, the usefulness of the desertification may be severely compromised by the ambiguity over its definition, and within this the inclusion of both causal mechanisms and identification factors.
Possibility related to the problem of definition of desertification, the theory and evidence used to support the concept has also been questioned (Mortimore, 1998). In this section the accuracy of the received narrative will be shown to have been challenged on three bases, the physical form of its occurrence, the processes that lie behind this form and the structural mechanisms that account for the continuation of these processes. The history of the narrative has been comprehensively reviewed by Swift (1996). Analysis of the process (without reference to the term) of desertification began in earnest in the late 1920s and 1930s, following a period of severe drought in the Sahel. One of the most influential writers on the subject was Stebbing, a forester who carried out fieldwork in the British and French colonies of West Africa in 1934. Stebbing asserted that the Sahara was moving southwards and estimated that this had occurred at a rate of 1km / yr for the previous three centuries (Stebbing, 1937 cited in Swift, 1996). Desiccation was ascribed as the process responsible for this change, with the increase in the use of indigenous agricultural land use through population pressure as the structural mechanism (ibid, 1937). This process was thought to have a subsequent feedback effect on rainfall, making it more intermittent (ibid, 1937). Although Stamp (1940) raised the severe shortcomings of Stebbing’s research, a number of elements of his argument can be traced in current consideration of desertification. In the 1970s and early 1980s, the desertification issue was revived, once again following a period of intense drought (Swift, 1996). The work of Lamprey (1975) and Ibrahim (1984) proved particularly important during this period, despite seriously flawed data analysis (ibid, 1996). Lamprey used ground surveys and aerial reconnaissance to compare contemporary boundaries on Kordofan and Darfur with those of a 1958 botanical survey. However, this distinction did not take account of the fact that the contemporary data was taken in a series of exceptionally dry years, whilst that 1958 represented a relatively wet period in the Sahel (Mortimore, 1998). In terms of the physical manifestation of this process, Lamprey (1975) concluded that ecological boundaries were shifting southwards and that sand was encroaching and threatening farmland. The work of Ibrahim (1984), based on observations in Darfur, Sudan over the period of 1976 – 1982, supported Lamprey’s conclusions and suggested that desertification had claimed over 650,000 km² of productive land in the previous fifty years. In both of these studies, man was seen as the cause of desertification, with the rate of the process controlled by the increase in population pressure. Actions such as excessive use of agriculture in unsustainable areas, extensive pastoralism and high herd numbers and the indiscriminate collection of firewood were suggested as the triggers for this process (Swift, 1995). In combination the ideas of Stebbing (1937), Lamprey (1975) and Ibrahim (1984) became the scientific basis of a received narrative in terms of the form, process and causation of desertification. These ideas were reflected in policy circles, culminating in the UN Convention on Desertification in 1977. The main elements of the desertification concept may be thus identified as the expansion of the desert at a measurable rate, with desiccation is main process of this change and that man (driven by population pressure) is its agent. A Malthusian spiral of population growth, resource mismanagement, environmental degradation and a reduced carrying capacity may then be extrapolated (Mortimore, 1998). Many aspects of this conception have been the subject of critique.
Any predictive and diagnostic model of environmental change is dependent on the accuracy of its data and the soundness of its premises (Mortimore, 1998). In terms of form (or physical manifestation of the process of desertification), the received narrative promoted images of moving deserts and the southward movement of the Sahara (Swift, 1995). Yet there is considerable evidence to suggest that rather than a linear encroachment of such conditions, desertification occurs at particular points (Bernus, 1977 cited in Mortimore, 1998). The limitations of the method of analysis used by Lamprey (1975) and Ibrahim (1984) have been further exposed by subsequent studies. Following a series of investigations by Lund University throughout the 1980s, Helleden (1991) was able to state that;
‘none of these studies verified the creation of long lasting desert-like conditions in the Sudan during the 1962-1984 period… there was no trend in the creation or growth of desertification patches around 103 examined villages and water holes over the period 1961-1985. No major shifts in the northern cultivation limit were identified… [and there were] no major changes in vegetation cover and crop productivity which cannot be explained by varying rainfall characteristics.’
Similar results were evident in the Manga Grassland survey by Mortimore in 1989. By comparing aerial plots of the dunes over the period 1950 –1969 with the boundaries delimited by the Forestry Commission Survey in 1937, he concluded that many of the basic characteristics of the area showed continuity over time (Mortimore, 1989). Also this time span covered a period of considerable settlement, agricultural intensification and expansion leading to doubt over the basic hypothesis of desertification (ibid, 1989). In terms of a process, it has already been noted that desertification may be more usefully considered with reference to its individual constituents of desiccation, drought and degradation (Warren, 1996). But a further point should also be raised here. Namely, that the desertification is often perceived as a disruption to a stable, equilibrial natural system. There is considerable evidence to suggest adequately represent dryland environments; they are unstable and disequilibrial in the short term and transitional in the longer term (Mortimore, 1998). In terms of both the form and process of desertification, the accuracy of its premises and data may be found wanting on both counts. The utility of the concept may be further questioned when the its structural causation mechanisms of population growth is considered.
The very definition of desertification automatically limits the conceptualisation of dryland sustainability, through the inherent assumption of the failure of human management systems to cope with increasing population pressure (Adams, 2003). Within a desertification narrative therefore, there is little room for the possibility of adaptation and flexibility of management techniques and practices by ordinary people (Mortimore, 1998; Adams, 2003). The dominance of large-scale studies that have an emphasis on quantitative analysis (such as remote sensing) rather than micro-scale perspectives that focus upon the social science aspect of the problem may explain this omission (Mortimore, 1998). A number of studies in the last decade have sought to de-link the implicit connotation of population growth and environmental degradation that has been central to the desertification narrative. Such analyses draw on the ideas of Boserup (1965), suggesting that increasing population pressure can provide the stimulus for innovation and agricultural intensification, for example through increased cropping intensities and the introduction of land saving techniques. Tiffen et al. (1994) examine the case of the Machakos District in Kenya, where there has been considerable concern over the sustainability of agriculture since 1930s colonial administrators attempted to implement soil conservation measures. They used a variety of historical and current sources, such as oral history, to undertake the study. They show that increasing population densities have facilitated more productive agriculture and greater specialization and exchange within society (ibid, 1994). Specific strategies include migration, the diversification of incomes (including non-agricultural incomes) and agricultural intensification (ibid., 1994).The area cultivated increased from 15 percent of the district in the 1930s to between 50 and 80 percent in 1978, and the land supports a population that has grown almost fivefold, from about 240,000 in the 1930s to about 1.4 million in 1989 (ibid, 1994). The photographs of Kiima Kimwe in 1937 and 1991 (below, left and right respectively) clearly illustrate the use of careful terracing and subsequent increases in productivity through the planting of banana and other trees (Drylands Research website, 2003).
Tiffen et al.’s (1994) study illustrates how local communities can respond spontaneously to land degradation and make land improving investments that significantly increase productivity over time. Applying the desertification framework in this situation would be of little utility in the explanation of population growth concurrent with continued or even improved prospects of sustainability.
Incorporating the idea of sustainable livelihoods and of social, human and human-made capital may be a further help to examining what the concept of desertification has missed through its biophysical sustainability bias (Serageldin, 1996). Such ideas open the possibility for a number of other inputs that may compromise, or indeed uphold, the sustainability of dryland production systems. An analysis of the social system in dryland production can point to the need for a sustainable social as well as natural system for the continuing use of the environment. Through the integration of this perspective, Murton (1997) is able to question whether Tiffen et. al’s (1994) ‘these examples of sustainable resource use have been compatible with the maintenance of sustainable livelihoods in such marginal African environments’ such as the Machakos. Murton’s research (1997) adds further dimensions the consideration of dryland production systems, including a requirement to consider how polarization and global markets can also impact upon the sustainability of this environment. The integration of the complex social and economic adjustments that embody the everyday decisions of local people has considerable potential to explain the disjuncture between the ‘doomsday’ predictions of desertification narratives and small-scale evidence on the ground (Mortimore, 1998).
An analysis of the history of the concept of desertification can easily lead to conclusions about how science ‘got it wrong’, with a consequent attribution of blame which is all too resonant with earlier desertification narratives (Thomas, 1997). A more thorough consideration will recognise that science necessitates the constant refinement and evaluation of ideas by default (ibid., 1997). This points to the need to ensure the transmission of uncertainty at the science-action interface and a careful reconsideration of how scientific concepts can be taken selectively or used out of context (ibid, 1997). In this way, the legacy of the desertification narrative may yet prove useful as an important reminder of the differential needs of science and policy and the need for a more cautious approach to scientific ‘truth’ and ‘objectivity’. This has been neatly conceptualised as the tension between models of environmental change ‘as heuristics or truth machines’ by Wynne & Sackley (1994, cited in Mortimore, 1998). From a slightly different perspective, an understanding of desertification may be considered critical precisely to move beyond it (Swift, 1996). Until the ghost of the received narrative is laid to rest in national governments and in major NGOs, the deconstruction (versus the understanding) of desertification will be key to the comprehension of dryland production systems (ibid., 1996).
In conclusion, the narrative of desertification may be considered as particularly unhelpful to an accurate understanding of the many facets of sustainability in dryland production systems. Definitions of the terms are problematic, contested and confused, leading to problems for clear and concise communication on the topic. Moreover, the scientific evidence and data upon which the narrative is premised has been shown to be seriously flawed and also coloured by ignorance and prejudice towards indigenous livelihoods and technologies. As such the consideration of dryland sustainability in the framework of desertification may be seen to incomplete and also misguided. However, this is not to say that credible work on drylands has not been performed, nor that real environmental problems do not exist in these ecosystems. Although the term has continued to be adopted in policy circles, the use of an alternative, such as dry land degradation, may prove useful in the longer term and particularly when trying to identify effective interventions. Knowledge of the desertification narrative however, may be seen to provide an important reminder of the need to actively manage the use of science as a basis for policy, particularly when in complex issues that contain a substantial element of uncertainty. An analysis of the way in which powerful institutions have harnessed the power of the desertification narrative is also important for its deconstruction and for the possibility of its succession by a concept that is more attuned to the real and substantive issues of dryland sustainability.
References
Adams, W.M (2001) Green Development: environment and sustainability in the Third World. Routledge: London
Adams, W.M & Mortimore, M.J. (1997) ‘Agricultural intensification and flexibility in the Nigerian Sahel’ Geographical Journal 163:150-160
Drylands Research Organisation Website (accessed 19/11/2003) ‘The Machakos Study’ (available online at )
ICIHI (1986) The Encroaching Desert: The Consequences of Human Failure – A Report for the Independent Commission on International Humanitarian Issues. Zed Book Ltd: London.
Mainguet, M. (1991). Desertification. Natural background and human mis-management. Berlin:
Springer-Verlag.
Mortimore, M. (1998) Roots in the African Dust- Sustaining the sub-Saharan drylands. Cambridge University Press: Cambridge.
Murton, J., 1997, 'Sustainable livelihoods in marginal African environments? the social and economic impacts of agricultural intensification in Makueni District, Kenya’, Conference Proceedings, Sustainable Livelihoods in Marginal African Environments, Sheffield University, April 1997
Swift, J. (1996) ‘Desertification: narratives, winners and losers’ in Leach, M. & Mearns, R. (eds.) The Lie of the Land: challenging received wisdom in African environmental change. James Currey/International African Institute: London
Thomas, D.S.G. (1997) ‘Science and the desertification debate’, Journal of Arid Environments (1997) 37: 599–608
Thomas, D.S.G. (2000) ‘Desertification’ in The Dictionary of Physical Geography. Blackwell: Oxford.
Turner, B.L., Hyden, G. and Kates, R.W., (eds), (1993) Population Growth and Agricultural Change in Africa, Gainsville: University Press of Florida
Tiffen, M., Mortimore, M. and Gichuki, F., 1994, More People, Less Erosion: Environmental Recovery in Kenya, Chichester: Wiley
Warren, A. (1996) ‘Desertification’ in W.M Adams, Goudie, A.S. & Orme, A.R. (eds.) The Physical Geography of Africa. Oxford University Press: Oxford.
