Contents - Previous - Next

This is the old United Nations University website. Visit the new site at

1. People, land management, and environmental change: The problems that a United Nations University programme is studying

Harold Brookfiel

Intensification and innovation
A case in East Africa
The management of degradation: A west African case
A case of stagnation in Papua New Guinea
Discussion and conclusion


The project described in this introductory paper focuses its attention on one specific part of the whole complex relationship between people, their society, and the land. We concentrate on small-farm regions of the tropics and subtropics and, within this, on how farmers manage or mismanage their land and its biological diversity, and on why they do so. Some such areas are well managed, and some suffer severe degradation. Dealing with small-farm regions, we encounter great diversity of farming practices, often within quite small areas, and we term this "agrodiversity" (Brookfield and Padoch 1994). Although PLEC (People, Land Management, and Environmental Change) includes some experimental research based on farmers' practices, it draws information principally from what farmers do and know, and from its own field research. PLEC is also a capacity-building project, with a main objective of creating a multidisciplinary network of specialists in environmental management, qualified also to understand the societal forces which create the conditions within which small farmers manage their land and biota.

We do not try to cover the whole of the tropics and subtropics, but instead concentrate our efforts on a few areas where these problems can be studied in greater depth. Six "clusters" of PLEC scientists have been formed, based in universities or government research organizations, in Africa, Asia-Pacific, and tropical American regions. They are linked together by the core project network. Four clusters work only in humid areas but two, in West and East Africa, extend into the dry subhumid margins. In mid-1995 there are 70 scientists in PLEC, 53 of them working for institutions in the developing countries. The distribution of the clusters, and the areas in which they work, is shown in figure 1. A good deal more about some of these clusters and their work is presented in the following papers.

To explain better what we are doing, we sometimes use supplementary titles for our project. One such is "Agrodiversity, Land Management, and Biodiversity." We argue that the greater part of the conservation problem, whether of land or biodiversity, lies in agricultural areas where crisis conditions have widely been generated by population growth, commercialization, deforestation, and land degradation. We urge a positive rather than a negative view of this situation, and propose that one of the most undervalued elements is the ability of a significant proportion of the world's small farmers to modify their farming systems in a comparatively short period of time, often in ways that make sustainable use of biodiversity as well as the land. Their object is production and not conservation, but an objective of sustained production does call for conservationist methods, and many farmers have or acquire the knowledge to adopt such methods and do. Going further, we also seek to persuade that farmers' adaptations and the knowledge on which they are based could often be successful, if societal conditions permit and assist. People manage the land and its biological diversity; moreover, they suffer the consequences of wrong decisions. One principal objective of PLEC is to give farmers' knowledge and practices an equal place with the findings of external science in the search for sustainability with conservation.

This paper does not describe the work of the project, though some of this is used, but discusses a part of the background. This paper looks at the specific question of intensification and innovation in land-resource management. It emphasizes some recent African material. PLEC argues that farmers can and do change their systems of management when pressures arise, and there is already abundant evidence to support this from within the project, as well as from a wider literature. My aim is to look outside PLEC's own evidence, and put it into a wider context. To discuss this question, I draw initially on two papers which I wrote about a decade ago, and which are here reconsidered (Brookfield 1984, 1986). This short paper is part of a rethinking process that I hope to develop further in other writing now under way. The reconsideration presented here is in the light of some recent literature concerning the way in which land management has adapted, with considerable success, to increase in population and to the need to manage land degradation. Some of this adaptation constitutes reclamation of degraded land and this aspect, not stressed in the mid-1980s' papers, is given greater weight here.

Intensification and innovation

To begin, it is necessary to define some basic terms. Intensification is properly measured against constant land, and means measures which will enhance productivity on that land. More specifically, what is usually implied is the intensification of inputs, meaning their increase on constant land up to - or even beyond - the point of zero marginal return. In most of the literature the inputs that are discussed are those of labour, since increases in capital input are generally "lumpy," and less easy to analyse. However, where changes are introduced into an agricultural system the inputs are applied in qualitatively new ways, and a new "curve" of intensification is originated. Such changes are best separated from progressive intensification, and thought of as innovations.

Figure 1 - The Clusters of PLEC

Innovations, in turn, need to be distinguished into those which are specific to a particular crop or crop cycle, and those of greater durability. The latter may be purely organizational, such as a change in land tenure arrangements, although these normally accompany changes in the farming system. A major set of innovations takes place when, for example, a mainly pastoral system becomes dominantly agricultural. Certain innovations create capital in the land, often termed "landesque capital." Mounds, ridges, terraces, irrigation and drainage systems, dams, created land and created soil, agroforests, fishponds and the like, constitute such capital. Their physical evidence may remain on the land after abandonment of upkeep, and some relinquished innovations may remain capable of being brought back into service after many years. Most successful physical innovations have the effect of changing the conditions under which soil fertility, structure, and soil water are managed, or of reducing or minimizing erosion from sloping land. They are changes in the direction of greater sustainability of production.

All innovations are not equal. Some are of only a minor nature, though they may be important. An example might be the deliberate fostering of a particular tree species as a part of general fallow management, improving nutrient recycling along with other benefits. Another might be the introduction of row-planting of rice in place of more random planting in a wet field, to facilitate weeding with higher planting density. Major innovations, however, include the terracing of hillsides, the introduction of a system of irrigation, the placement of mulch in mounds or under ridges, or anything which creates a major change in the system of management. The distinction is easier to draw conceptually than in practice but it is useful because major innovations are those which mark a real change in the conditions of production and, often, the creation of enduring landesque capital.

The relationship of major to minor innovation and of both to intensification is fairly easily understood in the case of wetlands. Wetlands may be agriculturally usable when water levels are low without any special works at all, but crops are at risk from unexpected variations in water level. Production is insecure. Once the wetlands are drained, or ridged, however, dry land is created. The effect of the major innovation is an immediate boost in crop security and often in per-crop yield as well. Such a system may then be elaborated with new minor innovations such as varying the height of the ridges, and working them more intensively by applying more labour. The initial transformation of the land surface was, however, the major innovation, greatly changing the conditions of production. There may be further major innovations, for example involving management of the water level. However, most subsequent elaborations are of a minor order, accompanying intensification of inputs into the new system.

It is common to see major innovations on difficult sites that could not have been managed sustainably without them. Wetlands and steeplands can both be used without innovation, and so can difficult soils, but their continued productivity cannot often be achieved without some form of physical transformation of the site. There are few steep hillsides that can be worked for a long period without loss of soil, unless there is some form of slope protection. While on naturally resilient slopes the problem may be considerably less than many observers, seeking to introduce high-technology management, may have believed, there are always advantages in the introduction of protective measures. However, it is not only in such obvious locations that innovations are important. Where land has been cleared from forest, for example, considerable changes take place in its micro-climate and, where the general climate is characterized by drought, the effect of this drought may be greatly intensified on treeless land. Innovations both to adapt agriculture for greater drought tolerance, and to restore some tree cover, may therefore be highly advantageous.

Rarely are innovations introduced all at once, unless by major external intervention. Both logical reasoning and observed experience show that they demand farmers' experimentation, either spontaneous or using new information they have received. Sustained and widespread adoption only takes place after clear advantages have been demonstrated. So little has the adaptive capacity of farmers been valued during the past two or three generations of development research and intervention, that the view of farmers as experimenters has until very recently indeed been derided, and it is still derided in some areas. Yet farmers experiment constantly, with new crops, new trees, new tools and practices, and new ideas. Many experiments, even if successful, are allowed to lapse, but a proportion survive and are more widely adopted. It is this proportion of enduring innovations, repeated across region after region, that is responsible for the enormous diversity of farming systems found across the world.

There is, however, a problem with some systems which involve a large inheritance of landesque capital from an earlier period of time, even a very recent period. The maintenance of this capital can be an onerous burden, absorbing a substantial amount of labour input, and income, that cannot therefore be devoted to current production and income-gaining. Modern change usually includes substantial growth of off-farm employment, whether or not it involves emigration from rural areas. The employment is often selective of young people, and the farm population grows older. The effect of a heavy dependence on landesque capital is most obvious in the intensive hydraulic systems of Asia, with their heavy dependence on irrigation, terracing, and their works. Writing of China, but with much wider potential application, Elvin (1993/4) writes of "technological lock-in," which can lead swiftly from apparent sustainability to unsustainability. While this is not yet a problem in the specific cases discussed below, it is already there as a threat.

One of the reasons why farmers' experiments have been so little recognized is that most external observers remain in a farming region only from hours to, at most, a very short period of years. Yet change in farming systems takes place across periods of decades and it is only now, when modern observations can be set against those of from 20 to 60 years ago, that the amount of largely spontaneous change can clearly be seen. It could have been realized sooner, by logical thinking about what must have taken place to produce present management systems in the many areas where whole new crop complexes have been introduced by transfers between continents within the past one to five hundred years. Except in an academic literature, however, the necessity to accept farmers' experimentation as a reality did not arise. And a good deal of the academic literature was also put aside as irrelevant by the practitioners of "development," or was used only in a highly selective fashion. One of PLEC's secondary objectives is to strive for larger incorporation of the results of good research into diagnosis, management, and development practice.

A case in East Africa

A useful body of literature has emerged in recent years on the modification of African agriculture, principally in the face of rising population pressure. In what follows, I concentrate heavily on a small part of this literature. A considered official view is also the classic view that, while African farmers may have the ability to change their farming systems, the process is too slow to cope with the rate of contemporary population growth (Cleaver and Schreiber 1992). A recent group of academic studies leads to a less negative conclusion, noting a considerable amount of successful adaptation, but leaves the case regarding population growth open (Turner et al. 1993). Several of its contributors still regard it as "inevitable" that continued growth of population will lead ultimately to land degradation. Very different is the basic conclusion of a detailed study of change in the Machakos district of Kenya, which concludes that increasing density of population over most of the present century has been a major force for improved environmental management (Tiffen and Mortimore 1992; Tiffen et al. 1994). Specifically, increased demand and closer interaction generated by a denser population, coupled with increased labour supply and economies of scale, have more than outweighed the negative effects of population growth. After a six-fold increase in population over 60 years, dryland farming in a drought-prone subhumid region "has been turned toward a new logic of sustainability" (Tiffen et al. 1994, p. 12).

There is more to the Machakos case than this, however. The processes of change involved first a progressive, indeed rapid, degradation which endured through the 1940s and later in some areas. Its most obvious, and best described, characteristic was soil erosion, but deforestation, declining soil fertility, and declining availability of water were also important elements. Already in the 1930s, however, improvement was recognizable in some areas, accompanied by a shift from largely pastoral farming supplemented by rotational cultivation toward mixed farming in the wetter areas, and with the beginnings of terracing and water control. External intervention was strongly directive and was concerned primarily with soil conservation, a pattern which continued until the 1950s. The imposed innovations were only selectively accepted by the farmers. Their family structure was at the same time shifting from extended toward nuclear families, more closely identifying the individual household with a specific tract of owned land, so that restorative management arose increasingly at the family-farm level, and not over whole landscapes. From the 1950s onward colonial government discouragement of cash cropping was replaced by encouragement of an already strong trend toward planting of coffee and other commercial crops and, coupled with incomes earned off the farm, cash crops began to provide money capital for investment on the land. Top-down intervention began to be replaced by persuasive agricultural extension by a service which continued to be strong until the end of the 1970s.

Positive change accelerated in the 1960s and has continued. It has included extensive terracing by different methods adapted to soil and slope conditions, cut-off drains to divert and conserve water, heavy though still insufficient use of livestock manure on the farms, adoption of a suitable plough, cash cropping of growing diversity, substantial increase in off-farm employment and business and, almost wholly by indigenous innovation with only highly selective adoption of external advice, the evolution of complex systems of agroforestry (Rochelau et al. 1993). By the 1990s, steeplands in central Machakos which were bare and eroded in the 1930s are well treed, intensively terraced, and managed by a diverse range of strategies. However, there are still clearly visible contrasts between individual holdings in the application of skills and labour to sustainable management.

Erosion is greatly reduced, even in the drier lower regions, though it has not ceased. Decline of soil fertility, due to soil loss and inadequate replacement of nutrients, also continues, although far less on the better-managed farms. Tree cover has increased almost everywhere, and in the more densely peopled areas it consists almost entirely of useful species, a high proportion of them exotics. Densities in some areas are now above 400/kmē, and population is still increasing at a high rate, but the on-farm density is now growing less fast because of the growth of off-farm means of gaining a living. However, land is now very scarce, encouraging the adoption of all possible methods to raise family-farm productivity. The net effect is an increase in per-capita production and incomes, and a great improvement in prospects for the future. All this has been achieved by a fruitful mix of externally derived and endogenous innovations, large and small, and the pattern of progressive change has not ceased.

Machakos is exceptional, but not unique. There are areas in Kenya and elsewhere in East Africa where management is yet more intensive, supporting still higher population densities in a sustainable manner (R.M. Kiome pers. comm.). But there are other areas in which management continues to leave much to be desired. Machakos has some special characteristics, including its proximity to the national capital with its large consuming population, as well as a good internal infrastructure. None the less, there are some important lessons of wider application to be derived. While externally derived innovations have been of major importance, they have been filtered through farmers' own perceptions of rationality, and several major innovations have been adopted without official encouragement. The most important lesson is that the capabilities of indigenous farmers should not continue to be undervalued, as they have been in the past.

The management of degradation: A west African case

By a wholly realistic interpretation, much of what has taken place in Machakos is the reclamation of a degraded landscape. There are large areas of degraded land in Africa as in other regions of the world. In many of them the fruitful combination of external intervention with spontaneous innovation which has characterized most of the Machakos story is absent. Government intervention, like international advice and aid, has been heavily oriented toward the production of commodities for the export market, and little assistance is given to farmers who lack the conditions for production of such commodities. The late-colonial concern with soil conservation, which was the true origin of most of the useful direct interventions in the Machakos case, was never strong where soil erosion was not a visible problem. Concern with deforestation, active in the post-colonial period and very popular today, has in most cases not got beyond the stage of unenforceable prohibition of clearance and use of fire.

Farmers' capability, and willingness to experiment, are however widespread elements in the situation. A growing literature is now emphasising this fact for policy makers, but it is very recent, and the pioneering writing of Paul Richards in West Africa is not yet overtaken (Richards 1983a, 1983b, 1985). If farmers have a well-integrated system of resource management developed in the past, and still evolving, they can make successful adaptations to change, even enhancing the quality of landscape and biodiversity. Such a case is described in Guinea, West Africa, by Leach and Fairhead (1994,1995). Valley swamps have been managed and enlarged for wet rice, large woodlands within which forest species have also thrived have been planted around villages, and the savanna has been mulch-mounded for cassava and has become more woody in consequence. Population growth has been accompanied by regression of the savanna, not by its enlargement. This conclusion runs in the face of still-widespread belief that savanna is everywhere extending at the expense of forest. Yet the conclusion is incontrovertibly supported by the comparative use of old and modern air photographs, supplemented by remote sensing imagery, and by descriptions found in the archives and in the memories of the farmers and their families.

Although there are also parts of West Africa in which intensive farming in subhumid areas has been notably successful, it does not seem that this is the case in much of Ghana. Large areas of semi-deciduous forest, once under-planted for cocoa and oil palm, have now become "savannized" and farmers experience declining yields, diminished availability of firewood, and inability to cultivate the more demanding crops that were common only one or two generations ago. Government support in both colonial and post-colonial times has strongly emphasized export cash crops, and continues to do so. Farmers face land degradation largely on their own.

Two recent studies by PLEC members, in adjacent areas of south-eastern Ghana, present somewhat differing interpretations. This is a region across which waves of pioneering land settlement have passed during some two centuries, almost from the outset strongly associated with production for the world market. Only the most naturally resilient parts of Ghana's original oil palm and cocoa heartlands remain in good productive condition, now for food crops grown for the urban market. Other areas, penetrated first for oil palm and then for cocoa farming, mainly in the present century, now suffer widespread deforestation, biodiversity loss, and impoverishment of soils. In the Mampong valley and adjacent areas, Gyasi et al. (1995) find farmers' attempts to modify cropping patterns and intensifying production thwarted by poverty and technological constraints, and by the high proportion of tenancy among the farmers which inhibits innovation and encourages over-exploitation. Commercial pressures, and rising population pressure, create a potentially critical situation.

In the adjacent Manya Krobo district, Amanor (1994) encounters many of the same phenomena and a very similar history, but also sees more positive elements. Tree selection is of major importance, as is the introduction of leguminous crops. Perhaps looking more specifically for innovation, Amanor finds it among a minority of farmers who introduce comprehensive management of the fallow so as to obtain better crops after it, or have developed an agroforestry system based on a particular tree (Newbouldia laevis). This tree has long been known to provide the right amount of shade for undercropping and to create a more favourable soil environment. Farmers' agroforestry treatments, spontaneous but very similar to those advocated by external specialists for use with introduced trees and shrubs, further enhance this environment for crops. A number of other experiments, by individual farmers, were uncovered during field research. Furthermore, the main centres of innovation appear to be in the most degraded parts of the region, where established production systems work least well, where there is longer experience of degradation and perhaps where the remaining forest stock is reduced to its hardiest species, assisting farmers to harness the remaining energies of the forest system. Over quite a small area, considerable variety has evolved out of an earlier greater uniformity of agroecosystem type. None the less, the management of degradation is still only marginal in the region as a whole.

A case of stagnation in Papua New Guinea

The author has worked for more than a decade with people and their agriculture in the highlands of Papua New Guinea, in a densely peopled area known as Chimbu where an elaborate farming system, complex and adaptive to different ecological conditions, was developed over some two hundred years in a pre-colonial period that ended only in the 1930s. With return visits, he has data on land use extending over a 26-year period from 1958 to 1984, with subsequent observation up to 1990 (Brown et al. 1990). During this period the population of the area specifically studied has grown by more than 50 per cent, but innovation tailed off rapidly after the first years of observation. During those early years coffee was introduced as a cash crop and by 1960 had already been concentrated on the most suitable land, while the regionally ubiquitous casuarina tree, which fixes nitrogen, had quickly replaced introduced exotic shrubs as shade. A number of related changes had been made in the farming system, though hindsight shows these to have been related as much to the beginning of concentration of settlement into quasivillages, with family residence taking the place of gender separation, as to the coffee innovation itself. After the mid-1960s, however, coffee development reached a plateau, other new crops came and went, while the basic farming system scarcely changed at all. All the population conditions that led to innovation in Machakos were present, but innovation in Chimbu virtually ceased.

The stagnation has social and political causes, for this has become a depressed region. However, two additional conditions which were present in Machakos have been absent in Chimbu. A very sustainable set of land management adaptations was already developed in pre-colonial times and, moreover, they continue to give successful results notwithstanding the rising density of population. While desire for higher incomes, and greater participation in economic opportunity, have been effective in bringing about change in some other parts of Papua New Guinea, in Chimbu there has been an absence of sustained opportunity, while there has also been no significant land degradation in modern times to act as a spur. This is of interest in the general context, because it supports a view that population growth is not a sufficient cause of either land degradation, or improved land management according to which theoretical position one adopts. Other elements need to be present in all cases.

Discussion and conclusion

Diverse management of diverse landscapes by small farmers is more than an inheritance from the past. It continues, and evolves. But adaptive diversity is also at risk, for both modern science and the pressures - and attractions - of commercialization have acted to reduce agrodiversity and replace it with simple, widely replicated systems of farming. Often, these modern systems are destructive of the biodiversity sustained under agrodiversity, not least the biodiversity of the soil. They may, at least for a time, support more production, but they manage the land less well. PLEC seeks to understand and explain these small-farmer adaptations, whether they are "traditional" or not. Indeed, the adaptations rather than the old systems themselves are our main focus of interest, for these adaptations are dynamic and responsive to change in external conditions. And, because we study change, we are also concerned with the history of change, in agriculture, society, and environment alike. Although many have tried to do it, conclusions about change cannot be established without studying history.

There are other elements in our project, some of which are mentioned in other papers in the collection, but those discussed here are among our distinctive characteristics. We emphasise the study of agrodiversity and its value, the study of adaptation and change, and the history of change. With them, and through our clusters of scientists spread across the tropical world, we seek, by building a capacity for informed, action-oriented research, to join a much larger endeavour, that of finding paths to a sustainable future for the growing numbers of people in the developing lands.


Amanor, K.S. 1994. The New Frontier: Farmers' Response to Land Degradation. A West African Study. Zed Books, London.

Brookfield, H. 1984. Intensification revisited. Pacific Viewpoint 25, 15-44.

Brookfield, H. 1986. Intensification intensified: "Prehistoric Intensive Agriculture in the Tropics" (review article). Archaeology in Oceania 21:3, 177-181.

Brookfield, H. and C. Padoch. 1994. Appreciating agrodiversity: A look at the dynamism and diversity of indigenous farming practices. Environment 36:5, 6--11; 37 45.

Brown, P., H. Brookfield and R. Grau. 1990. Land tenure and transfer in Chimbu, Papua New Guinea: 1958 1984 - A study in continuity and change, accommodation and opportunism. Human Ecology 18:1, 21-49.

Cleaver, K.M. and G.A. Schreiber. 1992. The Population, Agriculture and Environment Nexus in Sub-Saharan Africa. The World Bank, Africa Region, Washington, DC.

Elvin, M. 1994. Three thousand years of unsustainable growth: China's environment from archaic times to the present. East Asian History 6 [1993, publ. 1994], 7-46.

Gyasi, E.A., G.T. Agyepong, E. Ardayfio-Schandorf, L. Enu-Kwesi, J.S. Nabila and E. Owusu-Bennoah. 1995. Production pressure and environmental change in the forest-savanna zone of southern Ghana. Global Environmental Change: Human and Policy Dimensions 5:4, 355-366.

Leach, M. and J. Fairhead. 1994. The Forest Islands of Kissidougou: Social Dynamics of Environmental Change in West Africa's Forest-Savanna Mosaic (with D. Millimounou and M. Kamano). Report to ESCOR of the Overseas Development Administration. Institute of Development Studies, Brighton, UK.

Leach, M. and J. Fairhead. 1995. Reading forest history backwards: The interaction of policy and local land use in Guinea's forest-savanna mosaic, 1893-1993. Environment and History 1:55-91.

Richards, P.W. 1983a. Farming systems in West Africa. Progress in Human Geography 7, 1-39.

Richards, P.W. 1983b. Ecological change and the politics of African land use. African Studies Review 26:2, 1-72.

Richards, P.W. 1985. Indigenous Agricultural Revolution: Ecology and Food Production in West Africa. Hutchinson, London.

Rochelau, D., K. Wachira, L. Malaret and B.M. Wanjohi. 1993. Local knowledge for agroforestry and native plants. In: R. Chambers, A. Pacey and L.A. Thrupp (eds.), Farmer First: Farmer Innovation and Agricultural Research, 14-24. Intermediate Technology Publications, London.

Tiffen, M. and M. Mortimore. 1992. Environment, population growth and production intensity: A case study of Machakos. Development Policy Review 10, 359-387.

Tiffen, M., M. Mortimore and F. Gichuki. 1994. More People, Less Erosion: Environmental Recovery in Kenya. John Wiley, Chichester, UK.

Turner II, B.L., G. Hyden and R. Kates (eds.). 1993. Population Growth and Agricultural Change in Africa. University Press of Florida, Gainesville.

Contents - Previous - Next