SMALLHOLDER KNOWLEDGE, SOIL RESOURCE MANAGEMENT AND LAND USE CHANGE IN THE HIGHLANDS OF SOUTHWEST UGANDA
By
CARY S. FARLEY
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1996
LIBRARIES UNIVERSITY OF FLORIDA Copyright 1996
by
Cary S. Farley ACKNOWLEDGEMENTS
This research was supported by a Fulbright Fixed-Sum Grant, which was coordinated by the USIS in Kampala, Uganda. Additional field research funds were provided by The Rockefeller Foundation and CARE/Uganda, while technical and logistical support was also provided by the International Center of Tropical Agriculture
(CIAT), the Makerere University Institute of Environment and Natural Resources
(MUIENR) and CARE/Uganda. CIAT/Uganda played a particularly supportive role throughout the duration of the research project.
A great number of people have encouraged me in this effort. Most importantly,
I would like to thank parents and brothers who have provided encouragement both spoken and tacit these many years. I also owe a great deal to my fellow researchers in Uganda
who provided moral support, critical analysis and encouragement: David Edmunds, Paula
Davis, Jake Reynolds, Patrica Spittal and Joy Tukahirwa. Many others cannot be properly acknowledged, nor can words adequately express the debt that I owe.
I would like to thank my committee members for their guidance, and for
providing me with the latitude necessary to conduct and complete this work according
to my own manner: my chair, Dr. Edward Malecki, my co-chair Dr. Abe Goldman, Dr.
Joann Mossa, Dr. Peter Nkedi-Kizza (Soil and Water Science) and Dr. Art Hansen
(Anthropology). A number of other individuals at the University of Florida supported
my efforts on the technical front: Laura Hoffman helped create many of the tables, Ryan
iii Poehling produced the majority of the maps and graphics, and Ken Mease assisted with the analysis of the survey data.
In Uganda, I owe a debt of gratitude to a vast and diverse number of people.
They include Dr. Eldad Tukahirwa, Dr. Derek Pomeroy and the secretarial staff at
MUIENR; Dr. Mark Marquardt and Dr. Emmanuel Nabuguzi at MISR; Dr. Charles
Wortmann, Dr. Louise Sperling, Martin Fischler, Dr. Soniia David and the staff at
CIAT; Kim Lindblade, Philip Franks and the staff at CARE, and Fred Kyayi at the
Makerere University Institute of Applied Economics and Statistics.
Finally, in Kabale and Kisoro Districts, I am deeply indebted to my field assistants Zinne and Placidia, and of course, to the smallholders throughout the highland region who answered my many questions, and allowed my to follow them over hill and dale.
iv TABLE OF CONTENTS
page
ACKNOWLEDGEMENTS iii
LIST OF TABLES ix
LIST OF FIGURES xi
ABSTRACT xii
CHAPTERS
1 INTRODUCTION 1
Overview 1 Problem Statement 5 Research Objectives 9 Presentation of Research 14
2 LITERATURE REVIEW 16
Population Growth 16 Agricultural Intensification 19 Cultural Ecology 26 Political Ecology 30 The Colonial State and Land Management 36 Recent Approaches to Land Management 38 Land Degradation and Soil Science 42 Smallholder Knowledge 47 Summary 55
3 UGANDA AND THE SOUTHWESTERN HIGHLANDS 58
Population Growth in Sub-Saharan Africa 58 Demographic Trends in Uganda 59 Population Growth 59 Infectious Diseases 60
v 1 1
Refugees 64 Agricultural Production in Uganda 67 Agricultural-Based Economy 68 Land-Supporting Capacity 71 Purchased Agricultural Inputs 75 The Highlands of Southwest Uganda 78 Natural Vegetation 79 Landforms, Geology and Soils 79 Climate 83 Agricultural Production Systems 87 Cropping System 87 Land Tenure and Land Fragmentation 90 Labor 92 Demographic Trends 95 Summary 98
4 RESEARCH SITES AND METHODS 101
Research Sites 101 Research Methods 108 Preliminary Research 108 Field Research 1Q8 Participant Observation 109 Unstructured and semi-structured individual interviews 109 Semi-structured group interviews 1 10 Questionnaire Survey 1 1 Soil Classification Survey 1 4
5 COLONIAL INTERVENTIONS IN THE HIGHLANDS OF SOUTHWEST UGANDA 117
Introduction 117 Labor Reserves 118 Agricultural Cooperatives 121 Resettlement Schemes 122 Land Management Initiatives 124 Soil and Water Conservation Programs 124 Smallholder Soil Knowledge and Land Management 128 Colonial Management Approaches 129 Summary 132
6 SMALLHOLDER SOIL KNOWLEDGE 135
Scientific Soil Classification Systems 135
vi Smallholder Soil Classification Systems 136 Background 136 Results 137 Discussion 141 Soil Catenas 142 Background 142 Results 142 Discussion 143 Comparison of Local and Conventional Soil Classification Systems 145 Background 145 Results 146 Discussion 151 Smallholder Perceptions of the Fertility and Erodibility of Local Soils 154 Summary 158
7 SOIL DEGRADATION AND SOIL RESOURCE MANAGEMENT 162
Introduction 162 Soil Degradation 163 Popular Perceptions of Soil and Land Degradation 165 Scientific Evidence of Soil and Land Degradation 168 Smallholder Perceptions of Soil Degradation 171 Smallholder Soil Resource Management Practices 175 Soil and Water Conservation By-laws 175 Input Utilization 184 Agricultural and Resource Management Concerns 189 Summary 193
8 LAND USE AND AGRICULTURAL CHANGE 197
Environmental Change in Southwest Uganda 197 Land Use Change 198 Recent Vegetational Change 199 Land Fragmentation 203 Fallow Change and Intercropping 212 Weed Management 216 Cropping System Changes 218 Summary 222
9 SUMMARY AND CONCLUSIONS 226
vii Summary 226 Smallholder Soil Knowledge 228 Soil Degradation and Soil Resource Management 229 Land Use Change and Smallholder Adaptations 231 Colonial influences 232 Change and adaptation 233 Conclusions 234 Theoretical Implications 234 Applied Implications 236 Methodological Implications 239 Recommendations for Further Research 240
APPENDICES 242
A MEAN MONTHLY RAINFALL (MM) FOR KABALE, UGANDA - 1918 TO 1993 242
B ANNUAL RAINFALL (MM) FOR KABALE, UGANDA - 1918 TO 1993 243
C QUESTIONNAIRE SURVEY 244
D KABALE DISTRICT SOIL AND WATER CONSERVATION BY-LAWS 257
E KISORO DISTRICT SOIL AND WATER CONSERVATION BY-LAWS 260
F FERTILITY CAPABILITY CLASSIFICATION SYSTEM 264
G CHEMICAL AND PHYSICAL PROPERTIES OF TOP SOIL SAMPLES 267
BIBLIOGRAPHY 272
BIOGRAPHICAL SKETCH 332
viii LIST OF TABLES
Table page
3.1 Refugees and Asylum Seekers Need of Protection and/or Assistance 65
3.2 Land Use Trends in Uganda 69
3.3 Uganda's Population - Supporting Capacity (People/Hectare, Year 2000) 73
3.4 Fertilizer Consumption in East Africa in Metric Tons 77
3.5 Population Growth in Southwest Uganda, 1969-1991 96
3.6 Characteristics of Rural Households in Kabale and Kisoro Districts 97
4. 1 Characterization of Research Sites 105
4.2 Survey Sample Framework 112
5.1 Soil and Water Conservation Measures 125
6. 1 Smallholder Soil Classification, Kicumbi Parish, Kabale District 138
6.2 Smallholder Soil Classification Nyarurambi Parish, Kabale District 139
6.3 Smallholder Soil Classification, Muramba Parish, Kisoro District 140
6.4 Comparison of Soil Classification Systems 147
6.5 Cross-Site Comparison of Soil Classification Systems 150
6.6 General Soil Fertility Ranking 155
ix 6.7 General Soil Erodibility Ranking 155
7.1 Smallholder Perceptions of Soil Degradation 173
7.2 Smallholder Perceptions and Use of the Soil and Water Conservation By-laws 177
7.3 Smallholder Use of Agricultural Inputs 185
7.4 Smallholder Agricultural and Land Management Concerns 190
8. 1 Tree Planting Practices 201
8.2 Renting, Leasing or Borrowing of Plots 205
8.3 Fragmented Landholdings 207
8.4 Advantages and Disadvantages of Multiple Plots 210
8.5 Smallholder Fallow Practices 214
8.6 Cropping Profile Changes 219
x LIST OF FIGURES
Figure page
1.1 Uganda 6
1.2 Kigezi District 7
1.3 Kabale District 10
1.4 Kisoro District 11
3.1 Kabale Landscape 81
3.2 Kisoro Landscape 84
4. 1 Kicumbi Parish 103
4.2 Nyarurambi Parish 105
4.3 Muramba Parish 107
6.1 General Soil Catena for Kicumbi Parish, Kabale District 144
xi Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy
SMALLHOLDER KNOWLEDGE, SOIL RESOURCE MANAGEMENT AND LAND USE CHANGE IN THE HIGHLANDS OF SOUTHWEST UGANDA
By
Cary Farley
August, 1996
Chairman: Dr. Edward Malecki Major Department: Geography
In the highlands of southwest Uganda, the Districts of Kabale and Kisoro are characterized by extensive deforestation, intensive cultivation, land fragmentation and
2 one of the highest rural population densities in the country (>250 people/km ).
Concern over soil erosion and declining soil fertility in the region was first expressed by the British colonial government in the 1920's. Despite the continued widespread assumptions of land degradation in the highlands, the frequency, magnitude and scale of soil-related problems, as well as their causes and the ability of smallholders to manage for them have not been well understood. For the purposes of this investigation, three sites of differing geology, soils and population densities were selected in Kabale and Kisoro Districts for an in-depth study of smallholder soil knowledge and land management practices.
xii The investigation revealed that contrary to popular opinion, smallholders possessed a well-developed knowledge of soils and an array of land management practices for a diversity of environments. Additionally, cropping profiles, soil
resource management and land use have also changed in response to an array of
influences, including population growth, coercive soil and water conservation programs and agricultural policies, changing food preferences and market opportunities. Many smallholders have adapted to these influences by increasing tree planting, intercropping and the use of short term fallows, as well as relying on the fragmented landholding system to manage for agricultural risk.
Smallholder-cited constraints to improved resource management include land, labor and tool shortages, soil erosion and fertility problems, uncontrolled cattle grazing, and gender-based dimensions of agricultural production and land
management. Future initiatives to improve land management in the highlands will need to build upon the diversity of soil knowledge and land management practices among smallholders, and reflect an understanding of the multiplicity of land
management constraints they face. Initiatives will also need to include smallholders in the development and implementation of any resource management plans, provide
smallholders with a variety of technologies and land management options to select
from, and support the flexible and dynamic dimensions of the smallholder agricultural production systems.
xiii CHAPTER 1 INTRODUCTION
Overview
Over the last few decades, the image of agricultural and environmental crises in Sub-Saharan Africa has become a common fixture in both the popular media and academic literature. Africa itself has frequently been portrayed as a continent reeling from the effects of civil war, drought, desertification, deforestation, disease, land degradation, food shortages and famine (e.g., Glantz, 1987; Harrison, 1987;
1 Mabogunje, 1995; Timberlake, 1985). More specific to this study, soil erosion and soil fertility loss are considered to be undermining the productive capacity of agricultural systems across the continent (e.g., Brown and Wolf, 1985; Dregne, 1990;
Lai, 1987, 1988a; Oldeman, 1994; WRI, 1994). The numerous and enduring agrarian and environmental "crises" have been ascribed to an imposing litany of problems and constraints whose causes are often multiple and interactive. The causes may include social (e.g., class or gender inequalities, ethnic conflicts), political (e.g., colonial influences, government corruption, war), economic (e.g., unavailable or unaffordable agro-chemical inputs and technologies, limited market opportunities,
1 For example, between 1965 and 1985 it is estimated that while total food production in Africa grew by 54%, per capita production actually declined by 12% (WRI, 1989); this decline is also asserted in FAO (e.g., Alexandratos, 1988; FAO, 1992) and other studies (e.g., Biwas, 1994; Brown, 1988; Morgan and Solarz, 1994).
1 2 poor institutional support, ineffective agricultural research), biological (e.g., rapid population growth) and physical (e.g., fragile environments, climatic change) factors.
More specifically, agricultural scientists, environmental conservationists and development workers alike have expressed concern over increasing land degradation in the highlands of East Africa (e.g., Ford, 1990; Getahun, 1991; ICRAF, 1993;
2 Messerli et al., 1988; Stone, 1992). These highland environments support many of the most productive agricultural systems in East Africa and some of the highest population densities in Sub-Saharan Africa (Grosjean and Messerli, 1988; Stone,
1992). In these environments, it is also widely assumed that current levels of agricultural production have been achieved primarily by expanding cultivation into forests and marginal environments, including wetlands and steep hillslope areas, and intensifying cultivation of existing fields by decreasing fallow periods, eliminating crop rotations, and increasing labor inputs (e.g., ICRAF, 1993; Stone, 1992). Soil
degradation, including soil erosion and declining soil fertility, has been widely cited as one of the primary consequences of the extensive and intensive agricultural production systems and a serious threat to sustainable agricultural production in the region (Clay and Lewis, 1990; Grosjean and Messerli, 1988; Stone, 1992).
2 The highlands include land areas that are above 1,500 meters in altitude and have an average annual rainfall of at least 1,000 millimeters (e.g., ICRAF, 1993). East Africa is comprised of the countries of Burundi, Kenya, Rwanda, Tanzania and Uganda (e.g., Berry et al., 1990). The term "Eastern Africa" includes the aforementioned countries and Djibouti, Ethiopia and Somalia. 3
In the highlands of East Africa, population pressures are considered to be one
3 of the primary causes of soil and land degradation. Smallholders in the region have been widely impugned for the environmental crises, and often characterized as being
incapable of addressing resource management problems without outside assistance
4 (e.g., Clay and Lewis, 1990; Stahl, 1993; Weismann, 1992). Smallholders, however, also often have limited access to fertilizers, pesticides, improved seed, and new technologies, and supporting adaptive agricultural research programs are generally understaffed, underfunded or simply non-existent (e.g., Richards, 1985).
Unfortunately, the analysis of smallholder agricultural production and resource management systems has tended towards the simplistic (e.g., that cultivation of steep slopes results in excessive soil erosion) and thus perpetuated the trend to specify smallholders as the primary perpetrators of environmental degradation.
Recent investigations into human-environment interactions have expanded the scope of analysis to include interacting social, economic and political causes, and the historical context of agricultural production and environmental degradation (e.g.,
3 For the purposes of this study, the primary differences between soil and land degradation are one of scale, e.g., agricultural plot (soil) : watershed (land). Soil degradation is described as the decrease of a specific soil's potential to support agricultural or ecological systems. Land degradation is defined as "the substantial decrease in either or both of an area's biological productivity or usefulness due to human interference" (Johnson and Lewis, 1995: 2).
4 Smallholders are defined as rural cultivators practicing intensive, permanent, diversified agriculture on relatively small farms in areas of dense population. The family household is the major corporate social unit for mobilizing agricultural labor, and it provides for a significant part of its own subsistence needs (Netting, 1993:2). Many smallholders also produce food (e.g., beans) and cash crops (e.g., coffee) for the market. 4
Blaikie and Brookfield, 1987a; Chazan and Shaw, 1988; Jaeger, 1992; Watts, 1989).
Additionally, it is increasingly acknowledged that the scientific understanding of
tropical environments, particularly soils, is not as comprehensive as it is for temperate
regions, and many of the agricultural and land management models developed in
temperate climates and extended to smallholders have been inappropriate for many
production systems in Sub-Saharan Africa (Richards, 1985; Lai and Sanchez, 1992).
Furthermore, tropical agricultural production and ecological systems are also
characterized by dynamic features and resiliency, and many smallholders have
demonstrated an ability to modify their production systems to cope with and adapt to
stresses of both indigenous and exogenous origin (e.g., Brookfield and Padoch, 1995;
Goldman, 1995; Neimeijer, 1996; Netting, 1993; Mortimore, 1989).
Correspondingly, smallholders across Sub-Saharan Africa have demonstrated a
wide range of agro-ecological knowledge, resource management practices and
production strategies specific to their local environments and production systems
(e.g., Netting, 1993; Richards, 1985; Reij, 1991). Although smallholders have
demonstrated socially differential (e.g., gender-based) knowledge and management
capabilities, as well as incomplete or inaccurate knowledge for certain subjects (e.g.,
reproductive cycles of insect pests) (Bentley, 1989; Carter and Murwira, 1994;
Fairhead, 1990), they are characterized by a diversity of experience and management
abilities which they might contribute to agricultural and resource management research initiatives (Chambers et al., 1989; Netting, 1993; Pretty, 1995; Scoones and
Thompson, 1994). Future efforts to improve the sustainability of the highland agroecosystems will need to recognize and support their inherent diversity and flexibility, and to develop, test and promote a wide range of versatile and adaptable agricultural and land management technologies in collaboration with the primary users: smallholders.
Problem Statement
The highlands of southwest Uganda (Figure 1.1) exemplify a region with high population densities, intensive land use, deforestation and widespread speculation of
land degradation. Land use in the highlands is characterized by intensive agricultural production systems, which include the cultivation of wetland valleys and steep hillslopes. As such, the region represents a microcosm of the natural and cultural processes at work in East Africa, and provides multiple settings within which to investigate agricultural production and land management systems, and to examine how, and in response to what influences, they have changed. This highland region constituted part of Kigezi District (See Figure 1.2.) during the period of British colonial rule (1913 - 1962), and was eventually sub-divided into the Districts of
Kabale, Kisoro and Rukungiri after independence (1962). For the purposes of this study, smallholder agroecological knowledge, soil resource management and land use change were investigated at three different locations in Kabale and Kisoro Districts.
Soon after the establishment of the colonial administration, district-level staff began to express concerns that population pressures and inappropriate land management practices were causing serious environmental degradation. In response to those concerns, the District Agricultural Office (DAO) introduced soil and water 6
Figure 1.1: Uganda
Source : EIU, 1995; GOU, 1967. Legend _ International Borders
Current District i ZAIRE Borders
— Major Roads u Kigezi District • Towns
0 8 rniles^ 4 _L J T T . . kilometers 0 4 g
Rukungiri District
Kabale * Kisoro I District District
Kabale 'Town Kisoro Town
RWANDA
Figure 1.2: Kigezi District
Source : Uganda Protectorate, 1957. conservation measures into the region in 1937, and formalized them as Soil and Water
Conservation (SWC) By-laws in 1945 (Martin, 1945a, 1946; Purseglove, 1945).
Presently, despite the colonial land management interventions and the continued administration of the SWC By-laws, there persists a widespread perception of environmental degradation in the highlands. Outside experts and DAO staff continue to assume that smallholder agricultural and land management practices are altering hydrologic responses in the watersheds and damaging the long-term productivity of the soils, and hence undermining the sustainability of the agricultural systems.
There has, however, been a dearth of scientific research undertaken to investigate the nature of environmental degradation in the highlands of southwest
Uganda, and the claims of soil and land degradation often have been exaggerated.
Although soil and land management problems certainly exist in the highland region, neither the magnitude, scale, frequency or causes of soil and land degradation problems, nor the abilities of smallholders to manage for them are well understood.
The enduring assumptions of ineffective smallholder land management have been only marginally supported with empirical evidence and primarily based on qualitative observations, and simplistic analyses of land use practices. To date, the soil knowledge, and agricultural and land management practices of smallholders have been largely overlooked, or altogether dismissed. Additionally, the range of factors that have influenced soil resource management and land use change in the region since the initial colonial interventions, and the various adaptive responses of smallholders to those influences, are also poorly understood. 9
Research Objectives
In the highlands of southwestern Uganda, the Districts of Kabale and Kisoro are characterized by high populations densities, widespread forest clearance, intensive agricultural production systems (including the cultivation of steep hillslopes and wetland areas), and widely assumed land use pressures and soil and land management problems. To better understand the environmental, cultural, agricultural and land use diversity present in the highlands, three sites were chosen for in-depth investigation.
The sites selected were Kicumbi and Nyarurambi Parishes in Kabale District (Figure
1.3) and Muramba Parish in Kisoro District (Figure 1.4). (A detailed description of
the three sites and the rationale for their selection is presented in Chapter Four). At each of the three sites in the highlands, an investigation was conducted into smallholder agro-ecological knowledge and agricultural and resource management practices, and the general land use trends over the last five decades.
The primary research questions addressed were as follows:
1. How do smallholders classify local soils and conceptualize
soil fertility and soil erosion processes?
2. How do smallholders manage local soils to control erosion
and maintain fertility?
3. How and why have smallholders' soil resource management
practices and land use changed since the first colonial
soil and water conservation interventions in the highland
region? 10 11
Figure 1.4: Kisoro District
Source : GOU, 1965c. 12
In the highlands of Kabale and Kisoro Districts, one of the primary objectives
of the study was to document the local soil classification schemes, and compare them
to conventional, scientific (e.g., USDA, FAO) soil classification systems. The agricultural, soil resource and land management practices of smallholders were also examined. Additionally, these practices have also been subject to change as smallholders have adapted to population pressures, colonial soil conservation
programs and agricultural policies, new agricultural market opportunities, civil strife and an array of other indigenous and exogenous factors. Accordingly, another objective of this dissertation was to examine how smallholders have adapted to increased pressures on the soil resource base, to identify the various factors that have induced soil resource management and land use change, and to identify smallholder- cited constraints to improved soil resource and land management in the highland
region. The specific objectives of the dissertation study were to:
1 delineate . and characterize smallholder soil classification systems.
2. compare the local soil classifications with the scientific
classification systems.
3. identify and describe smallholder soil erosion control, soil
fertility management and general land use practices.
4. identify smallholders' perceptions of soil and land
degradation, and constraints to improved agricultural
production and resource management. 13
5. identify how, and in response to what influences, soil
resource management and land use have changed since
the colonial Soil and Water Conservation By-laws were
introduced in the region (ca. 1945).
This study provides evidence of a wealth of smallholder soil knowledge, soil
resource management practices and land use strategies in the highlands of southwest
Uganda. It also describes the location-specific dimensions of agricultural production
and land use, and highlights the various agricultural and resource management
adaptations of smallholders as they respond to changing conditions within the highland
agroecosystem. The theoretical implications of the dynamic and diverse interactions
between dense populations, agricultural production, soil resource management, and
land use change in highland environments are also explored.
There have been, and continue to be, numerous social, political and economic
influences on land use and resource management in southwest Uganda, but it is not an objective of this dissertation to examine all of them in detail. Some of the major exogenous influences, including the impacts of the colonial state on resource management and land use will be explored. Additionally, while the socially differential (e.g., gender-based) dimensions of agricultural production, resource management and environmental degradation in the highlands are recognized and discussed, they are not fully addressed in this work. 14
Presentation of Research
Following the introductory chapter in which the research problem was
introduced, Chapter Two presents a review of the literature relevant to this study.
Theories of population growth, agricultural intensification, and cultural and political
ecology are discussed to elucidate the dynamic, adaptive and multiple-scale
dimensions of human-environment interactions. Additionally, the literature on soil
science, soil conservation and smallholder ("indigenous") knowledge are also
examined, with special emphasis given to tropical regions.
Chapter Three provides background information on Uganda and the
southwestern highlands. In the first half of the chapter, the current data for
population growth, agricultural production and land supporting capacity in Uganda are
presented. The latter half of the chapter characterizes the physical environment,
agricultural systems and population trends in the southwestern highlands and provides
the context to examine the various issues currently affecting resource management in
the region. In Chapter Four, the three study sites are described in detail and the
materials and methods employed in this study are presented. The methods utilized included informal individual and group interviews, a questionnaire survey and a conventional soil classification survey.
In Chapter Five, the influence of the colonial state on land management and land use change in the southwestern highlands is examined through a review of archival materials and related literature, as well as interviews with smallholders and elders citizens. This chapter underscores some of the dynamic elements of the 15 highland agroecosystem and highlights the diversity of factors, including the historical, which continue to influence agricultural and land management practices among smallholders in the region.
The field research results are presented in Chapters Six, Seven and Eight. In
Chapter Six, the smallholder soil classification schemes are characterized for the three study sites. The smallholder schemes are then compared and contrasted with the conventional, scientific soil classification systems. In Chapter Seven the conventional perceptions of soil and land degradation in the highlands are presented, and then compared to the analogous and unconventional perceptions of smallholders. The
various soil erosion control and soil fertility management practices of smallholders, and their use of organic and inorganic inputs are also examined. In Chapter Eight, the various forms of environmental, agricultural and land use change in the highlands, the diverse factors that have influenced the changes, and the different manifestations of change among the three study sites are examined. The smallholder-cited constraints to improved agricultural and land management in the highlands are also discussed.
The final chapter, Chapter Nine, presents a summary of the research findings, conclusions of the dissertation project. The recommendations for future research in the highlands of southwest Uganda and the implications of the research findings for similar environments in East Africa are also presented. CHAPTER 2 LITERATURE REVIEW
Population Growth
The relationship between population growth and resource use and mis-use have
been investigated extensively for almost two centuries, and the subject continues to
attract considerable attention in the popular media and academic circles (e.g., Arizpe
et al., 1994; Lee et al., 1988). The outcomes hypothesized about the interactions
between populations and the environment have long provoked contentious debates that
are characterized by two extreme views. One position argues that expanding
populations often degrade the environment, and underscores the finite nature of the
resource base and the limits to economic growth, while the other position contends
that population growth often induces innovations and creates new opportunities for
production and economic development.
The debate over the balance between population growth and the availability and use of natural resources has expanded ever since Malthus (1960) first posited in
1798 that expanding populations would eventually exceed the ability of the natural resource base to sustain them. Malthus argued that population growth would always increase at a faster rate than food supply, and he maintained that "preventative," cultural constraints (e.g., delayed marriages) and "positive" controls (e.g., war, disease, famine, poverty) would check a population's tendency to outgrow its resource
16 17
and economic base. Herein lies the origin of the idea that the earth can only support
a limited population, i.e., that the earth has a finite "carrying" or "land supporting"
capacity and definite "limits to growth" (e.g., Higgins et al., 1982, 1984).
The view that population growth is the primary cause of resource exploitation
and environmental degradation has continued to attract advocates over the years (e.g.,
Barney, 1988; Ehrlich, 1970; Ehrlich et al., 1993; Meadows et al, 1974). The
contemporary champions of this viewpoint, often referred to as Neo-Malthusians,
contend that there is a direct link between population growth and environmental
degradation. For example, population pressures have often been causally linked to
land degradation in developing countries in the tropics and Sub-Saharan Africa (e.g.,
Allan, 1965; Moss and Rathbone, 1975; Stahl, 1993). The Neo-Malthusians argue
that the earth has a finite natural resource base that cannot support unlimited population growth and economic development, and they also consider overpopulation to be a primary cause of poverty (Logan, 1991; Mortimore, 1993). Accordingly, they often advocate that the affluent minority in (and within) the developed countries should reduce their consumption of resources so as to benefit the poor majority as well as future generations.
Fundamental tenets of the orthodox, Neo-Malthusian perspective include the beliefs that environmental degradation and poverty is a result of overpopulation, at current rates of population growth world-wide food shortages are imminent, and to mitigate these threats population control programs need to be implemented throughout the world and particularly in developing regions (e.g., Brown and Kane, 1994; 18
Brown, 1995; Ehrlich et al., 1993). On the whole, the Neo-Malthusians do not
ascribe significant importance to the social, political and economic influences on
population-environment interactions, and tend to be reductionist in their focus on
over-population as the primary cause of environmental degradation and agricultural
crises across the globe.
Other scientists and researchers, however, take an opposing view of population
growth and consider the neo-Malthusians to be modern-day "Cassandras." They
argue, contrary to the Neo-Malthusians, that people are actually the "ultimate
resource" and human imagination and inventiveness can surmount most problems.
The "anti-Malthusians" decry the attempt to explain environmental degradation and land use in complex socio-ecological systems by means of simplistic and mechanistic population growth alone.
Simon (1981), for one, has asserted that human ingenuity and motivation will invariably produce new technologies or strategies to extend the limits of the earth's population "carrying capacity" well into the future. For example, he points out that the world's rate of population increase is almost four times greater than it was some
200 years ago during Malthus' time, and that the now earth supports roughly five times as many people (i.e., approximately 5.8 billion). In the worst case scenario, advocates of this optimistic position assume that depleted resources can be replaced with alternatives that will allow populations to grow and enable economies to continue to extend their "limits to growth" (Simon, 1981; Simon and Kahn, 1984; 1989).
While the Neo-Malthusians generally concede that industrial and technological 19
advancements have undoubtedly extended the limits to growth, they maintain that the
"carrying capacity" of the earth's resource base has been extended only temporarily and production capacity cannot increase indefinitely (Simmons, 1988).
Both the Neo-Malthusian and the anti-Malthusian positions can muster evidence and cite case studies from around the world to support their arguments. The two contradictory perspectives have a heuristic value, however, and help to illustrate both the destructive and productive aspects of human impacts on the environment, as well as to improve our general understanding of human-environment interactions.
Their views, however, are oversimplifications of a complex and dynamic reality and they represent but opposing ends of a spectrum of potential outcomes of human- environment interactions.
Agricultural Intensification
Malthus' general theory concerning population growth and resource management has been widely drawn upon to explain the processes affecting land use and environmental degradation in developing countries (i.e., the "Third World").
Within this theoretical framework, increasing population pressures are considered to be the primary cause of environmental degradation and agricultural production problems (e.g., Camp, 1992; Jolly, 1994; Okafor, 1987; Stahl, 1993). Intensive land-use systems are still commonly linked to soil erosion, soil fertility loss and soil mining via continuous cultivation (e.g., Getahun, 1991; Lubanga, 1988; Smaling,
1991). A variety of researchers have questioned what they perceive to be a
reductionist, deterministic view of human-environment interactions and contest the
notion that environmental problems can simply be explained away as a result of
population growth (Lockwood, 1995; Logan, 1991, Mortimore, 1993). A number of
alternative explanations have been proposed, including the hypothesis that population
pressures might prompt development or increased food production through agrarian
change (e.g., Boserup, 1965; Jolly, 1994; Netting, 1993; Pingali and Binswanger,
1988; Tiffen and Mortimore, 1994; Turner et al., 1993).
One of the most prominent theories relating to agrarian change was developed
by Boserup (1965), who proposed that pressures on resources resulting from
population expansion could stimulate technological change and agricultural innovation.
While population growth often results in increased pressures on the resource base, it
can also provide additional labor for purposes of agricultural intensification.
According to Boserup, a growing population can increase food supplies through the
intensification of agricultural production, where output per capita is maintained, but
output per area is increased. This increase in food production is often initially
achieved through the reduction of fallow periods, and later with the adoption of new
technologies and increased inputs of labor, and organic and/or agro-chemical inputs
(Boserup, 1965). Advocates of this general hypothesis have often been associated
with the "pro-population" camp characterized above by Simon (1981), although
Boserup does not support such interpretations. She asserts that while food production 21
has increased in some indigenous production systems, it has also often been
accompanied by declining labor productivity (Boserup, 1965, 1981).
Whether increasing population pressures can be considered an impetus for or a
hindrance to agricultural development remains a complex and controversial issue (Lele
and Stone, 1989). A multitude of studies have explored Boserup's proposition over
the last three decades and produced findings that contradict, support and expand the
original hypothesis (e.g., Allan, 1965; Grigg, 1979; Martin, 1987; Turner et al.,
1993). For example, some researchers argue that while population growth might lead
to agricultural intensification, it can concomitantly result in further soil degradation
and deforestation (Shapiro, 1995; Bilsborrow and Ogendo, 1992). Others criticize the
intensification hypothesis for being insensitive to local socio-economic stratification
and gender differences that can also affect agricultural production and resource
management (e.g., Fairhead, 1990; Hakansson, 1989).
On the other hand, evidence from a number of regions around the world indicates that population growth can also induce agricultural intensification and improved resource management (Turner et al., 1977). In Sub-Saharan Africa, examples of population-induced agricultural intensification have been described for
Burkina Faso (Vierich and Stoop, 1990), Cameroon (Campbell and Riddell, 1984),
Tanzania (Attems, 1968; Feierman, 1993; Ludwig, 1968) and Kenya (Goldman,
1993b). Smallholders in a variety of areas in East Africa have also experienced increased population pressures and varying degrees of agricultural intensification and land use change in recent decades (e.g. Flury, 1988; Ford, 1990; Jones and Egli, 1984; Ludwig, 1968; Maro, 1988). Research from Machakos District, Kenya,
reveals improved agricultural production and resource management under increasing
population pressures. Population density in this semi-arid landscape has increased
over the last forty-odd years and yet land deterioration has been arrested and
investments in agriculture and resource management have increased. Additionally,
successful soil and water conservation programs have been implemented, tree planting
has expanded and agricultural productivity has increased throughout the region
(Mortimore and Tiffen, 1994; Tiffen et al., 1994; Tiffen and Mortimore, 1994). An
increase in tree planting following rapid population growth has also been reported
in other areas of Kenya (Holmgren et al., 1994).
Furthermore, a collection of case studies from Sub-Saharan Africa examined
the relationship between population growth and agricultural change and suggested that increasing populations can induce "positive" changes in an agricultural system (Turner et al., 1993). The findings from the aforementioned studies were by no means uniform across sites and the variety of responses to population pressures included the diversification of labor, expanded market-oriented crop production, increased capital and labor investment in agricultural production, and the increased adoption of new technologies and utilization of purchased inputs. Overall, the studies indicates that while population growth does not automatically and mechanistically induce agricultural intensification, improved resource management and increased agriculture production can~and often does-occur in densely populated and intensive agricultural systems (e.g., Hyden et al., 1993). The primacy of population growth as impetus to agricultural intensification
has also been viewed with skepticism, however, and the relationship between high
population density and successful agricultural intensification is not always clear. A
number of studies have demonstrated that the intensification of agricultural production
might occur in response to population pressures, other factors such as the
environmental conditions are also important. For example, increased intensity of land
use may occur as a result of ecological factors such as low soil fertility, or the need
for erosion control and thus terracing on steep slopes (Brookfield, 1972; Campbell
and Riddell, 1984). Correspondingly, in Bangladesh, while population pressures were
found to be an important influence in the intensification of agricultural production in
three different agroecosystems, the agricultural responses were different at each site
and modified in large part by site-specific environmental conditions and constraints
(Ali, 1995).
Research from the Mandara mountains of northern Cameroon has provided
evidence for yet another population-induced agricultural transformation (Campbell and
Riddell, 1984). In this highland region, at one time population pressures had led to
the development of intensive agricultural systems that included the soil erosion control
measures such as terraces. Overtime, as people emigrated from the region in search
of employment or to acquire farmland previously unavailable in the low-lying areas,
the dependence of the intensive systems on high labor inputs was revealed (Riddell
and Campbell, 1986). As the population density declined in the highlands, so did the abundance of labor required to maintain the terraces, and out-migration from the 24
region eventually led to deterioration of the terraces and the intensive agricultural
production systems. A similar scenario has also been reported for southwest Yemen
(Vogel, 1988).
In this mountainous region, an intensive, terrace-based agricultural system
developed over several thousand years has recently begun to deteriorate. The bench
terraces traditionally utilized to reduce soil erosion and optimize water usage are no
longer widely maintained, and the cost to restore them are probably now prohibitive.
The system also collapsed because of heavy emigration and the decline of the rural
labor force. The process described in these two studies, whereby a previously
intensive agricultural system deteriorates and is reduced to a lower level of productivity and resource management, has been referred to as "disintensification"
(Brookfield, 1972; 1984).
A recent study on Rusinga Island in Kenya has also described the discontinuation of intensive agricultural practices, including terracing (Conelly, 1994).
In this area, however, the population density did not decrease. Instead, the change in land use intensity was traced to the loss of labor to commercial fishing and migrant labor opportunities; both of these employment options provided men and women with increased remuneration for their labor investments. Other studies in developing countries have demonstrated that population growth can lead to off-farm employment, as well as stimulate out-migration (e.g., Bilsborrow and Geores, 1994). These examples of disintensification reveal the impact of external influences on a system, but also underscore the dynamic and temporal dimensions of intensive agricultural 25
production systems. Agricultural production is invariably subject to various internal
and external influences that can cause small changes in the nature of production, i.e.,
or even the induce the system to find another "equilibrium" state.
Alternatively, studies from East and West Africa have documented the
occurrence of agricultural intensification in areas of low population density. This
process primarily resulted due to opportunities provided through new land and market opportunities, as well as increased infrastructural support (Goldman, 1993; Netting et al., 1989). Goldman (1993b) has also distinguished between population-driven and
market-driven forms of agricultural intensification, and provided examples of the latter impetus to increased production in Kenya. In these cases, agricultural
intensification occurred not in response to demographic pressures alone or at all, but to various social and economic factors that were external to the production systems.
The aforementioned studies reveal that societies have responded to changes in population—either population growth or decline-in a variety of ways depending on the nature of agricultural production, social organization, local and regional economic structures, and geographic location (i.e., environment). In addition to population pressures, specific factors that have been found to influence the interactions between humans and their environment include land tenure, available technology, market access, social institutions, multi-level politics and environmental conditions and constraints. These studies underscore the dynamic nature of the agricultural systems, which are constantly changing in response to a variety of constraints and opportunities
(e.g., Berry, 1993; Mortimore, 1989). They have demonstrated not only the dynamic 26
nature of agricultural production systems, but also the resiliency of many
smallholders, as well as their varying abilities to adapt to changing social and
environmental conditions.
While population growth is often an obvious influence, it is infeasible to assign
primacy to any one of the multitude of ecological, social, political and economic
factors that stimulate agricultural change (e.g., Cleaver and Schrieber, 1992). There
are variety of factors influencing agricultural change for a given location, and it is
difficult to discern how various factors will interact in a given environment or prognosticate how the multiple and varied interactions will be manifested within an agricultural system. The nature of change in agricultural production and resource
management systems invariably will be location- and period-specific.
Cultural Ecology
Cultural ecology is a research paradigm which shares interdisciplinary roots in
1 both anthropology and geography. It has developed as a result of efforts in both disciplines to create a more realistic conceptual framework within which to investigate human-environment interactions and cultural adaptations. The study of the interaction of societies with one another and with the natural environment in order to understand and explain the processes of adaptation and transformation that operate to alter social institutions, human behavior and environment has characterized the cultural ecology
1 The term "human ecology" is also frequently used in contemporary nature-society studies, which focus on human adjustment to the natural environment while emphasizing the adaptive character of the human-nature interaction and its mediation by social institutions (Johnston et al., 1994: 258). 27
research tradition. Cultural ecology "focuses on the patterns of resource use and
production because they are the primary links between people and their environments;
the ability to cope with environmental change is a major determinant of the viability
of rural communities" (Grossman, 1984: 84). More recently, Butzer (1989: 686)
provided an outline of the predominant characteristics of the cultural-ecological
approach in geography:
1 Society . and nature are intimately interconnected and bound
by complex systemic interrelationships;
2. cultural behavior and diversity are explicitly considered in
their functional role and with respect to material and
non-material culture; and
3. food production, especially in relation to demographic
variables and sustainability, is a fundamental theme.
The role of people and the manipulation of resources within ecosystems, rather
than the delineation or simulation of such systems as a whole, is seen to be the main
concern of the cultural ecologist (Butzer, 1990: 686). For Turner (1989: 92), cultural
ecology could be described as a "research perspective on nature-society relationships
that are addressed largely, although not necessarily, at micro- and meso-spatial scales
in non-western settings. It has sought understanding primarily through systemic analyses and empirical examinations of problems and themes that link human activity and the physical environment ... the best of these studies have involved intensive, detailed field research, typically interdisciplinary and often team-based." 28
The cultural ecology paradigm has evolved through the close interaction
between geography and anthropology and their common interests concerning society-
nature relationships (Butzer, 1989; Ellen, 1988; Grossman, 1984). It has attempted to
create a conceptual framework within which to investigate human-environment
interactions and cultural adaptations, and it has a strong tradition in both disciplines
(e.g., Bennett, 1969; Geertz, 1963; Moran, 1990, 1991; Netting, 1968; Sauer, 1925;
Steward, 1955). Cultural-ecological research has characteristically involved both diachronic ("historical") and synchronic ("contemporary") studies, although the latter are more common in both disciplines (Butzer, 1989; Porter, 1991).
Investigations of human-environment interactions have a well-grounded tradition in geography (e.g, Butzer, 1989, 1990; Brookfield, 1984; Porter, 1978,
1984; Sauer, 1925). A cultural-ecological research framework has been widely employed to study human adaptations to the environment, generally in terms of the range of changes in agricultural and resource management systems, and the diversity of indigenous knowledge and resource management skills (e.g., Denevan, 1983;
Hardesty, 1986; Knight, 1974; Porter, 1965, 1970, 1978; Richards, 1985, 1986;
Turner et al., 1977). The studies generally have been conducted among "primitive" peoples and in the developing world. Research has also focused on agricultural and resource management change in highland and mountain regions, and stressed the unique nature of highland ecosystems and the human adaptations to them (e.g., Allan et al., 1988; Fedele, 1984, Jodha et al., 1992). Specifically, studies have been conducted in variety of mountain environments, including Papua New Guinea 29
(Brookfield, 1984), the Andes (Guillet, 1983), the Himalayas (Fricke, 1989; Zurick,
1989) and central and east Africa (Campbell and Riddell, 1984; Ford, 1990).
The cultural ecology approach, however, has never attained widespread
acceptance as a research framework within anthropology or geography, and a variety
of criticisms have been levied against it. The sub-field is characterized
by been an overall reluctance to undertake "prescriptive studies" advocating specific
management schemes, or policy-oriented studies focusing on "strategic action"
(Bennett, 1976). Netting (1986) attributed the limited impact of cultural ecological
research to the following factors:
1. an overconcern with societies in the past, or remote societies
whose environmental impact has been minimal;
2. a tendency to treat societies as cultural isolates with limited
attention paid to external influences;
3. an overemphasis on descriptive studies, and on culture rather
than ecology; and
4. the relatively small number of diachronic studies, resulting in
a tendency to conceive of ecological relations as
relatively stable and enduring.
While cultural ecology remains a viable research approach in both geography and anthropology, it has been rapidly supplanted by the burgeoning field of political ecology and the related regional political ecology. Political Ecolog y
The conventional perception of agricultural crises and environmental
degradation in developing countries generally involves the linkage of fragile
environments, overpopulation and the mismanagement of resources by local people
(e.g., Camp, 1992). This outlook is characterized by an inherently deterministic,
technocratic view of the relationship between population and the environment that
generally assumes a negative impact by the many and the poor on the natural resource
base (Broad, 1994; Logan, 1991). Environmental problems are usually depicted as
being physical in nature, and people (e.g., smallholders) are portrayed as victims of
their own inappropriate land management practices, indolence, ignorance and short-
term perspectives. To follow, it often has been assumed that environmental
degradation could be alleviated simply by providing technical assistance and
scientifically derived technological solutions (Blaikie, 1985; White and Jickling,
1995).
Conversely, efforts to expand the scope of cultural-ecological research beyond
community-level analysis have emphasized the need to consider a wider array of
factors that influence agricultural production and resource management systems. A
growing body of research has examined agricultural change, resource use and environmental degradation in terms of the larger socio-economic, political and historical context within which people pursue their livelihoods (e.g., Blaikie and
Brookfield, 1987a). The development of the political ecology approach, whereby
Marxist political economy theory is fused with ecology and the natural sciences, has 31
been one of the primary means of introducing these external factors into a cultural-
ecological research approach. For Bassett (1988: 469), the "detailed attention to the
interrelationships between the political economic and human ecological dimensions of
production processes distinguishes the political ecology approach from human
ecology."
In the political ecology framework, a social, economic and political analysis of
resource "mis-management and land abuse" has evolved and been widely applied in
developing countries (e.g., Redclift, 1987). For example, this approach has stressed
that soil and land degradation are not simply the result of demographic pressures,
ecological constraints or marginal environments. Rather, it demonstrates that
environmental problems are symptomatic of social, economic and political crises, and
underscores instead the coincidence between degraded environments and politically
marginalized and powerless people with few realistic alternatives (Baker, 1984; Thapa and Weber, 1991; Peluso, 1992a, 1992b).
In a broad social analysis of environmental destruction and soil erosion,
Blaikie (1985) addressed the wide range of political and economic interactions between the land-users and the state. Blaikie (1985: 117) argued that the adverse socio-economic and environmental conditions under which "marginal" peasants and pastoralists operate often force to them to commit "ecocide", despite their knowledge and land management skills. Correspondingly, a number of studies have also pointed out that rural people, researchers and government officials often have different perceptions of "environmental problems". The divergence of opinion among 32
smallholders and outsiders over the occurrence and causes of soil erosion, and the
resultant clashes over approaches to land management, have been reported in a
number of countries, including Thailand (Pahlman, 1991), Bolivia (Zimmerer, 1993)
and Sierra Leone (Millington et al., 1989).
In Latin America, the political ecology research perspective has been widely
employed over the last decade (e.g., Collins, 1986; Zimmerer, 1991). Schmink and
Wood (1987) utilized a political ecology perspective to study resource management
and land use patterns in Amazonia, and demonstrated how economic and political
processes have determined the way natural resources have been exploited. Hecht
(1985) has also examined the influence of national development policies on resource
use in Brazil. She blamed the regional development and cattle ranching schemes and
their specific patterns of capital accumulation on widespread ecological damage and
deforestation in Amazonia. More recently, Stonich (1993) has utilized a political
ecology framework to implicate the state and policies for much of the poverty and
environmental destruction in Honduras.
In Sub-Saharan Africa, a wide range of studies have utilized a political ecology
research framework. For example, the linkages between farming and environmental
degradation, and the interactive complex of socio-economic, political and physical
forces in Zimbabwe (Stocking, 1987), while the political, economic and planning
systems have been implicated for many of the environmental problems in Kenya (Fox,
1988). The importance of understanding political and economic forces in shaping the crises have been underscored in a variety of research projects, including studies of drought and famine in Nigeria (Watts, 1983) and Ethiopia (Kebedde and Jacobs,
1988), peasant-herder conflicts in the Ivory Coast (Bassett, 1988), overgrazing in the
Sahel (Turner, 1993) the gender dimensions of resource management and agrarian
change in the Gambia (Carney, 1993; Schoederer, 1993) and environmental resource
conflicts in highland Zimbabwe (Moore, 1993). Additionally, a number of studies
have examined the interaction of social, political, economic and historical (colonial) factors and their impact on environmental degradation in Ghana, Madagascar and
Sub-Saharan Africa (Amanor, 1991, 1994a, 1994b; Jarosz, 1993; Krokfors, 1989), development processes in the Senegal River Basin (Park, 1993) and livestock diseases in East Africa (Musere, 1990; Turshen, 1984).
These diverse studies have utilized the political ecology approach to emphasize the various interactions between resources, individuals, societies, economies, and institutions. They have attempted to go beyond the commonplace analyses of how humans respond to population growth or adapt to isolated environments, and began to address how social institutions at a variety of geographic scales interact with communities and the physical environment (e.g., Blaikie, 1988; 1989a; Little and
Horowitz, 1987; Watts, 1983; Wisner, 1989). While many political ecology studies have focused on the delineation of the multiple factors influencing environmental degradation and resource management, less attention has been directed to intra- regional differential responses to internal and external influences, or to the successful micro-level adaptations or coping strategies of local people. 34
To follow, the dynamic relationship between society and the environment has
been increasingly examined from within a regional perspective, an approach termed
"regional political ecology." Here, the regional dimension of environmental
"problems" are emphasized in order to take into account the environmental variability
and the spatial variations in resilience and sensitivity of the land, as different demands
are put on a given area over time (Blaikie and Brookfield, 1987b: 17). In addressing
ecological issues, the regional approach highlights the "institutions and the political
structures that set a matrix of limits, constraints and possibilities for resource
management" (Butzer, 1989: 202). Blaikie and Brookfield (1987b) recognize not only
the constantly shifting dialectic between society and land-based resources, but they
also emphasize the dynamic interactions within classes and other groups in society,
and thus their differentiated impacts on the environment.
With the regional political ecology approach, Blaikie and Brookfield (1987a)
stress the importance of understanding environmental variability as well as the
location-specific dimensions of environmental degradation. An analysis of the impact
of colonialism on resource management has also become a heavily underscored
component of the regional political ecology approach (e.g., Blaikie and Brookfield,
1987c, 1987d). In this approach, chains of explanation are developed to demonstrate
how distant historical events are linked to specific forms of degradation. Research applications of the "regional political ecology" framework have been varied and also included investigations into the agricultural crisis in a mountainous region of interior
Northern Portugal (Black, 1990), wetlands (dambo) cultivation in Zimbabwe (Bell and 35
Roberts, 1991), montane-bog agriculture and wetland production in the Peruvian
Andes (Zimmerer, 1991), and mescal production in Sonora, Mexico (Burwell, 1995), and extractive reserves in East Kalimantan, Indonesia (Peluso, 1992a, 1992b).
The political ecology and related regional political ecology frameworks have not evolved without their share of criticism. For some observers, the emphasis on the socio-economic and political variables have often tended to overwhelm the analyses
(Bryant, 1992). The environmental variables and variation are often neglected in favor of socio-economic and political factors, and many studies continue to be largely descriptive—albeit of a "larger social system." While the regional political ecology
approach has attempted to address some of these concerns, it too has been criticized.
Black (1991) has cautioned that the plurality characteristic of the regional political ecology perspective might become so broad as to open up a "Pandora's box of
processes and outcomes. " While the political ecology approach has been characterized by efforts to expand the methods of analysis and a more dynamic research framework, there have been few applications of the approach for prescriptive or policy-oriented purposes.
Overall, both the general political ecology and the regional political ecology approaches have developed broad, flexible and dynamic frameworks within which to examine human-environment interactions at a variety of geographic scales. They also effectively elaborate the interrelated ecological, social, economic, political and historical dimensions of resource use and environmental problems, as well as the social consequences of resource depletion and environmental deterioration. 36
The Colonial State and Land Management
Studies of land degradation have increasingly recognized the historical (e.g.,
colonial), as well as the social, economic and political factors which have influenced land
management on the continent (e.g., Critchley et al., 1994; Jiggins, 1989). Widespread
interest in soil erosion and land degradation first appeared around the world in the
1930's, a concern provoked by the occurrence of the Dust Bowl in mid-west of the
United States. The majority of the colonial soil conservation efforts were developed in effort to avoid the disastrous loss of top soil and crop failures that characterized the Dust
Bowl, and anti-erosion programs were implemented in a number of colonies in Sub-
Saharan Africa (e.g., Anderson, 1984; Gustafson, 1937; Jacks and Whyte, 1938;
Stockdale, 1937; Throup, 1988).
In depth examinations of the impact of the colonial state on agricultural production, land degradation and deforestation in a number of countries on the continent
(e.g., Richards and Tucker, 1988; Tucker and Richards, 1988). The role of the colonial state in formulating and implementing soil conservation policies has been examined in
Kenya (Pretty et al., 1995; Throup, 1988), Sierra Leone (Millington, 1987c), Zambia
(Wood, 1992), Zimbabwe (Whitlow, 1988) and Southern Africa (e.g., Beinhart, 1984;
Showers, 1989). Across the Sub-Saharan Africa, the colonial soil conservation efforts were characterized by a strong emphasis on physical constructions (e.g., terraces, grass bunds), a reliance on coercive or punitive measures to achieve implementation of soil conservation programs, and an overall neglect of, or insensitivity to, the traditional farm economies and environmental management practices. Many of the early colonial efforts, 37
such as those in Kigezi District in southwest Uganda, achieved what were lauded as
2 remarkable soil conservation successes (Masefield, 1962). Paradoxically, as is the case
in former Kigezi District, many of these externally-imposed, top down land management
programs either collapsed or were poorly implemented with the dissolution of the
colonial state.
The development of many of the colonial soil and water conservation/land
management programs were noted for their impressive constructions, but were also
remarkable in that they rarely included the participation of local smallholders (e.g.,
Critchley et al., 1994; Stocking, 1985; Wood, 1992). The balance between the local,
traditional farming systems and the environment was often undermined by colonial land
management interventions and export-oriented agricultural policies (e.g., Anderson,
1984; Blaikie and Brookfield, 1987a). For example, in Lesotho resource management
regimes implemented under colonial administrations were blamed for undermining
traditional resource management systems, and often as a result inducing or exacerbating
land degradation and affecting attitudes towards soil conservation (Showers and
Malahleha, 1992; Showers, 1989).
Additionally, in Zambia Wood (1992) reported that a number of colonial policies led to changes in traditional farming practices and eventually induced significant soil erosion. These policies included establishment of the Native Reserves (1927-28), which restricted the indigenous population to reserves, the promotion of semi-commercial
2 A closer examination of the success, failures and impacts of the colonial soil and water conservation efforts in the highlands of southwest Uganda is undertaken in Chapters Five and Seven. 38
agriculture (e.g., maize production) and the introduction of "improved" technologies like
the plow-which concomitantly led to expansion of cultivation and further exacerbated
land pressures. The plow also broke up the soil structure and generally reduced its
resistance to erosion, and facilitated the cultivation of heavier clay soils that were more
susceptible to surface erosion and gulleying (Wood, 1992: 157-159).
Studies have also begun to examine the validity and accuracy of historical
accounts and conventional perspectives of environmental degradation, and to offer new
interpretations of human-environment relationships in areas of West and East Africa
(Fairhead and Leach, 1994, 1995; Rocheleauetal., 1995). This expanded social analysis
of landscape histories has not only re-examined colonial reports and research data, but
most importantly included and underscored the often differing perspectives of local
people in the "new" interpretations. As discussed in the sections above, many
researchers now emphasize the impact and the interactions of various historical, social,
economic and political factors on environmental mismanagement and resource use and
abuse (e.g., Baker, 1984; Blaikie, 1985; Blaikie and Brookfield, 1987a).
Recent Approaches to Land Management
Smallholders often have been portrayed as poor farmers and resource
managers, and many environmental problems in developing countries have been depicted as being physical in nature. Accordingly, it has often been widely assumed that soil and land degradation could be alleviated with technical assistance and new technological solutions, including improved extension services, increased agro-
chemical and labor inputs, and improved physical constructions such as terraces (e.g., 39
Baker, 1984; Timberlake, 1986). While a variety of technical measures and land
management practices to arrest erosion and improve soil conservation have been
available for decades, the adoption of soil conservation measures and land
improvement technologies in many developing countries has been quite limited
(Blaikie, 1985; Critchley etal., 1994; Hudson, 1992).
The culpability of smallholders in environmental degradation has often been
linked to their "short-term perspectives," whereby they are assumed to exploit the
environment in an "unsustainable" manner for immediate gain. For example,
smallholders have often been blamed for "mining" soil nutrients without considering
3 the "sustainability" of their actions (e.g., Dregne, 1990; Smaling, 1991). While soil
mining is often widely criticized, in some circumstances smallholders may consider
the practice "good economics" and an acceptable cost of agriculture production
(Ruthenberg, 1980). Such practices are also undertaken in attempt to meet immediate
household food needs or ever-changing and often short-lived market opportunities.
In recent years many investigations into the multiple dimensions of agricultural and environmental "crises" in Sub-Saharan Africa have been undertaken within a political ecology framework that has been influenced by marxist, social analysis.
Other studies, however, have analyzed the socio-economic and political dimensions of agrarian crises, environmental degradation, and ineffective resource management programs employing neo-classical economic approaches. Bates (1981; 1989), for
3 Soil mining is defined here as the extraction of soil nutrients through continuous cultivation without the addition of organic or chemical amendments. 40
example, has combined political and economic analyses to examine the linkages
between changing political factors and institutions, and their impact on the agrarian
economy in Kenya. From similar theoretical perspectives, several studies have
examined the economic aspects of land degradation (Bojo, 1991), the cost and benefits
of the adoption of soil conservation measures (Lutz et al., 1994) and related
agroforestry practices (Current et al., 1995).
To follow, a number of studies have examined the successes and failures of
both colonial and contemporary top-down soil and water conservation programs across
Sub-Saharan Africa. They have identified a wide variety of factors that affect land
management and the adoption of soil conservation technologies at the farm or
household-level. The factors include farm size, land fragmentation, security of land
tenure, access to credit and agricultural inputs, labor availability, gender roles,
cultural biases and lack of awareness of a problem. The studies have also identified
factors that were essentially external to the smallholder production system included
poor infrastructural development and market access, inadequate institutional and
research support (e.g., Anderson, 1990; Blustain, 1982; 1985; Hudson, 1992; Kelley,
1983; Napier et al., 1994; White and Jickling, 1995). Furthermore, soil and water
conservation programs were characterized by "top-down" development approaches, an
ignorance of local ("indigenous") knowledge and management practices, and minimal participation of local people in all stages of the project development (e.g., Critchley et al., 1994; Hallsworth, 1987; IFAD, 1992; Reij, 1991). 41
More recent soil conservation programs have moved away from explicit
technical and engineering strategies and instead promote a mixture of technologies, an
expansion of biological methods (e.g., agroforestry, green manures) and increased
smallholder participation in land management initiatives (e.g., Critchley, 1991; Meelu
et al., 1994; Moldenhauer and Hudson, 1988; Young, 1989). For example, the "land
husbandry" approach advocates a multidisciplinary research framework to increase
production through improved resource management initiatives (Shaxson et al., 1989).
Such efforts increasingly promote low-input, multipurpose technologies that are more
appropriate for low-resource farmers and their environmental and socio-economic
constraints (Lai, 1990). Agroforestry is one multi-purpose technology that has been
widely promoted in Sub-Saharan Africa (Sanchez, 1987; Young, 1989), although
there has been less adoption of many agroforestry technologies (e.g., alley cropping)
than anticipated (Carter, 1995).
The "catchment approach" to soil conservation, initially implemented in Kenya in the 1980's, is one example of a successful land management initiative. It is often cited as an example of a "participatory", community-based, multi-sectoral resource
management initiative. In this approach, communities within a watershed, i.e., catchment area, participate in an interdisciplinary program to develop and implement a regional soil and water conservation programs. The approach is characterized by a grassroots mobilizing effort and which generally achieves widespread involvement from all sectors of the participating communities. To date, the catchment approach 42
has been widely touted as successful, multi-disciplinary land management program
(Admassie, 1992; Pretty et al., 1995).
Land Degradation and Soil Science
Soil degradation, particularly soil erosion and soil fertility loss, has
consistently been reported as a serious constraint to improved agricultural production
throughout Sub-Saharan Africa (e.g., Brown, 1989, 1994; Dregne, 1982, 1986, 1990;
Lai, 1987, 1988a; Popenoe, 1986; Smaling, 1991; Smaling et al., 1993; Stoorvogel
and Smaling, 1990; Stoorvogel et al., 1993; UNEP, 1992). Soils across the continent
are often characterized as inherently infertile, and susceptible to erosion and
mismanagement: "African soils tend to be shallow, have poor texture, are inert and
have low water-holding capacities" (Opio-Odongo, 1992: 10). The low levels of agricultural production in Sub-Saharan Africa also have been linked to environmental degradation and numerous "ecological constraints" (e.g., Allan, 1965; Lai, 1987,
1998a; Matlon and Spencer, 1986; Phillips, 1959; Popenoe, 1986; Weischet and
Caviedes, 1994).
Research on soils in Sub-Saharan Africa, however, have been quite limited when compared to the work undertaken in temperate regions, and as a result there has been a tendency to over-simplify discussions of "African soils" (e.g., Dregne, 1990;
Lai, 1988a). Scientific knowledge of the chemical and physical properties of soils in
Sub-Saharan Africa is far from complete, and many of the common assumptions concerning the inherent infertility and high erodibility of tropical soils have been increasingly questioned. Many of the misconceptions about soils and land degradation 43
processes in the tropics have been based on extrapolations of temperate soil data,
limited and qualitative field observation, and simplistic interpretations of complex
population-environment relationships. The most common ecological models and
agricultural and land management technologies applied to tropical environments are
also derived from scientific research conducted in temperate regions, and are often not
directly transferable. Correspondingly, the adoption of many introduced land
management technologies in Sub-Saharan Africa have been limited in part been due to
their inappropriateness for the unique environmental conditions and agricultural
systems in the region (Allan, 1965; Austen and Hendrick, 1983).
In terms of the physical environments in Sub-Saharan Africa and the tropics in
general, there is "inadequate information on principal soils of the region, interaction between soils and the prevalent climate, soil physical and mineralogical properties, soil chemical and nutritional characteristics, soil biota and their effects on productivity" (Lai and Sanchez, 1992: ix). The information for tropical soils is incomplete when compared to the wealth of data that has been accumulated for temperate soils, but recent research efforts have begun to contribute new insights into the different and dynamic processes affecting soil erosion and soil fertility and their management in the tropics (e.g., Mulongoy and Merckx, 1993; Ross, 1993; Woomer and Swift, 1994).
Discussions of soils in Sub-Saharan Africa have often been replete with overgeneralizations and misinformation, and investigations of land degradation on the continent have been equally prone to mis-measurement, exaggeration and erroneous 44
extrapolation. Many investigations into land degradation in developing countries have
been characterized by limited field observations, inappropriate methods and scale of
analysis, and data of questionable accuracy (Blaikie, 1985, 1989a, 1989b; Carter,
1993; Mortimore, 1989). Investigations of environmental degradation often have
been undertaken using limited, unreliable and even irrelevant data. There also have
been problems associated with both defining the "soil-problem" and "measuring" the
frequency, magnitude and scale of soil erosion and land degradation (Blaikie, 1989a,
1989b; Blaikie and Brookfield, 1987b; Stocking, 1987). Additionally, numerous
studies of resource management and land degradation have been conducted at
4 inappropriate scales and using macro-level data. Studies utilizing macro-level data
have been conducted on population-environment relationships in Sub-Saharan Africa
(e.g., Lele and Stone, 1989), assessments of the global and African-continental extent
and causes of soil and land degradation (e.g., Dregne, 1986; Oldeman, 1994;
Stoorvogel et al. , 1993) and determinations of the land supporting capacity for a
variety of developing countries. 5
Macro-level studies of various topics such as land supporting capacity can provide a general index of land use intensity for a given region or country, or play a
The oversimplifications inherent in macro-scale studies of land degradation have been noted in some studies (e.g., Smaling et al, 1993).
5 Using macro-level data, the United Nation's Food and Agriculture Organization (FAO) has concluded that under traditional systems of agricultural production, populations have already exceeded the land-supporting capacities in many countries in Africa (e.g., Higgins et al., 1982; Ho, 1990). The FAO studies and their specific implications for Uganda are further reviewed in Chapter Three. 45
variety of heuristic roles in research and policy development (e.g., Arrow et al,
1995). However, while many of the macro-level studies may help to identify and/or
publicize changing environmental and land use trends, the data on which they are
based are also often of questionable accuracy or simply estimates (Raikes, 1988).
Additionally, while the macro-level studies represent one of the few sources of data
for many developing countries, they are too often the only source and thus drawn
upon to develop land use and resource management policy in lieu of more costly
micro-level studies. Projects like the FAO/UNESCO World Soil Map at 1:5 million
scale (FAO, 1993a) provide important contributions to our knowledge of the nature
and distribution of soils around the world, but they are by design small-scale maps
that generalize about regional soil and other environmental characteristics, and
overlook farm-level soil and management variations (Carter, 1993; Grosjean and
6 Messerli, 1988).
Furthermore, regardless of the scale of analysis, a number of investigations
have disputed the idea that soil erosion and land degradation can be attributed
uniquely to human action. Recent studies in the Himalayas have questioned whether
land degradation in mountainous areas can be blamed solely on smallholder ignorance and poor land management practices, and have posited that natural hazards and ecological factors might play a larger role in the processes than previously assumed.
6 The FAO soil maps are approximations, and attempts to extrapolate from them are fraught with potential mapping-scale, classification or data source errors (e.g. , Andriesse, 1988). The simplifications inherent in macro-level studies have been acknowledged (e.g., Smaling et al., 1993), but there remains a common tendency to generalize such data across a diversity of environments. 46
In contrast to the long-held assumptions that humans were the primary causal-agent,
the studies argue that significant natural erosion occurs in the mountainous regions as
a result of natural but often extreme climatic events (Ives and Pitt, 1988; Ives and
Messerli, 1989; Stone, 1992). Additionally, the causes of environmental degradation as well as the potential land management solutions are generally location-specific
(e.g., Blaikie and Brookfield, 1987a). Researchers have also pointed out the importance of differentiating between the natural processes for a given environment and the effects of human interference, and of further distinguishing between the harmful and positive elements of that interference (Blaikie and Brookfield, 1987b).
The prevailing focus of soil and land management studies on soil erosion and
soil productivity problems has also been expanded to include discussions of soil resiliency: that is, a soil's capacity to recover after a disturbance (e.g., Greenland and
Szabolcs, 1994). The social and agricultural sciences also recognize that all social, agricultural and ecological systems are characterized by varying levels of resiliency, as well as inherent capabilities for adaptation and change (e.g., Conway, 1987;
Dommen, 1988; Dovers and Handmere, 1992; Holling and Bocking, 1987;
Mortimore, 1989).
It is now widely accepted that an ecosystem, or an agro-ecosystem, has a certain "resiliency" or "adaptability"~that is, the capacity to absorb changes and to endure stresses (Denevan, 1983; Holling, 1973, 1986; Holling and Bocking, 1987;
Moran, 1990). While the resilience limits of a system are not easy to identify, once they are exceeded, the system rapidly seeks equilibrium at a different level of production (Johnson and Lewis, 1995). From a management perspective, an agroecosystem's capacity to absorb change also implies certain levels of flexibility, i.e., change is a systems' property that can be accommodated and even managed for, as opposed to being a property to be avoided or possibly eradicated (e.g., Conway,
1987; Dovers and Handmere, 1992). Smallholders are often changing the nature of their agricultural production systems in response to a variety of ever-changing conditions (Brookfield and Padoch, 1995; Dommen, 1988; Neimeijer, 1996).
On the whole, the western scientific knowledge of tropical soils and land management remains deficient. Efforts to expand the basic, scientific understanding of tropical soils and how best to manage them are ongoing, but far from complete.
An improved understanding of smallholder knowledge, experiences and management practices of can underscore their unique insights and abilities, and also complement ongoing conventional soil science investigations.
Smallholder Knowledge
Our programs of agricultural aid pay little attention to native ways and products. Instead of going out to learn what their experiences are, we go forth to introduce our ways and consider backward what is not according to our pattern. (Sauer, 1956: 68)
Investigations into indigenous knowledge and traditional agricultural systems of smallholders have been undertaken for decades, and many of the early studies underscored the rational and optimizing dimensions of smallholder management systems (e.g., Allan, 1965; Conklin, 1954; Sauer, 1956; Schultz, 1964). The studies were largely valued for their academic or theoretical contributions, however, and the 48
insights into traditional agricultural and resource management systems were not often
incorporated into rural development efforts.
Since the early 1980's a diversity of academic disciplines, including
anthropology and geography, have demonstrated a renewed interest in indigenous
(local) knowledge (e.g., Brokensha et al., 1980; Brush and Stabinsky, 1996;
7 Chambers, 1983; Richards, 1985; Warren et al., 1991). The more recent interest in
indigenous knowledge has included both academic, and increasingly, applied or
development-oriented endeavors. The recognition of the contributions that indigenous
knowledge might make to sustainable development efforts is gaining increasingly
widespread acceptance among research scientists, development workers and
environmental conservationists alike.
The profusion of publications that address the subject attests to the widespread interest in the practical value of indigenous knowledge, as well as the variety of disciplines involved in this growing initiative. Examples of recent works include general overviews of traditional ecological knowledge (Inglis, 1993; Johannes, 1989;
Johnson, 1993), studies of indigenous knowledge within common property resource management systems (Berkes, 1989; Jodha, 1992) and investigations into the potential role of indigenous knowledge in forest resource management (Clay, 1988; Oba, 1994) and sustainable development efforts (e.g., Bodley, 1988; Clarke, 1990; Freeman and
7 There are a wide range of terms in use that loosely correspond with "indigenous knowledge". Some examples include: ethnoscience, people's science, traditional ecological knowledge, folk ecology, indigenous technical knowledge, local knowledge, rural people's knowledge and farmer knowledge. Chambers (1983) provides an in-depth discussion of the various "meanings" and implications of many of these terms. 49
Carbyn, 1988; Gladwin, 1989a, 1989b; Thurston, 1992; Warren et al., 1991; World
Bank, 1994). 8
The complexity and diversity of indigenous knowledge for agricultural
production and resource management has been recorded for many regions of the
world (e.g., Beyer, 1980; Warren, 1991; Hallsworth, 1987). In Asia, data have been
collected on indigenous knowledge for agroforestry systems (Clarke and Thaman,
cropping 1993) , systems and shifting cultivation (Conklin, 1954, 1980) and soil
knowledge and land management (Colfer and Newton, 1989; Dvorak, 1989; Martin
and Vityakon, 1986; Muller-Boker, 1991; Perrot-Maitre and Weaver, 1992; Scott and
Walter, 1993; Weinstock, 1984).
In Latin America, extensive research has also been conducted on indigenous or
traditional agriculture (Altieri, 1991; Posey, 1983; Wilken, 1987), tropical forest
management (Clay, 1988; Posey, 1993), and soil classification and land management
(Behrens, 1989; Bellon and Taylor, 1993; Bellon, 1995; Bocco, 1991; Furbee, 1989;
Guillet, 1987; Pitalbo, 1981; Ryder, 1994; White and Jickling, 1995; Zimmerer,
1994) . More specifically, Hecht and Posey (1990) provide a detailed description of the ethnopedology, agronomic management practices and the effects of the practices on crop production and soils for the Kayapo in the Amazon basin. Hecht (1990) has defined "ethnopedology" as the study of native land classification systems,
8 Indigenous knowledge has been documented for a wide variety of topics, including medicine (Riley and Brokensha, 1988), veterinary medicine (McCorkle, 1989a, 1989b), agroforestry (Hoskins, 1984), natural pesticides and botany (Warren et al., 1991). management techniques and their variations, and how the practical and theoretical knowledge is developed, expanded and encoded.
In Sub-Saharan Africa, a variety of early studies described a diversity of agriculture production systems and traditional resource management practices (e.g.,
Allan, 1965; de Schlippe, 1956; de Wilde, 1967; Phillips, 1959; Matheson and
Bovill, 1950; Trapnell, 1953; Trapnell and Clothier, 1957; Thomas and Whittington,
1969). For example, studies from Zaire have described a number of admirable management practices: shifting cultivation, land fallow and rotation, use of ash and compost soil inputs, crop rotation, mixed-cropping-even local soil taxonomies (Jurion and Henry, 1969; Miracle, 1967). More recent research has also revealed an abundance of indigenous knowledge for agricultural production and resource management (e.g., Porter, 1984; Knight, 1980). Examples include indigenous technical knowledge for traditional irrigation (Adams, 1989; Adams and Anderson,
1988; Sutton, 1984; 1989), cropping systems (Allan, 1965; Belshaw, 1980; Richards,
1985), and the traditional use of plants (Knight, 1974; 1980; Osunade, 1994).
Investigations of local soil knowledge, and soil and land management practices reveals a diversity of expertise and specialization among smallholders. Examples include the use of plants as indicators of soil fertility (Knight, 1974; 1980; Osunade, 1988), local soil classification schemes (e.g., Almy et al., 1991; Carney, 1991; Carter, 1993; Edje etal., 1988; Fairhead, 1990; IFAD, 1992; Osunade, 1992a, 1992b; Sikana, 1993a, 51
1993b), and soil and land management such as terracing (Grove and Sutton, 1989;
9 Osunade, 1992a, 1992b, 1994; Reij, 1991).
Researchers and agricultural scientists interested in the practical, functional
elements of smallholder knowledge emphasize the parallels between and the
complementarity of indigenous and scientific knowledge and experience (den
Biggelaar, 1991; Pawluk et al, 1992; White and Jickling, 1995; Walker and
Wortmann, 1993). Correlations between conventional scientific soil analysis and classification and local soil knowledge and taxonomies have been demonstrated in
Sierra Leone (Richards, 1985), Kenya (Tabor et al., 1990), Mexico (Bellon and
Taylor, 1993) and Peru (Behrens, 1989). In Burkina Faso, local knowledge of the characteristics of land degradation was shown to be reflective of western, scientific logic (Lindskog and Tengberg, 1994). Local smallholder knowledge was also found to be strongly analogous to the knowledge derived through systematic scientific research for traditional agroforestry systems in China (Chandler, 1994).
The agro-ecological knowledge and experiences of smallholders and indigenous people are not, however, simply systems of agriculture and resource management that have been perfected over time and thus provide flawless alternatives to models advanced by scientific research. Indigenous agricultural and resource management practices may cause environmental degradation, and both sustainable and unsustainable management practices can coexist in the same agroecosystem (e.g.,
9 Historical studies in East Africa have also identified a variety of traditional soil conservation measures, although many of these systems have long since been abandoned (Sutton, 1984; 1989). 52
Bellon, 1995). Smallholders often know more about some aspects of their production
systems than others, and it is just as important to understand what farmers don't know
as it is to document what they do know (e.g., Chambers, 1983; McCorkle, 1989a).
For example, Bentley (1989: 25) notes that smallholders in Honduras "know more
about plants, less about insects, and less still about plant pathology."
Correspondingly, in Nepal researchers have pointed out the limitations of both
local knowledge (evolves over long-term, large areas) and conventional soil science
(short-term, site-specific) to mitigate erosion in the Himalaya mountain areas (Scott
and Walter, 1993). The researchers propose that the two "knowledge-bases" are
complementary, and they contend that the implementation of the two approaches in concert have reduced sedimentation and increased agricultural production in the study area. Attempts to categorize indigenous and western knowledge systems as unique and distinct are problematic and fraught with contradictions, and while they can be complementary, it is also useful to view them as occupying different and dynamic positions along a spectrum of human experience (e.g., Agrawal, 1995).
Indigenous knowledge is not uniformly systematized, static or immutable, but rather dynamic and adaptable (e.g., Browder, 1995; Lewis, 1993; Thompson and
Scoones, 1994). The knowledge and behavior of rural people and smallholders often change as their local agro-ecological systems change, and/or their larger socio- economic and political systems are transformed-and the change is not always a positive one. Agricultural production system are often sustainable given a certain set of conditions, and they may become unsustainable as a result of ecological 53
perturbations, the introduction of new pests into the agroecosystem or other external
influences (e.g., economic, political) (e.g., Goldman, 1995; Hansen, 1994). In the
Amazon basin, Behrens (1989) compared observed Shipibo land use with their
traditional soil classification and stated rules governing land use in order to portray
the effect of the adoption of cash cropping on the traditional systems. Behren found
that as the Shipibo farmers were further integrated into the market system and became
more dependent on rice cash cropping, they started to break their own cultural rules
about selecting the appropriate soils for the cultivation of food crops.
Additionally, not all smallholders can be considered good "farmers" or
resource managers, and the knowledge, experience and management skills often vary
greatly among smallholders. Researchers have underscored the socially-differentiated
dimensions of knowledge, and reveal that knowledge can vary among farmers in
general, and according to differences in gender, age, wealth, social position and
ethnicity (e.g., Brouwers, 1993; Fairhead, 1990; Colfer and Newton, 1989). The
importance of understanding the gender-differentiated roles in agricultural production,
and control of and access to resources, has been widely researched (Carney, 1993;
Fairhead, 1990; Fortmann, 1990; Rocheleau, 1988, 1991; Schroederer, 1993). The
agricultural production management constraints specific to women have been demonstrated in a number of African countries, including Gambia (Carney, 1993;
Scroederer, 1993), Zaire (Shapiro, 1989; 1990; Schoepf and Schoepf, 1988), and
Rwanda (Randolph and Sanders, 1992). Recent studies have also emphasized the 54 dynamic, site-specific and institutionally-situated aspects of local knowledge (e.g.,
Agrawal, 1995; Browder, 1995; Fairhead, 1993; Thompson and Scoones, 1994).
The movement to re-evaluate indigenous knowledge and to consider its potential contributions to development projects has arisen in response to the dissatisfaction with the results of green revolution technologies and sustainable development efforts, the socially-differentiated distribution of the benefits of those efforts, the persistent institutional bias against working with rural people found in many development projects and research centers, and the general disregard for local knowledge, experience and management practices (e.g., Berry, 1984; Biggs and
Farrington, 1991; Friis-Hansen, 1995; Pawluketal., 1992; Richards, 1983; Shiva,
1988; Vermeer, 1983; Yapa, 1993). In attempt to address some of these shortcomings, an ever-increasing number of researchers have advocated the greater inclusion of indigenous knowledge and smallholder participation in agricultural and resource management research efforts (e.g., Chambers, 1983; Chambers et al., 1989;
Kloppenburg, 1991; Richards, 1985). Some concerns have been expressed, however, that in the rush to identify useful ideas and technologies, the intellectual property rights of indigenous peoples will not been adequately protected and they will not receive any compensation or entitlements (Brush and Stabinsky, 1996; SAA, 1994;
Wijk, 1993).
The expanded involvement of farmers in research efforts may help to facilitate data collection, provide detailed information about existing production systems, enhance communication between researchers and farmers, improve the acceptability of 55 introduced technologies, and provide new insights into decision-making processes and
management practices. For example, research in Zimbabwe revealed that traditional farming practices exhibited fewer signs of soil erosion than practices on commercial
farms (Stocking, 1987). An improved understanding of indigenous knowledge, and experimental and management skills might also lead to the identification of agricultural and land management practices appropriate for wider diffusion, and local technology generating and diffusing capacities that might be further strengthened (den
Biggelaar, 1991; Edje et al., 1988; Lubaga, 1988; Reij, 1991).
The collaborative role that smallholders can play in efforts to mitigate
environmental degradation and improve agricultural production is increasingly recognized (e.g., Critchley et al., 1994; de Boef et al., 1993; Pawluk et al., 1992), but researchers and others also stress the need to provide smallholders with greater input into and control of the research process, as well as develop alternative research delivery systems (e.g., Scoones and Thompson, 1994). Unfortunately, despite the diversity of evidence, the knowledge, experience and management abilities of smallholders are neglected and underutilized in most research programs.
Summary
This chapter sought to review the literature relevant to the research project and to further elucidate some of the conceptual and theoretical issues that might strengthen this study's examination of the interrelationships between demographic pressures, environmental degradation, agricultural production and the adaptability of smallholders. The theory of agricultural intensification outlined a range of 56
smallholder responses to population pressures, while the cultural and political ecology
approaches provided holistic, multidisciplinary perspectives to examine human-
environment interactions. The regional political ecology approach in particular
provided a broad and flexible framework to examine the multi-scale, multi-factor and
temporal dimensions of this study.
To follow, the majority of the soil and water conservation programs in East
Africa have been characterized by a reliance on technological fixes and "top-down"
management approaches, simplistic and mechanistic analyses of the causes of
environmental degradation, a general failure to recognize the wider array of factors
impacting on land use and resource management, and an incomplete scientific
knowledge of tropical environments. The common reliance on macro-level studies
has also obscured the cultural, agricultural and environmental diversity present, and
the varying abilities of people to manage their production systems often have been
underestimated or disregarded altogether as well. Unfortunately, the preoccupation
with the resource management failures as opposed to successes of smallholders have
also contributed to an overwhelmingly negative outlook for agricultural production
and resource management in many regions of East Africa. The promulgation of pessimistic and "Cassandra-esque" perspectives has often had a cumulative and negative effect: the "disempowerment", or disenablement, of local people, whereby their ability to develop, test and implement potential solutions-with or without outside assistance-has been undermined (e.g., Thrupp, 1989; Wisner, 1989; Watts, 1983,
1987). 57
The efforts of western-scientific research to improve agricultural production for the majority of the population across Sub-Saharan Africa have produced mixed results, and the general shortcomings of conventional, top-down research programs underscore the need to re-examine the knowledge and experience of smallholders.
The knowledge of smallholders, however, is not a commodity simply to be adopted, coopted or freely incorporated into agricultural and resource management initiatives.
Smallholders are more than reserves of utilitarian knowledge that can be extracted and utilized by outsiders-they also have experience, ideas and innovative and experimental abilities to contribute to research and development efforts. Accordingly, smallholders should be provided the opportunity to participate in agricultural development efforts as collaborators and colleagues, and/or supported in their own management and research endeavors.
In Chapter Three which follows, the first half of the chapter provides a description of the demographic, agricultural and land use trends in Uganda. In the latter half of the chapter, a general description of the highland agro-ecosystem, including the physical environment, agricultural systems and population trends, in
southwest Uganda is provided. CHAPTER 3 UGANDA AND THE SOUTHWESTERN HIGHLANDS
Population Growth in Sub-Saharan Africa
Until the early 1990's, Sub-Saharan Africa was considered the only major
region in the world that had not yet experienced a decline in fertility levels; continued population growth for the region was projected well into the next century. Numerous demographic studies investigated the high fertility levels on the continent, and the
anomaly that Sub-Saharan Africa represented was often explained in terms of its unique cultural factors and a preference for large families (e.g., Caldwell and
Caldwell, 1987; 1990; Goliber, 1985; 1989; World Bank, 1986).
More recently, however, there is evidence that fertility has begun to decline in a number of regions in Sub-Saharan Africa (e.g., Foote et al., 1993; Locoh and
Hertich, 1994). Countries that have demonstrated a fertility decline include
Botswana, Ghana, southwestern Nigeria, Senegal, South Africa and Zimbabwe
(Caldwell et al., 1992; Caldwell and Caldwell, 1993b; Diamond and Rutenberg,
1995; Onuhua and Timeous, 1995). Evidence for similar demographic transitions have also been reported for Kenya and Tanzania (Caldwell et al., 1992; Murti and
Hinde, 1995; Robinson and Hardison, 1995; Tiffen, 1995).
Where fertility declines have emerged in a country, however, they have not necessarily occurred uniformly throughout the population. In countries like Nigeria,
58 59
there are intra-country variations in fertility declines. Additionally, the responses also
often differ across social groups (e.g., ethnic, rural/urban) within a given country
(Caldwell etal., 1992; Lockwood, 1995).
Demographic Trends in Uganda
While data indicating a decline in fertility levels have been reported for a
number of countries in East Africa, the population continues to grow in Uganda.
Population Growth
In Uganda, there has been little evidence of a decline in fertility levels.
Results from recent national population censuses, as well as the Demographic and
Health Survey (DHS) indicate that fertility levels in the country are still high,
approximately seven births per women during her childbearing years (GOU, 1992a;
Kaijuka et al., 1989). For more than thirty years, estimates for the average total fertility rate (TFR) in Uganda have fluctuated only slightly: from 6.81 in 1959 to 7.1 in 1991 (GOU, 1993a).
A number of factors have contributed to the high fertility levels, including a low level of contraceptive use, early marriage (over 50% of women marry before 18 years) and high infant and childhood mortality. Education has been found to play a moderating role on a woman's fertility: with increasing levels of education there is evidence for a trend towards later marriage and a woman with higher education has on average five births, as compared with seven births for women with only a primary education (Bertrand et al., 1993; Kaijuka et al., 1989; Lesthaeghe and Jolly, 1995;
Stephens et al., 1991). 60
Infectious Diseases
There a number of diseases that heavily affect mortality rates in Uganda,
including malaria, sleeping sickness and AIDS. Unfortunately, while much is known
about the epidemiology, and morbidity and mortality rates for malaria and sleeping
sickness, there is considerably less known about the impact of HIV/AIDS on
population growth. Consequently, the potential impact of HIV/AIDS on demographic
trends in Uganda is reviewed in this section.
The levels of HIV-1, HIV-2 viruses and Acquired Immune Deficiency
Syndrome (AIDS) in Sub-Saharan Africa are among the highest in the world
(Anderson et al., 1991; Becker, 1990; Caldwell et al., 1989; Caldwell and Caldwell,
1990, 1993a; Goliber, 1989; Miller and Rockwell, 1988; Shannon and Pyle, 1989).
In Uganda, the prevalence of HIV-1 infection amongst heterosexual populations is one of the highest in Africa (Anderson et al., 1991; Becker, 1990). The long-term demographic impact of AIDS in Uganda is difficult to predict, however, due to the unknown effect of education and prevention programs, and questions related to how soon a vaccine might be developed. Changing behavior patterns and the quality of available health care will likely have a significant impact on future trends as well.
The impact of the HIV/AIDS epidemic on population growth in Uganda and East
Africa is not sufficiently clear at this time, although the death toll and the general outlook remains tragic (Way and Stanecki, 1993).
The current HIV infection rates in Central and East Africa range from 1 % to
7-9% of the national populations (Becker, 1990), and AIDS appears to be the leading 61
cause of adult mortality in a number of urban areas in Africa (Anderson et al., 1991).
In Uganda, it has been estimated that as much as one-third or more of the general
sexually active urban population has the HIV-1 virus (Becker, 1990). Given these
tragic figures, Becker (1990) outlines a possible future scenario for Uganda: if
Uganda, with a population of roughly 16 million, had an HIV seroprevalence rate of
6% in the population and a conversion rate to AIDS of 2%, AIDS-related mortality
would be 19,200 people per year. Even with such a high death rate, however, and
the potential for negative population growth trends in some of the more heavily
impacted central districts, Uganda's overall population would continue to increase-
albeit at a lower rate (Becker, 1990; Goliber, 1989). Becker (1990) foresees no
circumstances where the population growth rate in the region would fall below 2% per year. More recent projections of mortality trends also suggest that while the
AIDS epidemic will continue to grow rapidly and effect death rates in Africa, population growth in the region will remain high (Bongaarts, 1996).
The impact on population growth rates aside, the HIV/AIDS epidemic has precipitated, and will continue to exact, heavy economic as well as social costs in
Uganda. First, the overcrowding of hospitals with HIV/AIDS patients means that people with curable diseases often do not receive adequate medical attention. Second,
AIDS is most widespread amongst the sexually active population (i.e., people aged
15-40) and it has hit the parental generation the most severely. There has been an increase in single-parent households and a growing number of children living with grandparents or other relatives. The loss of a large percentage of this mainly adult 62
age-group to AIDS underscores the trend towards a more youthful population in
Uganda. The youthful age structure in Uganda, where about 50% of the population
are below the age of 16, will mean that a large share of development resources will
need to be directed to the country's youth, especially for education and health
purposes (EPL, 1995).
Thirdly, the sexually active population and the prime-productive population is
one and the same. Correspondingly, it has been projected that the HIV/AIDS
epidemic will have a negative impact on labor availability in the coming years, especially in urban areas where the disease is more widespread (Way and Stanecki,
1993). Lastly, the disease has also been shown to have a higher death rate among women than men (Becker, 1990; Valleroy et al., 1993), and this could have an impact in areas where women provide the majority of the agricultural labor.
A number of studies have also recently examined the influence of HIV/AIDS on agricultural production in Uganda and East Africa (Abel et al., 1988; Barnett and
Blaikie, 1992; Barnett, 1994). In a study of three central districts in Uganda, communities affected by HIV/AIDS showed a sensitivity to labor loss and a "marked
shift to more basic and less varied food and other crop production" (Barnett et al.,
1995: 168-9). While the impact of AIDS in the central districts and urban areas of
Uganda has been widely studied in recent years, less is known about the prevalence or impact of the epidemic in the more rural regions of the country. The population densities and the farming systems of the southwestern highlands are also markedly different from those in the central districts, and thus comparisons between the regions 63 are fraught with complications. Additionally, while there is no data available to indicate the prevalence of HIV/AIDS in these more remote areas, the HIV/AIDS epidemic appears to be less severe in Kabale and Kisoro Districts as compared to the
south-central districts.
Research with a comparative value, however, has been conducted on the
influence of the HIV/AIDS epidemic on a variety of agricultural production systems in Rwanda (Gillespie, 1989). (Two of the multiple sites examined in the study are located near the southern borders of Kabale and Kisoro Districts, and are generally representative of the sites in this study.) At two locations in the highland regions of
northern Rwanda, it was judged that the agricultural production systems were susceptible to HIV/AIDS-related labor loss. The study determined that women were generally affected more severely by the disease than men. Considering that women contribute the majority of the labor in these highland farming systems, the impact of the labor-loss on agricultural production and food supplies might be disproportionately large given the actual incidence of the disease. The study also concluded that the labor-loss could have the potential to induce changes in local agricultural practices and cropping profiles, and consequently alter local diets. For example, changes in the production of labor-intensive legumes like beans and peas to tubers such as sweet and white potatoes could alter the nutritional make-up of local diets. Some concern was also expressed over the availability of labor for established resource management practices, such as terrace maintenance (Gillespie, 1989). 64
The highland region in Rwanda is characterized by smallholder agricultural production systems and high population densities very similar to those found in
Kabale and Kisoro Districts. Correspondingly, the agricultural production systems in
Kabale and Kisoro Districts are potentially sensitive to the HIV/AIDS epidemic and related labor-loss. However, little information exists to indicate whether the
HIV/AIDS epidemic has had, or is having, a similar impact on agriculture production and resource management in the highlands of southwest Uganda.
Refugees
The wars and civil unrest that characterized Uganda throughout the 1970' s and early 1980's forced thousands of people to flee the country and seek refuge in neighboring East African countries and even Europe. The social chaos also made
Uganda one of largest refugee-producing countries in the world during that period.
While it is difficult to derive accurate figures, it is estimated that Uganda has been the source as well as the haven for hundreds of thousands of refugees over the past three decades (Foster, 1989; Goliber, 1989; USCR, 1986; 1990; 1991a; 1992; 1993; 1994).
In the late 1980's, when the government formed under the rebel military leader
Yoweri Museveni (1986) began to stabilize, many Ugandans began to repatriate
(USCR, 1990). According to the World Refugee Statistics (See Table 3.1), there are currently no sizeable populations of Ugandan refugees remaining in any one country.
Over the last decade, Uganda has changed from a refugee-source country to a refugee-host country (Table 3.1). In 1995, refugees seeking haven in Uganda came from a number of countries in Eastern and Central Africa, including Sudan, Zaire, 65
Ethiopia, Somalia and Rwanda. Given the close proximity of Rwanda to the highlands of southwest Uganda, the recent chronicle of the Rwanda refugees has particular relevance to this research project.
Table 3.1. Refugees and Asylum Seekers in Need of Protection and/or Assistance.
Asylum Country # of Refugees/Source Country
Uganda 300,000 - Sudan
Uganda 15,000 - Zaire
Uganda 5,000 - Rwanda
Uganda 3,000 - Other
Other None - Uganda
Source : USCR, 1994. Note: Figures do not include refugees permanently settled in other countries.
People from the Great Lakes Highlands region, (i.e., Burundi, Rwanda and
Eastern Zaire) have historically migrated to throughout the highland region and southwestern Uganda for centuries, primarily in search of land, autonomy and peace
(Ngologoza, 1969; Rugyema, 1974; East African High Commission, 1959). More recently, in the early 1960's refugees fleeing a revolution and related ethnic conflicts in Rwanda fled to Uganda in great numbers (Foster, 1989; USCR, 1991b). Over the following two decades, many of the Rwandan exiles established settlements in the mixed-shrub and grassland environment that lies to the north and east of the southwestern highlands. 66
In the early 1980's, many Rwandan refugees were expelled from Uganda for
reasons stemming from fear, jealousy and related political factors. Although the
impact of the refugees on local communities and resources in Uganda over the last
three decades is poorly understood, the unremitting circulation of refugees between
Uganda and Rwanda undoubtedly created tensions between the Rwandan and Ugandan
Governments. In the latter half of the 1980's, however, under Museveni's new,
"open" government there was hope that the remaining "Rwandans" in the country
would become Ugandan citizens.
Unfortunately late in 1990, before a permanent solution could be achieved,
the situation exploded. A force of first and second generation Rwandan refugees
living in Uganda, and comprised primarily of former and/or active Ugandan military
personnel, invaded Rwanda (USCR, 1990; 1991a; 1991b; 1992). The fighting
between Rwandan government forces and the rebels continued intermittently up
through early 1994, when the rebels initiated an aggressive offensive campaign. The
fighting was fierce and widespread, and large numbers of Rwandan refugees fled into
southwest Uganda, including Kabale and Kisoro Districts. The rebels achieved a pyrrhic victory in mid- 1994, and many exiled Rwandans living in Uganda-some had even become Ugandan citizens-chose to return to Rwanda.
The invasion further exacerbated tensions between the two governments, caused a wave of Rwandan refugees to flee into Uganda, and provoked appeals within
Uganda to expel all people of Rwandan origin. Additionally, the invasion precipitated the closing of the Rwanda-Uganda border in 1990, both to halt the movement of 67
refugees and to stem the flow of supplies to the rebels. (The Rwandan government
accused Uganda of assisting the rebel forces.) Both formal and informal cross-border
trade were important sources of income for both Uganda and Rwanda throughout the
1970's and 1980's, but with the invasion the trade essentially halted. In Kabale and
Kisoro Districts, the border closure had a particularly negative impact on smallholders
and merchants who relied on an open border to sell and trade agricultural products
(e.g., dried beans) to Rwanda. While the evidence is primarily anecdotal and precise
figures are unavailable, many people in Kabale and Kisoro Districts claim that the
closure of the border has seriously undermined the local economy.
Agricultural Production in Uganda
Uganda is a country of verdant landscapes, with climates generally hospitable to human habitation, soils moderately productive and precipitation normally sufficient to support moderate levels of rain-fed agricultural production. Much of Uganda
experiences a pattern of bi-modal rainfall that can be attributed to its position within the Inter-Tropical Convergence Zone (Jameson and McCallum, 1970; Stephens,
1970). The northern and north-eastern regions of the country are semi-arid, while the eastern, central, western and southern regions normally receive rainfall sufficient to support mixed pastoral and agricultural systems~as well as the densest populations.
The major food crops cultivated in Uganda, planted in a variety of sole and mixed cropping systems, include bush and climbing beans, peas, grams, pigeon peas, cowpeas, chickpeas, sorghum, millet, maize, wheat, sweet potatoes, Irish (i.e., white) potatoes, squash (i.e., gourds or pumpkins) rice, groundnuts (i.e., peanuts), 68
cocoyams, cassava, bananas or plantains, and a variety of vegetables. Cash crops
include coffee, tea, cotton, sugarcane, oilseed and tobacco (McMaster, 1962; Mukasa
and Thomas, 1970; Thomas, 1970).
Generally speaking, the tea, coffee and sugarcane are grown on estates, while
other cash crops are often produced in smaller commercial ventures; smallholders do
engage in some mixed production of cash crops such as coffee or sugar cane. In
recent years, the production of food as a cash crop has also dramatically risen. The
move to grow food crops for the open-market has expanded the potential to provide
needed cash for smallholders, but concomitantly might also lead to the
impoverishment of local diets~by reducing both the quantity and the diversity of
foods for home-consumption (World Bank, 1993).
Agricultural-Based Economy
Uganda possessed a strong and prospering agricultural economy during the
colonial period and was often referred to as the "Pearl of Africa". With the arrival of
Idi Amin in 1972 and the political chaos that followed through the mid 1980's,
Uganda slide into a period of political, social and economic decline. 1 The general
chaos that characterized the period helped to undermine what was a previously
productive and vital agricultural sector, particularly in the rural areas of the country.
Agricultural production, particularly in the export sector, declined dramatically in the
1970's and early 1980's (Cleave and Jones, 1970; Jamal, 1991).
1 Additional information concerning the social, economic and political turmoil in Uganda during this period may be obtained in a variety of publications, including: Hansen and Twaddle, 1988, 1990; Mudoola, 1993; Mutibwa, 1990. Since the establishment of Museveni's government in 1986, however, peace
and stability have returned to much of the country. Agricultural production has
increased in recent years and there has been a trend towards the expansion of land
under agricultural production (Table 3.2) (WRI, 1994). Since the mid-1980's, there
has been an annual increase in agricultural and food crop production, and most
importantly there has also been an annual increase in per capita food crop production
(World Bank, 1995; WRI, 1994). More recently, crops with both food and cash
values such as maize have experienced dramatic increases in both production and
yields (World Bank, 1995).
Table 3.2. Land Use Trends in Uganda: 1,000's of Hectares.
1977 1982 1987 1992
Total Area 23,588 23,588 23,588 23,588
Land Area 19,965 19,965 19,965 19,965
Arable & Permanent Crop Land 5,538 5,840 6,705 6,770
- Arable Land 4,023 4,180 5,000 5,040
- Permanent 1,515 1,660 1,705 1,730
Permanent Pasture 1,800 1,800 1,800 1,800
Forest & Woodland 6,210 5,960 5,710 5,500
Other Land 6,417 6,365 5,750 5,895
Source : FAO, 1994a.
Agriculture remains an overwhelmingly important part of Uganda's economy, and export data from the last five years reflects the importance of this sector (EIU, 70
1995; EPL, 1995; FAO, 1994a; 1994b; GOU, 1994b; World Bank, 1993). In 1991,
the agricultural sector accounted for 51% or more of the Gross Domestic Product,
and for over 90% of exports (World Bank, 1993). The principal exports of Uganda
in 1993 were coffee, maize, beans and tea, respectively (EIU, 1995). Since 1986, the
country has been relatively stable politically, and is again becoming an important food
exporter in the Eastern Africa region. The agricultural sector is largely supported by
a rural population of some 2.5 million smallholders, 80% of whom cultivate less than
2 2 hectares each (World Bank, 1993). Unfortunately, the rural agricultural producers
are generally not well remunerated for their efforts, and the living standards are low.
For Uganda (1991), the per capita Gross National Product (GNP) was $US 163, while
the per capita Gross Domestic Product (GDP) was $US 142 (WPJ, 1994). 3
While agricultural production has increased over the last decade, many
observers believe that the combination of mounting population pressures,
environmental degradation (e.g., soil erosion, deforestation) and ineffective rural and
agricultural development programs threaten the sustainability of agricultural
production and resource use in the country (e.g., Dorsey et al., 1990; GOU, 1994b;
Yost and Eswaran, 1990; World Bank, 1987, 1993).
2 Some estimates are as low as 0.37 hectares of cropland per capita (WPJ, 1994).
3 For comparative purposes, the per capita GNP for Kenya and Tanzania are $US 350 and $US 95 respectively, while the per capita GDP for the two countries are $US 339 and $US 1 also 18 respectively. The exchange rate in Uganda is approximately 950 /= shillings Ugandan per $1 dollar U.S. 71
Land-Supporting Capacity
Land-supporting capacity is a recurrent theme in the literature addressing the
relationships between agricultural productivity and population growth in Sub-Saharan
Africa (e.g., Brown, 1994; Hardin, 1968; Ho, 1990; Mitchell, 1985). It has also
been a popular research topic within the discipline of geography (Bernard, 1985;
Bernard and Thorn, 1981; Bernard, Campbell, and Thorn, 1989; Grossman, 1977;
Mitchell, 1985; Stamp, 1958; Zelinsky et al., 1970). The term land-supporting capacity (alternatively called "carrying-capacity," "environmental potential," or
"population-supporting potential") refers to the number of people a specific area of land can support given its various environmental attributes and systems of management. Factors considered in the determination include a region's climate, precipitation, soil type, vegetation, potential cropping systems and growing seasons
(i.e., the "supply side"), and the demographic factors, including management, level of mechanization and use of agricultural inputs (i.e., "demand side").
Over the last decade and a half, the Food and Agriculture Organization (FAO) in particular has undertaken research to determine the potential land-supporting capacities of developing nations and their larger regions (FAO, 1978, 1980, 1983,
1991; Higgins et al., 1982; 1987; Higgins et al., 1984; Higgins and Kassam, 1985).
The main objective of the FAO studies has been to assess the physical potential of a country's land resources for food production, to determine appropriate land uses for a given region and consequently to determine the capacity for a given region or country to support its resident population. The "population-supporting capacities" for land areas are calculated for three different levels of agricultural "inputs", which are defined as follows (Higgins et al., 1982):
(1) Low-level inputs: assumes crop production on all available rainfed land
using only manual labor and hand tools, no fertilizers or pesticides
applied, and no soil conservation measures employed. The cropping
systems would be characterized by subsistence production, high labor
inputs, low capital intensity, fragmented landholdings and no infrastructure
technical assistance.
(2) Intermediate-level inputs: assumes crop production on all available rainfed
land using improved hand tools and/or draught implements, limited
fertilizer and pesticide applications, and the use of simple soil
conservation measures. The cropping systems would be characterized
by subsistence production with some commercial sales, high labor
inputs, intermediate capital intensity with credit available, occasionally
consolidated landholdings, and some infrastructure and technical
assistance.
(3) High-level inputs: assumes complete mechanization of agricultural
production and harvesting, full use of optimum genetic plant material,
and the necessary chemical and soil conservation measures on all
potentially cultivable rainfed land. The cropping systems would be
characterized by commercial-market sales, low labor inputs, high 73
capital intensity with credit available, consolidated landholdings, good
infrastructure support and expert technical assistance.
The population supporting capacity projected for Uganda in the year 2000, assuming a land area of 20 million hectares and a population of 24.3 million, was calculated from the FAO data according to these criteria (Higgins et al., 1982).
Although the FAO model collects intra-country, or regional, differential agricultural production potentials, the regional figures are aggregated to obtain a country-level figure. The projections (Table 3.3) indicate that with low-levels of inputs, Uganda
will surpass its population supporting capacity by the turn of the century; given the
current primacy of low-input agricultural production systems in the country, it is probable that this has already occurred.
Table 3.3. Uganda's Population-Supporting Capacity (People/Hectare), Year 2000.
Input Year 2000: Projected Year 2000: Potential Levels Population Density Population Density
Low-Level 1 0.5
Intermediate 1 2
High-Level 1 7
Source : Higgins et al., 1982.
A separate study by the World Bank calculated that at low-input levels, the population of Uganda already exceeded its land-supporting capacity well over a decade ago (Ho, 1990). According to the FAO projections, only with advanced 74
agricultural technology and intermediate- or high-levels of agricultural inputs will
Uganda be able to meet the future food needs of its growing population. The World
Bank's study echoes this viewpoint: "much of the population-resource imbalance
could be eliminated if the agricultural sector advanced to improved agricultural
techniques" (Ho, 1990: 41).
Not all researchers, however, are convinced of the validity of the FAO
prognostications for the land-supporting capacity of Uganda, or elsewhere. A number
of researchers have cautioned about the problems involved with applying the land-
supporting capacity models to the real world (Hare, 1980; Mitchell, 1985). For one,
the data available to determine a given country's agricultural production and crop productivity per land area are most often estimates and frequently inaccurate (Raikes,
1988). This critique is particularly pertinent to Uganda. Civil wars, social unrest and economic instability were characteristic of the country throughout the 1970's and mid-1980's and the widespread turmoil renders suspect most data collected during that period. The country-level data are also often produced by aggregating regional data, and as a result the intra-country differential agricultural potentials and constraints are often overlooked when discussing land-supporting capacity.
It is difficult to forecast the multiple and interrelated factors which might constrain agricultural production or even cause food shortages in a country, or conversely, stimulate the human resourcefulness which can contribute to agricultural innovation and improved production. While the land- or population-supporting models cannot account for the political instability, institutional constraints, poor 75
infrastructure, climatic change, or natural disasters which might negatively affect the
agricultural production of a country, or a region, they do have some heuristic value.
Carrying capacities in nature are not fixed, static, or simple relations. They are contingent on technology, preferences, and the structure of production and consumption [and] on the everchanging state of interactions between the physical and biotic environment. A single number for human carrying capacity would be meaningless because the consequences of both human innovation and biological evolution are unknowable. Nevertheless, a general index of the current scale or intensity of the human economy in relation to that of the biosphere is still useful. (Arrow et al., 1995: 92-93)
With the growing integration of social, economic, political and environmental
data into digital databases, computer decision-models and geographic information
systems (GIS), the models used to determine population-supporting or carrying-
capacity have become increasingly complex and dynamic (e.g., FAO, 1992b, 1993b;
Jagannathan et al., 1990; Smith and Dumanski, 1993). The most recent FAO
projections have a heuristic value and can provide an indications of the approximate
number of people a region can support under a given production system. They
cannot, however, provide absolute determinations of a country's land supporting
capacity and care should be taken that they are not used as absolute prescriptions for
agricultural or land management policies.
Purchased Agricultural Inputs
FAO considers the increased utilization of agricultural inputs (e.g., fertilizers, pesticides) a necessity to improve agricultural productivity and increase the land-
supporting capacity of countries throughout Sub-Saharan Africa. In Uganda prior to
1972, while input usage was low, there was a well established agricultural inputs supply 76
and distribution system. After 1972, however, many of the input-related businesses
closed and the distribution system essentially collapsed (GOU, 1994b). At present,
limited supplies of fertilizers and other agro-chemical inputs can be purchased in the
larger towns, and although the government ostensibly establishes retail prices for
fertilizers, actual store prices can be higher and they also vary widely between regions.
At present, the Government of Uganda (GOU) does not provide any type of fertilizer
credit or subsidy support programs to smallholders (FAO, 1994c).
Purchased inputs are most commonly utilized in large-scale plantation schemes,
particularly for cash crops such as coffee, tea and sugar cane. For agricultural producers
involved primarily in food crop production for household consumption, the use of
purchased agro-chemical inputs is more limited, although there is regional variation in
input use. In rural areas like southwest Uganda, the use of agricultural inputs among
smallholders is quite low (CARE, 1994; World Bank, 1992b). On the whole, the use
of purchased inputs such as improved seed, pesticides, herbicides and fertilizers make
up only a small fraction of the total cost of crop production in Uganda (GOU, 1994b;
WRI, 1994). Unfortunately, even if smallholders were willing, able and economically-
situated to increase their use of inputs, there remain serious supply and distribution
problems and poor infrastructure in much of the country.
For comparative purposes, recent fertilizer consumption in Kenya, Tanzania and
Uganda is presented in Table 3.4. The figures indicate the exceptionally low levels of fertilizer consumption in Uganda, especially when compared to neighboring Kenya and 77
4 Tanzania. While a variety of factors contribute to the differing levels of consumption
in the three countries (e.g., availability of hard currency for import purchases, favorable
exchange rates, institutional support, infrastructure), fertilizer use in Uganda is close to
an order of magnitude less than the other two countries.
Table 3.4. Fertilizer Consumption in East Africa in Metric Tons.
Fertilized 1988/89 89/90 90/91 91/92 92/93
Nitrogen (N) -Kenya 67242 45000* 57000* 51100* 46000* -Tanzania 26956 28700* 36678 33644 33078 -Uganda 107 250 92 500* 400*
Phosphorous (P205) -Kenya 51062 62000* 51000* 35600* 44000* -Tanzania 11107 16300* 11694 11318 9983 -Uganda 12 55 100* 300* 100*
Potassium (K2 0) -Kenya 6295 9800* 8000* 8700* 10500* -Tanzania 3036 4000* 2877 4717 4776 -Uganda 12 50 n/a 400* 300*
Source : FAO, 1994c. Note: # = Mixed and complex fertilizers sources; * = Estimate.
In terms of the future availability of fertilizer materials, it appears there will be
some shortages. For example, the current shortages of nitrogen and potash (i.e.,
potassium) fertilizer materials are projected to continue for the next few years for much of Sub-Saharan Africa. The projections indicate that only phosphate materials will be
4 Some data sources actually indicate virtually no imports of fertilizers over the last decade, and the extremely low to non-existent use of fertilizers and pesticides (World Bank, 1995; WRI, 1994); accurate information remains difficult to obtain. 78
in sufficient supply to meet demand in the coming years (World Bank, 1992a). Uganda,
however, does have the potential to provide some of the raw materials to meet its own
fertilizer needs. Some mineral-materials suitable for fertilizer production have been
identified, including phosphate rock, potash and volcanic ash. However, a domestic
fertilizer industry has not yet arisen to develop those mineral resources (Kisitu, 1991).
At present, the vast majority of the agro-chemical inputs and related technologies
are imported into Uganda, and this trend will most likely continue for the foreseeable
future (FAO, 1994b). Additionally, neither the development of a fertilizer industry nor an increase in imports will ensure that smallholders in the country will have equal and timely access to them, or will even be able to afford them. Although commercial farmers and select smallholders can benefit from the use of purchased agro-chemical inputs, at this time the increased use of purchased inputs is not a viable means of increasing agricultural production, and correspondingly improving resource management, for the majority of smallholders in Uganda.
The Highlands of Southwest Uganda
Under the early British colonial rule, the highlands of southwest Uganda were comprised the administration unit of Kigezi District, which was also included in the
Western Province. In 1974, long after the rest of the country had been reorganized into smaller districts, Kigezi District was divided into North and South Kigezi
Districts (Langlands, 1971). A few years later in 1980, North Kigezi was renamed
Rukungiri District and South Kigezi became Kabale District. In 1991, Kabale District was further divided into Kabale and Kisoro Districts-the focus of this study. 79
Natural Vegetation
A forest-savanna mosaic ecosystem characterized the majority of the medium
altitude environments in Kabale and Kisoro Districts, while moist lower montane
forests inhabited the higher elevations. However, the mosaic forest types have all but
disappeared, while few examples of the montane forests exist outside of protected
reserves or natural parks in the region (e.g., Cunningham, 1992; Cunningham et al.,
1993; Hamilton, 1969; GOU, 1993b). Wetland ecosystems were also a common
feature of the landscape in the highlands, and they were characterized by papyrus-
dominated ecosystems (e.g., Cyperus papyrus, Cyperus latifolias) and mixed
grasslands (e.g., Hyperrhenia-Pteridium climax communities) (Gibb and Partner,
1967; GOU, 1993b; Nyamaguru, 1992; Thompson and Hamilton, 1983). The vast
majority of these wetland areas have also been brought into cultivation in recent
decades.
Landforms. Geology and Soils
The topography in Kabale District is characterized by ridges of uplifted Pre-
cambrian rock that generally run along N/NW - S/SE axes, with elevations roughly
ranging from 1,500 - 2,700 meters. The ridges have inter-locking spurs and are
fluted with small step-valleys that gradually descend into swamps and wetlands. The
main ridges are separated from one another by steep-sided, U-shaped valleys
containing broad wetlands. Aside from a few lakes that have formed in flooded
valleys, there is little surface water or overland flow (Harrop, 1970; Hughes and
Hughes, 1992; Temple, 1971). Kabale District is roughly divided into two hydrologic 80
drainage basins; the eastern part of the district drains into Lake Victoria (altitude:
1,130 meters), while western portion drains into Lake Edward (altitude: 910 meters)
on the Uganda-Zaire border (GOU, 1967).
The geology in the region is part of the Karagwe-Ankolean (uplifted lake-beds)
system and includes folded sandstones, quartzites, shales and phyllites. With the
exception of a few areas of volcanic materials, most soils (e.g., USDA system: oxisols and ultisols) are old and weathered, with medium agricultural potential (e.g.,
Chenery, 1960; Doomkamp, 1968; Harrop, 1960; Langlands, 1974a; Stephens, 1970;
Yost and Eswaran, 1990). The landscape diagram in Figure 4.1 typifies the landforms present in Kabale District, and is representative of the landscape at both the
Kicumbi and Nyarurambi Parishes study sites.
In the southwest area of Kisoro District, the eastern extent of the Virunga volcanoes and a variety of volcanic features dominant the landscape. 5 The volcanoes rise as high as 3200 meters, and are surrounded by a wide volcanic plain interspersed with various volcanic intrusions, lava flows and small cones of pyroclastic materials.
There is little surface water or overland flow, and the region forms part of the Lake
Edward drainage basin. The geologic materials in the southwest are primarily volcanic, and the soils (e.g., USDA system: andisols) have moderate to good levels of fertility (e.g., Chenery, 1960; Harrop, 1960; Stephens, 1970; Yost and Eswaran,
1990).
5 The borders of Zaire, Rwanda and Uganda meet in the middle of the Virunga Volcanoe range. The Ugandan portion of the mountains are situated in Mgahinga National Park, home to a small and endangered population of mountain gorillas. 81 >
82
2 > I *1 I * 2 ' i 1 s SI a X T3 .3: 1 > 2>
•si > 8 S > S o w so Sx 3 S 3 U U, Jo
I 1 X 1
1 i i
1 5 - 2 E a B •5 «» ~ o to •
2 2 > = SO 9 0. > © 1 x w 3 5 « > I Si n > S «— r• S a o 2 1 I 3 §• i. 2 .S 2 J3 -5 60 so 3 ca I 1 8 d '5 "a HI s i II > 2 8 o 83
The gently sloping plain extends to the north and north east of the volcanoes
and is flanked in the north and east by the older, weathered uplifted highlands of the
6 Districts of Rukungiri and Kabale, respectively. The landscape diagram in Figure
4.2 typifies the general landforms on the volcanic plain, and is characteristic of the
landscape at the Muramba study site.
Climate
The climate in the highland region is equatorial yet temperate, with a mean
annual maximum temperature of 23.2 and a mean annual minimum temperature of
10. 1 degrees Celsius. Throughout the year, the daylength in Kabale varies less than
fifteen minutes from the equatorial daylength of 12 hours and 4 minutes. There is,
however, considerable cloudiness in the region, and the mean annual hours of
sunshine is only 5.1 hours/day-the least in Uganda (Jameson and McCallum, 1970).
The rainfall pattern in the highlands is bimodal, with significant precipitation
occurring from February to May, and from August to December~a pattern reflective
7 of the region's location within the Inter-Tropical Convergence Zone (ITCZ). The
January "dry" season is less pronounced than the one which occurs in June and July
(Harrop, 1970; Jameson and McCallum, 1970). Meterological records from the DAO
in Kabale Town (location: 1.15 S, 29.98 E; elevation 1,867 meters) were collated for
the period 1918 - 1993. For that 75 year period, the annual rainfall average was
6 For further descriptions of the geology and volcanology of the southwestern highlands, refer to the following works: Cahen and Snelling, 1966; Holmes and Harwood, 1937; Oilier etal., 1969; Stheeman, 1932; Wayland, 1956a, 1956b).
7 A graph of the annual rainfall pattern is provided in Appendix A.
85
§ § a 3 > > 3 5 o 9 > 60 a "3 3 ^ a 11 1 a > > 60 > .fi 0 3
0 > I II
60 O
i 1 i 3 d o I 13 £ e i 1 1 1 * -3 Ml d c 9 2 a 0 a o > 00 2 E •S i 5 o i |l si I I s > > 5 d HI 0 9 > k a.
60 I 1
o
g
d 1 > c a. .-8
0
IT! I [L, 86
1000 millimeters (mm). However, the year-to-year total rainfall variation over that
8 period was also considerable. For example, an annual rainfall low of 637
millimeters was recorded in 1921, while an annual rainfall high of 1485 millimeters
was reported for 1941 (DAO, 1994). Additionally, the monthly rainfall averages
were also determined for the period 1918 - 1993. Over that period, the wettest
months were April (139 mm of rainfall), March (118 mm), November (110 mm) and
October (104 mm), while the driest months were July (20 mm), June (28 mm),
August (54 mm) and January (61 mm)~a pattern reflective of the bimodal rainfall
pattern (re: ITCZ) described above. The rainfall intensity is reported to be low in
9 Kabale, with occasional intensive storms (Purseglove, 1945; Tukahirwa, 1995).
On many days, particularly in the rainy seasons, cool mists blanket the valley
bottoms until mid-day, providing additional moisture through condensation on
vegetation. The highlands also experience cold katabatic drafts at night, and frost
damage to vegetation has been recorded-although not in recent decades (Jameson and
McCallum, 1970). Although there is only one meterological station at Kabale town,
there appears to be significant variation in rainfall patterns and temperature in the
highland region. In the very southwestern portion of the highlands (e.g., Kisoro
District), the annual rainfall has been reported to be higher, occasionally reaching upwards of 1,500 mm/year (Hamilton, 1969). Records from Kisoro Town in 1992
8 A graph of annual rainfall totals from 1918 to 1993 is provided in Appendix B.
9 Daily precipitation data collected during a recent soil erosion research indicates that rainfall intensity is quite low for one area of Kabale District (Tukahirwa, 1995), and corroborates the similar findings of Purseglove (1945). and 1993, however, reveal annual rainfalls of only 635.3 mm and 772.1 mm,
10 respectively.
Agricultural Production Systems
The majority of the population (> 85%) in the highlands of Kabale and
Kisoro Districts are described as subsistence farmers (GOU, 1992a; 1992b), although many smallholders are also involved in some form of semi-commercial agricultural production. Semi-commercial activities, often include the brewing and selling of sorghum-beer, and the selling of crops and other goods in the local periodic markets.
Cropping systems
Throughout the highlands, there are two cropping seasons per year which correspond with the February to May, and the September to December rains.
Smallholders' food crops include sorghum (Sorghum bicolor), finger millet (Eleusine coracana), maize (Zea mays), wheat (Triticum aestivum), field peas (Pisum sativum), bush beans and climbing beans (Phaseolus vulgaris), Irish potatoes (Solanum tuberosum), sweet potatoes (Ipomoea batatas), squash (or pumpkins) (Curcurbita spp.), taro (Colocasia antiquorum), cooking and dessert bananas (Musa spp.), and vegetables (e.g., cabbage - Brassica oleracea); cash crops have included coffee
(Coffee arabica), tobacco (Nicotania tabacum) and pyrethreum (Hyde, 1975; Kisakye,
11 1987; McMaster, 1962; Parsons, 1960, 1970). Although there is no uniform
10 There is obvious intra-regional variation in rainfall, and additional research is required to determine the extent of micro-climatic variations in the highlands.
11 Other local crops include various amaranths (Amaranthus spp.), peppers (Capsicum spp.), African eggplant (Solanum macrocarpon). Cassava (Manihot esculenta) is grown 88 pattern of cultivation, most cereal (e.g., sorghum) and pulse (e.g., bush beans) crops are grown on the hillslopes, while vegetables and tubers (e.g., Irish potatoes) are grown in the wetlands.
In terms of cultivation practices, smallholders double-dig most hillside plots,
while raised beds are utilized in the wetland areas. During tillage, weeds are directly incorporated into the soil during double-digging, piled to decompose and incorporated at later time, burned, or piled along the contour to reduce soil erosion. Smallholders utilize two main methods to plant: the broadcast and the "chop and plant". With the broadcast method, seeds (e.g., sorghum, millet) are dispersed by hand across the topsoil; however, with this method seeds are often inadequately covered with soil and
the germination rate can be low, while plant density is also difficult to regulate. With
the "chop and plant", a small hoe is used to randomly cut into the soil, a few seeds
(e.g., beans, maize) are deposited into the hole, and soil is kicked to cover the seed.
The germination rate is higher using this planting pratice, but the method is much more laborious. Smallholders also weed plots once or twice a season, depending on the availability of labor. There are often labor shortages, however, and delays in weeding can both impact on crop yield as well as on the scheduling of later activities and preparation for the next seasons' planting.
Traditionally both sole- and mixed-cropping were practiced in the highlands, although mixed- and relay-cropping systems have become increasingly common. For example, beans are interplanted with a number of crops, including maize, sorghum
sparsely and only in the lowest elevations. 89
and Irish potatoes. Additionally, beans and maize may be planted as a mixed crop,
and after the beans are harvested sorghum may be relay-planted into the stand of
maize. While intercropping provides lower yields per crop, smallholders note that the practice allows them to both diversity production and reduce their risk if a particular crop fails. Finally, although intercropping has increased in both districts, sole- cropping remains a common practice in the highlands.
Livestock such as cattle, goats, sheep and chickens are raised in small numbers throughout the region (McMaster, 1962; Parsons, 1970). On the whole, the livestock do not provide a significant amount of protein to smallholders (protein comes primarily from pulses, e.g., beans), but do represent wealth, an extant savings account and potential bridal payments. The livestock are primarily free-ranging, and there are virtually no cut-and~carry systems in the region. For smallholders with
livestock, manure is generally applied directly to their plots, although its occasionally composted with other household refuse and crop detritus and then applied.
The majority of smallholders in both Kabale and Kisoro Districts do not commonly use purchased inputs (e.g., fertilizers, pesticides, herbicides, fungicides, etc.) in their agricultural systems. Some smallholders, however, do purchase improved seed when availability and utilize agro-chemicals on a few cash crops (e.g., fungicides for Irish potatoes). In Kabale District, one study found that out of 180 households, 34% had purchased some type of agro-chemicals and 53% had purchased seed (World Bank, 1989b), while another study reported that 11.5% of respondents (n 90
= 79) had used fertilizers (CARE, 1994). Neither study, however, provided any
information about the actual management of the purchased inputs.
Land tenure and land fragmentation
At the turn of the century, most of the land in the southwestern highlands was
loosely held and managed as communal property. Individual user rights (e.g.,
usufruct), rather than private ownership rights were customary and land use was
overseen by the family, clan and community leaders. The individual user rights were,
however, were commensurate with private ownership, and land could be managed
according to the user's preferences. Grazing land, on the other hand, was a
communal resource and continues to be managed as such (Ngologoza, 1969;
Rugyema, 1974).
In 1958, a freehold-title scheme was proposed in Uganda, whereby if
ownership for a piece of land could be established, a certificate of ownership could be
obtained. Although the proposal was widely opposed in the country, in Kigezi, where
the scheme was initiated as part of a pilot project, the private freeholds were
welcomed (Opio-Odongo, 1992). Later in 1969, the Public Lands Act provided for public or statutory leases of land.
In 1975, however, the nature of land tenure in Uganda changed again when via the Land Reform Decree, the government proclaimed that all land in Uganda was public land, and would be administered by the Uganda Land Commission. The Land
Reform Decree remains legitimate, although land tenure in Kabale and Kisoro
Districts is characterized by a confusing amalgamation of de jure (public) and de facto (traditional) systems (e.g., Mugisha, 1992; Kamugisha, 1993; and Tukahirwa, 1994).
Some individuals have taken advantage of this confusion and bought up and
consolidated smallholders' landholdings to establish large dairy farms (Mugisha,
1992), while others (e.g., village chiefs) have illegally sold off communal grazing
lands for personal gain.
Throughout the highlands of Kabale and Kisoro Districts, there has been a
history of fragmented landholdings (Kagambirwe, 1972; Langlands, 1974b;
Purseglove, 1950). This phenomenon remains characteristic of land use in the
highlands, and is a result of both population pressures and patrilineal inheritance
traditions. In the patrilineal system, landholdings were divided amongst a man's
children, predominantly his sons. This tradition endures and the land is further
divided with each passing generation (e.g., Byagagaire and Lawrance, 1958; Kateete,
1976; Kururagire, 1969; Tindituuza, 1976).
The majority of smallholders in the highlands own multiple, and generally dispersed plots that often include land on both hillsides and in valley bottoms. While
some plots can be adjacent to the homestead, others can be kilometers away. Most
smallholders own roughly from three to ten plots, although some smallholders have upwards of 30 or more plots. Those smallholders with insufficient land for cultivation purposes generally attempt to borrow plots, and a few individuals rent plots. The rental fees are paid either in cash or "in kind" (e.g., crops).
It is difficult, however, to determine the actual size of smallholder landholdings given the dispersed nature of the plots and the various forms of tenure 92 claims on them. Recent studies by the Government of Uganda and the World Bank have provided some indication of the small size of landholdings in the highlands. Of the rural households in the regions, 62 % have landholdings of less than one hectare
(Ha) each, while 23% have somewhere between 1 and 2 Ha each. In total, 85% of the rural households produce crops and raise livestock on holdings which average less than two hectares each (GOU, 1994b: 19). Additionally, in Kabale District the per capita available cultivable area has been variably reported as 0.39 Ha (GOU, 1994b) and 0.286 Ha (World Bank, 1993), which in either case represent some of the lowest available land areas in the country.
Labor
In Kabale and Kisoro Districts, traditionally the social structure was based on clans and clan lineages. Among the Bakiga in Kabale, clan units, comprised of multiple, male-headed households of related individuals, were often situated in a
specific "home" territory, and clans were often different from hill to hill (Edel,
1969). The Bafumbira in Kisoro were also organized into clans. Presently, while the tradition of clan affiliation has not altogether disappeared in either area, the social structure has been re-oriented around the household unit (e.g., Baxter, 1959; Edel,
1969; Motoyoshi, 1973; Yeld, 1969a, 1969b).
Polygamy was also prevalent in earlier periods, but due to the costs of
supporting more than one household and economic constraint, it has become less
common in both Kabale and Kisoro Districts. In cases where men do still have multiple wives, generally speaking each wife maintains her own house, cultivates her own plots and meets the household needs of herself and her children. The husband
normally provides each wife a house and agricultural land, and ostensibly some
financial support.
Throughout the highland agricultural systems in Kabale and Kisoro Districts,
women traditionally have provided the bulk of the agricultural labor. Presently,
women continue to shoulder the responsibility for managing most aspects of
agricultural production, including the production of food-crops for household
consumption. Women perform some of the tillage work, and the majority of the
planting, weeding, harvesting and post-harvest crop processing. Unfortunately,
women are also often unable to fully address many of the local problems due to
12 gender-inequalities in the region. Men, on the other hand, assist in the initial
heavy tillage of agricultural plots and harvesting of crops, although their labor
contributions are generally small when compared to those of women. The hoe is the
primary agricultural tool and among smallholders in the highlands, there are no
tractors and virtually no mechanization of any agricultural work, with the exception of
small grinding mills used to process locally grown cereals (e.g., maize, wheat).
Households with the financial means often hire labor to undertake initial plot
tillage, weeding, harvesting and other agricultural tasks. Women also often organize
themselves into womens' groups and they sell their labor for agricultural work,
including tillage, weeding and harvesting. The funds raised are generally shared
12 For example, local land tenure and cultural traditions effectively bar women from planting trees or implementing soil conservation measures without the consent of their husbands. the among women, or occasionally invested in community projects (e.g., schools,
health units). Both women and men are organized into cooperatives in many areas of
the highlands. These cooperatives are predominantly agricultural in orientation, and
most are devoted to agricultural production in the wetland areas. The cooperative
members generally cultivate group plots for the purposes of marketing the harvest
(e.g., Irish potatoes) and individual plots for household use.
Smallholders are also active in some market-oriented agricultural production,
which primarily involves the marketing of food crops (e.g., Irish potatoes, beans,
cabbage) as opposed to other common cash crops (e.g., coffee or tea) grown in the
region. While women cultivate most of the household food crops and many of the
market crops, men generally control the earnings from all crop sales, as well as the
overall household income. The agricultural earnings for many smallholders in the
region, however, are often insufficient to meet household expenses (e.g., school fees
for children) and many men have increasingly pursued off-farm employment in recent
decades. As mentioned, the per capita Gross National Product (GNP) for Uganda
(1991) was $US 163, and the per capita Gross Domestic Product (GDP) was $US 142
(WRI, 1994). These are national averages and the incomes for rural populations in
the highlands are generally even lower (e.g., GOU, 1994b; World Bank, 1993).
Unfortunately, while men have increasingly turned to off-farm employment in recent decades, there have been few job opportunities available in the region.
Smallholders report that throughout the 1970's and 1980's, many men in both districts were involved in the informal trade of agricultural goods and other products with 95
Rwanda and Zaire. However, the onset of the civil war in Rwanda, the closure of the
Rwandan-Ugandan border throughout much of the early 1990's (roughly late 1990 to
late-1994), and the continued instability in Zaire have made these unprofitable options
for many men. These difficulties have also meant that the markets for beans and
other crops have largely disappeared, or have been only sporadically accessible to
smallholders.
Demographic Trends
The Districts of Kabale and Kisoro comprise one of the most densely
populated rural regions in Uganda, and more than 85% of the population in the
highlands are described as subsistence farmers (e.g., smallholders) (GOU, 1992a,
1992b). Results from the 1991 census, presented below in Table 3.5, reveal that the population has continued to grow in recent decades in both Kabale and Kisoro
Districts.
The total fertility levels in the region indicate that population growth will continue at least into the next century. The total fertility rate (TFR) in the highland
region exceeds that of the national average of 7. 10. In Kabale District the TFR is
7.99, while in Kisoro District the TFR is 8.35 (GOU, 1993a). Unfortunately, it appears that these numbers will remain high for some; local health officials have noted that both contraceptive use and female education in Kabale and Kisoro Districts are below the national average. Women in both districts begin to have children at an early age (mid-teens), and over 50% of the female population above the age of six has never attended school (GOU, 1992a, 1992b). 96
Table 3.5. Population Growth in Southwest Uganda, 1969-1991.
Kabale Kisoro Uganda
ropuiation census District District (000 's)
1 , r 1 A sir ")v2 1) Land Area (Km ) 1,695 620 197,096
2) 1969: Population 288,602 114,798 9,535
Population Density 170 185 48 2 (People/Km )
3) 1980: Population 328,757 126,664 12,636
x ULmlaUUll ucilMly 1 QA Odd 04A/1
4) 1991: Population 417,218 186,681 16,761
Population Density 246 301 85
5) Average Annual Growth Rate: 1969- 1980 13 1.25 % 0.95 % 2.71 %
1980-1991 2.17 % 3.53 % 2.52 %
Source : GOU, 1992a; 1992b; 1994a.
According to the 1991 national census, in Kabale District, 93% of the entire
population, and 92% of all households were described as rural, while 7% of the population and 8% of the households were situated in Kabale Town (GOU, 1992a).
Among the rural population, the average size of a household was 5.1 members. Among the rural population, approximately 1/3 of all households were headed by women and
2/3's were headed by men (Table 3.6). The census describes subsistence agriculture as the basis of the economy in the region: 88.2% of male- and 89% of female-headed
13 The low rates of population increase during 1970's can be partly ascribed to out- migration, but it is most likely the product of an inaccurate census conducted during this period of political turmoil. 97
households (88.5% of total households) in the district reported agriculture as the main
source of household earnings and food supplies. Furthermore, 81.8% of all households
in the district reported the lack of any household cottage industries to supplement their
farming incomes. 14
Table 3.6. Characteristics of Rural Households in Kabale and Kisoro Districts.
Rural Households By Kabale Kisoro Sex of Household Head District District
Female 24,688 (32.5 %) 14,157 (36.1 %)
Male 51,330 (67.5 %) 25,037 (63.9 %)
Total 76,018 (100 %) 39,194 (100 %)
Source : GOU, 1992a; 1992b.
In Kisoro District, a similar profile of an rural and agricultural population
appears. In the district, 95.9% of the entire population, and 95.6% of all households were described as rural, while the remainder of the population resides in Kisoro Town
(GOU, 1992b). The average size of the rural household was 4.6 members, while slightly more than 1/3 of the households were headed by women and 2/3 's were headed by men (Table 3.6). The census describes subsistence agriculture as the basis of the economy in Kisoro District: 91.3% of male and 89.9% of female headed households (Total HH's: 90.4%) reported subsistence agricultural production as the
main economic activity. Furthermore, 83. 1 % of all households in the district
14 For example, carpentry, metal work, brick making, or agricultural processing. reported the lack of any cottage industries to supplement their farming incomes
(GOU, 1992b).
Both Kabale and Kisoro Districts have been characterized by intensive land use
and continued population growth since the early colonial period (ca. 1920). A
number of efforts were made to address population pressures and related land
shortages in the highland region, including the implementation of soil and water
conservation and resettlement programs in the 1940's and 1950's (Purseglove, 1950,
15 1951). In recent decades, in response to land shortages as well as the lack of
employment opportunities, Kisoro and Kabale Districts have also experienced a net
out-migration of people (GOU, 1994b; World Bank, 1993). Additionally, rural youth
in the highlands have become increasingly less interested in agrarian work and many
are moving to towns in search of other employment opportunities (viz., Bryceson,
1996). These population movements aside, the reduction in population growth has not
been significant enough to mitigate rural land use pressures in the highland region.
Summary
Despite evidence for a demographic transition in other countries in Sub-
Saharan Africa, recent census surveys indicate that the population of Uganda, and
particularly the Districts of Kabale and Kisoro, will continue to expand well into the
next century (GOU, 1992a, 1992b; Stephens, 1991). The available economic data
also indicate that agriculture will remain an important sector in Uganda's economy,
15 The soil and water conservation programs and resettlement schemes are discussed in greater detail in Chapter Five. while a substantial majority of the population in Kabale District and Kisoro Districts
will also continue to depend on agricultural production for their livelihoods for the
foreseeable future.
The highland landscapes of southwest Uganda are currently characterized by a
patchwork of agricultural plots-a pattern representative of a combination factors that
include high population densities, intensive agricultural production systems, patrilineal
inheritance customs, and colonial soil and water conservation programs. The high
population densities and land shortages in Kabale and Kisoro Districts have lead some
observers to suggest that the land supporting capacity of the highland region has been exceeded-particularly at low levels of purchased input use (e.g., Bagoora, 1988;
Ngabirano, 1993; Peden and Kakuru, 1993; Tukahirwa, 1988; Were et al., 1992).
At present, the majority of smallholders throughout Kabale and Kisoro
Districts have few opportunities to expand their use of purchased inputs. They have limited access to not only agro-chemical inputs, but improved seeds, agricultural tools and related technologies, and the supply and delivery systems for the various inputs are also undeveloped. Furthermore, most smallholders do not have the capital (per capita GDP: $143) to make additional investments in agricultural production if inputs were available. Given the lack of subsidies or other support programs, purchased inputs and related agricultural technologies will most likely remain prohibitively expensive for the majority of smallholders for some time. Finally, applied and adaptive research programs, and agricultural extension support are quite limited, if 100
if not often altogether non-existent in the highlands of Kabale and Kisoro Districts
(e.g., CARE, 1994; World Bank, 1992b).
Purchased inputs could provide smallholders with important nutrient supplies
and help increase crop production. Unfortunately, for the immediate future neither
purchased inputs nor the mechanization of production will be viable agricultural
management options for the majority of smallholders in Kabale and Kisoro Districts.
Future agricultural and resource management initiatives in the highlands will need to
consider management alternatives (e.g., organic inputs, green manures, agroforestry)
that might complement the utilization of purchased inputs (e.g., fertilizers) and that are also readiliy available and affordable for smallholders. The experimental and adaptive capacities of smallholders should also be more fully supported with extension assistance and increased access to agricultural tools and inputs.
In Chapter Four, which follows, the first section of the chapter provides a
detailed description of the three study sites, while the remaining sections describe the research methods employed throughout this study. CHAPTER 4 RESEARCH SITES AND METHODS
Research Sites
To better understand the diverse nature of the highland environments, the variations in smallholder knowledge and agricultural and soil resource management practices, as well as the range of land use responses to indigenous and exogenous land use pressures, the research was conducted at three different sites. In the process of selecting the study sites, field visits and some twenty group meetings were conducted with smallholders at eight prospective locations in Kabale and Kisoro Districts. 1
Only three of the initial eight sites were selected for further investigation in this study, and they were Kicumbi and Nyarurambi Parishes, located in Kabale District, and Muramba Parish, located in Kisoro District (Re: Figures 1.3 and 1.4).
The study sites were primarily selected to encompass a variety of soil types and geological materials, with differences in population density as a secondary factor.
The sites also vary to some extent in ethnic composition, agro-ecological conditions, general cropping systems, infrastructural development (roads, agricultural inputs, markets) and institutional support. Additionally, over 85% of the population in each of the three study-sites are considered to be smallholders involved in intensive
1 Sites were eliminated for a number of reasons, including their general inaccessibility, dangerous proximity to the Rwandan border and the Rwandan civil war, and uncooperative populations.
101 102 agricultural production. The range of differences in inter-site environmental and agricultural characteristics provided the basis for interpreting several variables but, given the multiple dimensions of variation, cannot represent a fully controlled natural experiment (e.g., Goldman, 1993b). A brief description of each site follows and their significant features are summarized in Table 1.1.
Site 1 . Kicumbi Parish, Kamuganguzi Sub-County, Ndorwa County, Kabale
District, (Figure 4.1): This site is located approximately 12 kilometers south of
Kabale Town. It is rural, agricultural and the population consists primarily of
2 smallholders; the population density, at 420 people/km , is mid-range between the three sites. The Bakiga are the major ethnic group in the region, and the main
language spoken is Rukiga.
The landscape is comprised of steep, fluted hillslopes and step-valleys in the west, and the wetlands of Katuna valley in the east; elevations roughly range from
1,800 to 2,225 meters (approximately 5,900 to 7,300 ft.). The hillslopes are moderately steep (10-20 degrees), although some slopes in the region are severe (30
- 40 degrees). With the exception of the hydromorphic soils in the wetlands, the soils are derived from older, weathered geologic materials and are characterized by moderate levels of fertility (Chenery, 1960; Harrop, 1960; Langlands, 1974a;
Stephens, 1970; Yost and Eswaran, 1990). Both the hillslopes and the wetlands are widely cultivated (e.g., sorghum, beans, potatoes), while the ridgetops and areas of rock-outcrops are generally used as communal pasture or woodlots. Small populations of livestock (e.g., cattle, goats) are also raised. 103
Figure 4.1: Kicumbi Parish
Source : GOU, 1965a; 1965b. 104
Site 2 . Nyarurambi Parish, Muko Sub-County, Rubanda County, Kabale
District (Figure 4.2): This site is roughly located mid-way between sites 1 and 2, at
approximately 35 kilometers west of Kabale Town. It is rural, agricultural and the population is predominantly one of smallholders; the population density, at 322
2 people/km , is the lowest of the three sites. Again, the Bakiga are the predominant
ethnic group and Rukiga is the main language.
The landscape is similar to that of Kicumbi Parish. It is characterized by steep, fluted hillslopes, step-valleys and bordered to the west and north by the wetlands of the Kigeyo Swamp; elevations roughly range from 1,950 to 2,175 meters
(approximately 6,500 - 7,200 ft.). The hillslopes are moderately steep (10 - 25 degrees), with a few steeper slopes. As in Kicumbi, the upland soils are generally derived from older, weathered geologic materials and are characterized by moderate
levels of fertility, while hydromorphic soils are found in the wetlands. There is, however, one unique geologic feature found here: the remnants of a cinder cone, as well as lava flows and related volcanic ejecta are located in the eastern portion of the parish (Chenery, 1960; Harrop, 1960; Stephens, 1970; Yost and Eswaran, 1990).
The hillsides are widely cultivated (e.g., sorghum, beans, potatoes), intermixed with a few scattered communal grazing areas and woodlots. The wetlands are characterized by small areas of cultivation and large expanses of the native- papyrus ecosystem. Small populations of livestock (e.g., cattle, goats) are also husbanded in the region. 105
Figure 4.2: Nyarurambi Parish
Source : GOU, 1965d. 106
Table 4. 1 . Characterization of Research Sites.
1. Kicumbi 2. Nyarurambi 3. Muramba Parish Parish Parish
Predominant Geology Old, Weathered Old, Weathered Volcanic and Soil Materials Materials Materials
t oaa ^icn Elevation (meters) 1800-2250 1950-2175 1825-2150
2 Land Area (Km ) 8.7 16.8 12.9
Population 3,350 5,423 8,059
Population Density 2 (People/Km ) 406 322 621
Major Ethnic Group Bakiga Bakiga Bafumbira
Source: GOU, 1992a; 1992b
Site 3 . Muramba Parish, Nyarusiza Sub-County, Bufumbira County, Kisoro
District (Figure 4.3): This site situated near Mt. Muhavura (Virunga Volcanoes) in the extreme southwest corner of Uganda, approximately 4 kilometers southwest
Kisoro Town. It is rural, agricultural and upwards of 90% of the population are
2 smallholders; at more than 620 people/km , it is the most densely populated of the
sites. The primary ethnic group is the Bafumbira, and the language is Rufumbira, which is closely related to Rukiga (and KinyaRwanda in neighboring Rwanda).
The landscape primarily consists of a plain sloping moderately downwards as it extends north and east away from the Virunga Volcanoes. The gently undulating plain is interspersed with lava flows, seasonal run-off gullies and small cinder cones, and elevations roughly range from 1,825 to 2,150 meters (approximately 6,000 -
7,000 ft.). Unlike sites 1 and 2, there are no significant wetland areas on the 107
1 1825 Legend
Major Roads
Parish Boundaries
-If Volcanic Cone
N t
.X % 2125 * ) ^1975 ^/(meters)
Figure 4.3 : Muramba Parish
Source : GOU, 1965c. 108 volcanic plain. The landscape is primarily characterized by intensive cultivated (e.g., maize, sorghum, beans, potatoes), with a few communal grazing areas. Small populations of livestock (e.g., cattle, goats) are husbanded. The soils are predominantly volcanic in origin, often interspersed with volcanic stone and ejecta and stone, and characterized by moderate to good levels of fertility (Chenery, 1960; Yost and Eswaran, 1990).
Research Methods
Preliminary Research
Research for this dissertation began with a preliminary visit to Uganda in the summer of 1993. The initial trip was primarily undertaken for the purposes of visiting the research area and investigating potential study sites, as well as identifying likely individual and organizational collaborators. Additionally, the trip was utilized to research and review academic literature and historical materials at Makerere
University (Kampala), various government ministries (Kampala and Entebbe) and the
Kabale and Kisoro District Offices (Kabale and Kisoro Towns). The compilation of a bibliography of sources that characterize the cultural, agricultural, ecological and land use history of Kabale and Kisoro Districts was a secondary objective of this study.
Field Research
The field research was undertaken in Kabale and Kisoro Districts from January
1993 to June 1994. The field data were collected utilizing a variety of research
methods, including participant observation, unstructured individual interviews,
unstructured, semi-structured and structured group interviews (e.g., focus groups), a 109
unstructured, semi-structured and structured group interviews (e.g., focus groups), a questionnaire survey and a conventional soil classification survey.
Most aspects of the fieldwork were undertaken with the aid of two field assistants who were from Kabale District and fluent in the local languages. The assistants, one male and one female, helped to identify key informants, mobilize the
focus groups, and conduct the questionnaire and soil classification surveys.
Participant observation
Participant observation was a fundamental methodological component of the fieldwork throughout the year and a half of the study. The method primarily involved lengthy periods of observation of, and interaction with, smallholders doing various
aspects of agricultural work. It also included observations of smallholders managing household activities (e.g., marketing, crop processing, cooking, fuel collection).
Given the distances involved and transportation constraints, participant observation
and other methods were generally employed for periods of two to five days per site, on a rotational basis between sites.
Unstructured and semi-structured individual interviews
Unstructured and semi-structured interviews were conducted on a variety of topics across the three sites throughout the duration of the study. Unstructured interviews were conducted to investigate a loosely-specified topics in an informal and open manner; this approach was primarily employed with individual smallholders and others who could be easily visited again to continue discussions. Semi-structured
interviews were conducted in order to address a more specific set of questions (i.e., 110
"interview guide") while allowing some latitude for further questioning; this approach was most appropriate for interviews with time constraints and were thus often employed with individual professionals (Bernard, 1994).
A variety of topics were discussed in the unstructured and semi-structured interviews. Firstly, discussions about the history and present status of the Soil and
Water Conservation (SWC) By-Laws were conducted with district level agricultural, forestry officers and extension agents, independent and university researchers, and smallholders. Secondly, these individuals were questioned about their opinions of the benefits, drawbacks and current effectiveness of the SWC By-Laws, as well as their perceptions of environmental degradation, and agricultural and resource management problems in the highlands. An effort was made to interview equal numbers of male and female smallholders, although women were often reluctant to speak openly and frankly. Additionally, the majority of the professionals contacted were male.
Semi-structured group interviews
The majority of the interviews with smallholders were conducted in group settings, or focus groups (Bernard, 1994; Chambers et al., 1989). To arrange the group meetings, the field assistants generally spread the word through the village chief or well-known smallholders, and they also delivered notices to local churches
and women's groups. Across the three sites, approximately 85 focus groups were conducted over the course of the study. The size of the groups varied from 8-20 individuals, but occasionally ranged as high as 30-40 participants. The majority of the group meetings were mixed-sex and mixed-age in composition. However, men Ill dominated many of the discussions and women were often reluctant to voice their ideas and opinions. Consequently, single-sex groups of females were organized in order to provide women with the opportunity to more fully participate. Mixed groups of elders were also organized to investigate historical issues and agricultural and land
use change.
The informal focus groups (single- and mixed-sex) were used to investigate a broad range of subjects. The primary focus of the investigations were the local soil classification systems, soil resource management practices, cropping systems and constraints to improved agricultural production and resource management.
Additionally, mixed-sex groups of elders (age > 65 years) were arranged to discuss the history of the soil and water conservation programs in the region. In many of the
semi-structured group interviews, elements of structured interviews such as rating and ranking exercises were also employed. For example, smallholders were asked as a
group to rank and/or rate the fertility and erodibility levels of soils, the preferred crops for a given soil type and the constraints to improved agricultural production and
resource management. These latter methods were also used during a limited number of individual interviews.
The results of these various types of interviews are presented in Chapters Five,
Six and Seven.
Questionnaire survey
After the completion of more than one year of fieldwork using informal
methods, a questionnaire survey was conducted in March and April of 1994. The 112 survey was undertaken to complement the informal and participatory methods
described above, as well as to test the representativeness of those methods. The questionnaire survey was developed from the information provided in the informal individual and group interviews, and further refined after conducting a pre-test of 30 respondents. (The questionnaire survey is provided in Appendix C).
The survey was conducted with the assistance of one university-trained survey coordinator, the aforementioned field assistants and 21 enumerators that were divided
into three teams (i.e., one team per parish). The team of enumerators in each parish were selected from among local individuals (primarily smallholders) who successfully completed a two-day survey training workshop.
Table 4.2. Survey Sample Framework
1. Kicumbi 2. Nyarurmabi 3. Muramba Parish Parish Parish Total
Population 3,350 5,423 8,049 16,822
# Villages 10 12 11 33
# Households 740 1052 2,053 3,845
Stage 1: (5) (5) (6) 16 # Villages a. Omurindira a. Bugunga a. Kanyenka Sampled b. Kasenyi b. Katosa b. Ruhango
c. Nyamabale c. Kanyamatembe c. Nango d. Omurujanja d. Rushambya d. Murinzi
e. Rushongati e. Rutoga e. Ruhandanzovu
f. Burere
Stage 2: 67 95 185 347 # Households Sampled 113
The survey population was defined as the total number of households in the parishes of Kicumbi, Nyarurambi and Muramba. 2 The survey population totaled
3,845 households. A sample of 9% of the total population of households, or 347 households, was selected in a two stage process. (A summary of the sample
framework is presented in Table 4.2.) In the first stage, five villages in Kicumbi and
Nyarurambi parishes, and six villages in Muramba parish were selected. (The number of villages per parish was purposively selected, primarily due to the logistical and distance constraints involved in the survey work.) In the village-selection process, the villages were weighted according to their size (# of households) and selected with variable probability, wherein a large village (a village with many households) stood a greater chance of being selected than a smaller village.
In the second stage, a sample of 9% of the total number of households in each parish were selected from the villages selected in stage one. (For example in
Kicumbi Parish, 67 respondents were selected from five villages.) In the household- selection process, the pre-selected villages in each parish were weighted according to their size, or number of households. The households were then selected with variable probability, wherein households in a large village (one with many households) had an increased chance of being selected over those households in a small one (Cochran,
1977). After respondents who were absent or unwilling to participate were removed from the pool, the survey was administered to 347 households/people.
2 A list of households for each village in each parish was compiled by the parish chiefs, a task facilitated by updating household lists used in the 1991 national census. >
114
A household unit was defined as all individuals who lived and ate together, and shared financial and material resources. In the cases where men had multiple
wives who managed separate households, each female-headed household was
considered to be an independent household regardless of the man's primary residence
(e.g., Casley and Kuma, 1988). The household head was the preferred respondent,
but if he or she was not present then the survey was administered to another adult (
18 years of age) household member. If no adults were present during the initial visit,
then a second or third visit was made until an adult was contacted. Where adults
were recently deceased or absent for long periods, a replacement household was
identified. In instances where the respondent was determined to be uncooperative, or
there were too many errors made during the administration of the survey, the survey
was rejected and replaced with a household from a previously defined reserve list.
Finally, when the survey data were "cleaned" and coding errors and missing
values were uncovered, the anomalous responses were rejected. If a survey had a few
mistakes (1 - 10) it was included in the sample, while a survey with a large number
of errors (11 >) was rejected. Consequently, the number of responses for a given
question do not necessarily reflect the size of the original sample population. The
survey data were analyzed using the SAS statistical software package. The results are
presented in Chapters Six and Seven.
Soil classification survey
A conventional soil classification survey was conducted in each of the three
parishes in order to make comparisons between the western, scientific soil taxonomies 115 and the smallholder soil classification systems. The survey was undertaken with two field assistants, two staff members from the Soil Science Section, Kawanda
Agricultural Research Station (Kampala) and a small (5-8 members) group of smallholders per parish. At each of the three parishes, the smallholders identified via group consensus typical examples of the various soil types according to the local classification scheme.
After the soil types were identified, soil pits were dug to a depth of two meters, with the exception of those areas where bedrock was close to the surface.
For each soil pit, the soil profile was described according to both the USDA and FAO soil survey specifications (e.g., number and width of soil horizons, soil color, particle size, etc.) (FAO, 1977, 1993a; Langdon, 1991; Soil Management Support Services,
1987; Soil Survey Staff, 1975, 1993, 1994). Soil samples were taken from each horizon of each soil profile, and analyzed in the lab of the Soil Science Section,
Kawanda Agricultural Research Station, Kawanda, Uganda. In total, 72 horizon samples were analyzed for the purposes of soil classification. The soil characteristics that were tested in the lab included soil pH, organic matter content, cation exchange
capacity (CEC), and percent sand, silt and clay. With the data collected from the completed soil profile descriptions and the laboratory analysis, the soils were categorized according to the USDA and FAO soil classification systems, and the three systems are then compared (e.g., FAO, 1977; Soil Management Support Services,
1987; Soil Survey Staff, 1993, 1994). The results are presented in Chapter Six. 116
Additionally, three top-soil samples were collected at each of the soil pits.
The samples were collected from top-soil (<20 cm in depth) within five meters of
each of the soil pits (twenty-three pits in total), and analyzed at Kawanda's soil
3 laboratory. The lab results were utilized to classify each of the soils according to the
Fertility Capability Classification System (FCC). The FCC system emphasizes the
chemical and physical properties of soils, and is particularly oriented towards the
4 agronomic management of soils (Sanchez et al., 1982). These results are also presented in Chapter Six, as well as Appendix G.
In the following chapter, Chapter Five, an overview of colonial interventions in the highlands, and a discussion of their impact on current agricultural and land use
systems in Kabale and Kisoro Districts is presented.
3 The soil analysis methods employed were as follows: - a) soil pH 2.5:1 H20 (Rhoades, 1982); b) soil particle size/distribution - hydrometer method of soil mechanical analysis (e.g., Landon, 1991); soil organic matter and extractable soil P, Ca and K - Rapid Routine Soil Analysis (Foster, 1971).
4 Given its uncommon usage, the nomenclature of the FCC system is provided in Appendix F. CHAPTER 5 COLONIAL INTERVENTIONS IN THE HIGHLANDS OF SOUTHWEST UGANDA
Introduction
In the latter years of the 19th century, European explorers and missionaries
first began traveling through southwest Uganda and the surrounding Great Lakes
1 Highland region . Around the turn of the century, the highlands of southwest
Uganda were variously divided and exchanged between the European colonial powers
(i.e., Belgium, Britain, and Germany)~a result of boundary disputes as well as local and international political conflicts (e.g., GOU, 1967). The British established administrative control in the southwestern highlands in 1913, and thereby introduced external-administrative control in a previously autonomous, and locally-ruled region
(Hopkins, 1968; GOU, 1967; Mateke, 1970).
Under the British colonial authority, dramatic change was to become a
common feature of all aspects of political, social and economic life in the highlands.
The Kigezi District Administration implemented a number of policies and management plans that would affect the development of the region and dramatically alter land use and management in the highlands. They included the designation of
Kigezi District as a labor reserve, the implementation of resettlement schemes to
1 The Great Lakes Highlands is imprecisely defined as the mountainous regions of Burundi, Rwanda, Eastern Zaire and Southwestern Uganda.
117 118
alleviate population pressures, the development of smallholder vegetable cooperatives
and the implementation of a variety of agricultural and land management regulations
to mitigate land use pressures in the highlands.
Labor Reserves
In the early 1920's, the British colonial state developed a number of official
policies to enhance the production of cash crops in the districts centrally located and
peripheral to the capital of Kampala. For example, one policy attempted to
discourage cash-crop production in less favored, outlying areas so as to ensure a
continued flow of labor to the more central areas where cash crops such as cotton
were produced and more easily processed. One area ideally suited to provide labor to
the cash crop production efforts, given that it was outlying and rural, with poor
infrastructure and dense populations, was Kigezi District. Eventually, Kigezi District
was designated a "labor reserve" for cotton, coffee, tea and sugar estates operating in
the western and central districts (i.e., Buganda areas) (Powesland, 1954; Wrigley,
1959). There was even a Kigezi Recruiting Agency (later the Kigezi Voluntary
Employment Bureau) established at Kabale town to recruit labor; migrant laborers often left the region for a few months but many also stayed away for years
(Turyagenda, 1964).
The British colonial state discontinued its formal policy of suppressing cash
crop production in the labor-exporting areas in 1926, shortly after its initial implementation. The policy, however, implicitly assigned a "less- favored" status to
Kigezi District that has proven difficult to overcome. The colonial state changed its 119
formal policy, but at the same time it did little to encourage or support agricultural
development in the outlying areas. The lack of transportation infrastructure and
markets for agricultural products were serious obstacles for smallholders in outlying
areas who desired to sell any agricultural products. Nor did the colonial state actively
discourage people from pursuing work in the central districts.
Collectively, the lack of employment and market opportunities coupled with
poor socio-economic conditions and land pressures often induced men in Kigezi
District to seek work other districts. For example, migrants from Kigezi sought work at the copper mines at Kilembe in western Uganda starting in the 1930's and continuing through the 1960's. Kigezi residents also continued to seek work on the
sugar estates and tea and coffee plantations in the central districts well into the 1960's
(Erlich, 1965; Jorgensen, 1981; Langlands, 1971; Lury, 1976; McMaster, 1968;
Turyagenda, 1964; Wrigley, 1959).
The British colonial state was also characterized by a pervasive discrimination in assessing taxes against rural households. For example, many men in the labor reserves like Kigezi District were induced to migrate to the cash crop-producing areas in order to pay the colonial poll tax (Young, 1988; Jamal, 1978). Workers who migrated from the designated labor reserves also often paid poll taxes in their area of work, a trend which further enriched the "host" cash-crop area and the central government, and further impoverished or "underdeveloped" the labor reserve area.
As migrant laborers were generally responsible for providing their own living expenses and wages were meager, few laborers were accompanied by their families. 120
Given the costs of the poll tax and living expenses in the work-area (laborers were
generally required to pay their own way) and a meager salary, most migrant laborers
were unable to send remittances home to their family. The migrants endured a
variety of hardships during the long periods away from home (e.g., poor working and
living conditions), but the families who remained behind often suffered as well. In
areas like Kigezi District, although women undertook the majority of the agricultural
work, men were still an important source of labor and their absence often meant an increased work load for household members (Jorgensen, 1981).
The colonial state did reveal concern over agricultural production in the labor reserves in terms of the implications for local food shortages. In order to avoid endangering local food supplies and inducing or exacerbating famines, the colonial state mandated that not more than 10% of the adult male population could emigrate from a labor-producing area (Erlich, 1965; Jorgensen, 1981; McMaster, 1968;
Wrigley, 1959). Reports from the labor-producing areas in 1952, however, revealed the general ineffectiveness of this strategy: in Kigezi District, for example, 46% of adult males were found to have emigrated. In other areas like West Nile, in the
north, the shortage of male labor was so great that it actually induced changes in the cropping system (Jorgensen, 1981). While the impact of the colonial state's migrant labor policies on agricultural production and soil conservation practices in Kigezi cannot be fully determined, the implications are that the high degree of male out- migration translated to additional labor burdens for the women and children struggling to meet household food needs (Turyagenda, 1964; Yeld, 1969a, 1969b). 121
Although the colonial state underwent policy changes after the second world
war and eventually encouraged the production of cash crops for export purposes
throughout Uganda, infrastructural support and research assistance in the central
districts were generally superior to that available in the more remote and rural
districts. Additionally, agricultural research efforts were (and remain) biased in favor
of export crops and the introduction of "improved" agricultural technologies, and against traditional food crops and management practices (Opio-Odongo, 1992;
National Research Council, 1996). In Kigezi District, for example, the traditional crops such as sorghum and millet have not benefitted from new varietal introductions, or adaptive research programs focused on improving production under smallholder conditions.
A gricultural Cooperatives
During the colonial era, however, despite the minimal agricultural investment,
Kigezi District was able to produce small quantities of crops for export to other regions of Uganda and East Africa. Crops that proved marketable include maize, sorghum, finger millet, wheat, some barley (destined for the beer breweries in Jinja), bush beans and field peas, (Martin, 1963; Masefield, 1962; Wrigley, 1959). Another notable exception, based around Kabale town in Kigezi District, was the smallholder agricultural cooperatives which produced vegetables for Kampala and other East
African urban markets.
The Kigezi Vegetable Growers Cooperative Union, the primary cooperative which had branch societies in each county, was established in 1961 and flourished 122 through the early 1970's (Anusionwu, 1977; Scherer, 1969). The field crop and vegetable trade decreased dramatically, however, with the onset of economic and
2 social decline that occurred throughout the country in the early 1970's. Many of the wetland areas utilized for vegetable production in the 1960's were also eventually converted to large-scale dairy farms during the 1970's; smallholders lost access to both land and economic opportunities (Rutunga, 1992). The highland region, particularly the peri-Kabale town area (located in Kabale District), experienced economic growth again during the late 1980's--primarily a result of informal trade with Rwanda~but that boom ended in October, 1990 with the onset of the civil war in
Rwanda. 3
Resettlement Schemes
The British colonial administration first expressed concerns about overpopulation and land degradation in Uganda-particularly in highland regions like
Kigezi District-in the late 1920's and early 1930's (e.g., Brasnett, 1936; Fiennes,
1939; Wayland and Brasnett, 1938). The colonial administration in Kigezi District was particularly concerned with increasing land use pressures, especially given the rapid rate at which the population was expanding in the region. Between 1931 and
1948 population increased by an estimated 75% (Anderson, 1984). The rapid population growth was a result of both a high rate of natural increase (estimates range
2 This decline was precipitated by Idi Amin's successful coup in 1971.
3 Since the early part of this century, the informal trade of agricultural products such as bush beans with neighboring Rwanda and Zaire has been an important part of the local economy. 123 from 3.2% to upwards of 4% during in the 1930's and 1940's), as well as in- migration from Rwanda and Zaire (Denoon, 1971; Kabera, 1982; Langlands, 1971;
Ngologoza, 1969).
In the 1930's and 1940's, land degradation, land fragmentation and even landlessness were prominent concerns with the British colonial administration in
Kigezi District. The colonial administration implemented a soil and water conservation program in 1945 (see following section), but believed it would not suffice to mitigate population pressures on the land base in the highlands.
Consequently, a voluntary "resettlement" program was introduced to divert some of the population from southern Kigezi District and to settle them in less densely populated areas to the north and west (Langlands, 1971; Purseglove, 1950, 1951).
The first group of "migrants" moved from South Kigezi to North Kigezi (now
Runkungiri District) in 1946, and other groups continued to resettle in North Kigezi for a number of years. Later groups also resettled in Ankole (currently, the Districts of Mbarara and Bushenyi) and Toro (now Kabarole) in western Uganda up through
1955 (Ngologoza, 1969; Purseglove, 1950, 1951; Yeld, 1965, 1969a). While the
initial groups resettled voluntarily, in later years people were increasingly "induced" to move (Kabera, 1982; Kakarikawe and Yeld, 1966).
For a number of years, the sponsored out-migration was considerable: in 1949 alone, the District Agricultural Officer estimated that some 15,000 people were
resettled (Purseglove, 1950). Overall, it is estimated that between 1946 and 1976, there were upwards of 80,000 sponsored migrants from Kigezi (Kabale) District, 124
although many more people moved voluntarily and without government assistance
(Kabera, 1982). In the end, however, the resettlement programs did not effectively
reduce land pressures in the highland region. Even with the out-migration of laborers
and some of their families, the organized resettlement programs and voluntary
migrations induced by land shortages, the population in the Kigezi highlands grew
throughout the colonial period, and the growth has continued unabated to the present.
Land Management Initiatives
The colonial administration in Kigezi maintained a longstanding concern about
the expanding population in highlands, and the implementation of a district-level soil
and water conservation program actually preceded the resettlement schemes. The
origins of the concern over land degradation during this period can be traced to the
environmental disaster in the United States known as the "Dustbowl", where extensive
soil erosion ruined thousands of acres of farmland in the 1930's (e.g., Jacks and
Whyte, 1938). The developing awareness of the hazards of soil erosion and the
potential threat to agricultural production and food supplies among the "native populations" was reflected not only in Uganda, but throughout the British colonial
empire (Anderson, 1984; Blaikie, 1985; Stockdale, 1937; Wrigley, 1959). It was in
Uganda, however, that the British colonial system administration was to achieve one
of its greatest soil conservation "successes".
Soil and Water Conservation Programs
The British colonial administration originally expressed concern that poor agricultural management practices and decreasing fallow periods were leading to soil 125 erosion and declining soil fertility in the 1930's. This emerging concern ultimately lead to the implementation of soil conservation measures in the southwest highlands in the late 1930's (Martin, 1945a, 1945b; Purseglove, 1945, 1946; Tukahirwa and Veit,
1992). More specifically, in 1937 the first grass bunds (i.e., strips) of napier grass
(pennisetum purpurem, a.k.a. elephant grass) were introduced into smallholder plots to control for soil erosion, and by 1945 a comprehensive land management plan was implemented at the district level. The package of land management regulations was formalized as the Soil and Water Conservation (SWC) By-laws and smallholders throughout the district were expected to adopt them as soon as possible. The SWC measures are summarized in Table 5.1.
Table 5.1. Soil and Water Conservation Measures.
Mandated Prohibited Physical Constructions Activities
1. Terraces 1. Grass burning
2. Grass bunds 2. Cultivation within 9 feet of (e.g., contour strips) watercourse or roadway
3. Intra-plot trash or 3. Use of gullies or furrows plant detritus lines to demarcate plots
4. Run-off channels 4. Cultivation of erosion- prone areas, e.g. steep 5. Soakway (Overflow) pits slopes
6. Check dams
Source : DAO, 1992, 1993. 126
The Soil and Water Conservation By-laws actually required considerable labor
investments from smallholders, and regulations included the planting of crops along
the contour of the hillside, the establishment of grass bunds (strips) along the down-
slope ridge (i.e., riser) of a plot, the construction of modified terraces (versus radical-
bench terraces), as well as the prohibition of a number of common practices. The
construction of terraces along the steep hillsides was one of the most notable land
management technologies mandated under the SWC By-laws~and the most exacting
4 for smallholders. The terraces were constructed according to strict specifications:
1 . vertical drop across the bench of the terrace is not to exceed
6 feet [1.83 meters];
2. bench widths of 16 yards [14.63 meters] are recommended
for slopes of 6 to 15 degrees, bench widths of 12 yards
[11 meters] are advised for slopes of 15 to 20 degrees,
and slopes above 20 degrees are considered unsuitable
for cultivation;
3. if land shortages are severe, plots of greater than 20 degrees-
slope can be constructed with a bench width of 12 yards
[11 meters]. However, these plots are not to be
cultivated for more than two years, after which they are
to be fallowed for four years; and
4 The steepness of hillsides generally range from 10-20 degrees slope, although occasionally they can exceed 30 degrees slope. 127
4. grass bunds on the front-riser are to be 3 feet [0.9 meters] in
width and at least 1 foot [0.3 meters] in height. (DAO,
1992; Martin, 1945a; Purseglove, 1945, 1946).
The Soil and Water conservation By-laws in Kigezi District were administered
at the district-level by the colonial administration staff, and at the village-level by the
native chiefs and their subordinates. The village chiefs were responsible for
informing their respective village populations of the obligatory nature of the by-laws,
as well as for enforcing them. The DAO staff worked hard to achieve widespread
implementation of the By-laws; they publicize the program via the village chiefs,
made numerous field visits, and even held soil conservation competitions among
smallholders to promote the program (Scherer, 1969). Early DAO documents
indicate that the majority of smallholders readily adopted the SWC measures and recognized their value, and that the widespread adoption of SWC measures reduced
soil erosion and improved overall resource management in the highlands of Kigezi
District (e.g., GOU, 1965e, 1966; Martin, 1945a; Purseglove, 1945, 1946).
The extensive construction of terraces and grass bunds across the landscape also created hillsides covered in a geometric pattern of terraces and vegetation- bordered plots. The patchwork of terraced-plots that characterized the landscape in
Kigezi was considered evidence of a successful soil and water conservation initiative in East Africa, and even cited as "the most spectacular work on soil conservation anywhere in the British empire" (Masefield, 1962: 97; also, McMaster, 1962;
Wrigley, 1959). 128
However, for all intents and purposes, the adoption of the By-laws were
mandatory rather than voluntary and there was a significant degree of coercion exercised to ensure that smallholders abided by the SWC regulations. For example, if
smallholders failed to "voluntarily" adopt the measures, the By-laws were enforced
using coercive and/or punitive measures, i.e., smallholders were fined for the first
infraction, and fined and/or imprisoned for subsequent violations (DAO, 1992,
1993). 5 While smallholders certainly recognized some positive aspects of the By- laws, they had no alternative but to adopt all of By-laws or face punishment. A number of smallholders remembered with resentment the fines imposed they paid for
failing to employ the By-laws during the colonial period; the "top-down" management approach irritated many smallholders who did not agree with the need for such labor- intensive regulations, or the coercive methods employed to enforce them.
Smallholder Soil Knowledge and Land Management
In contrast to the colonial and prevailing post-colonial concerns over land degradation in the southwestern highlands of Kigezi, historical accounts provide some evidence that smallholders in the region employed notable soil and land management practices. Early travelers in the highland region described both intensive agricultural production practices and a variety of soil conservation practices. An account from an expedition in 1919—well before the colonial land management interventions- characterized the highland region as follows:
5 The specific fines and punishments are outlined in the By-laws, which are provided in Appendices D and E. 129
These people of Kigezi are mountaineers and find the steep hillslopes no difficulty; their fields extend up the slopes of the mountains and are marked off from each other by ridges where the weeds and stones are gathered together. After a few seasons, the fields become regular plateaux, for the rains wash the earth from the higher ground against these ridges and formed terraces raised above the lower fields. [Gazing across the valley], the fields looked as though they were laid out in terraces and fenced. (Roscoe, 1922: 101)
Other accounts also suggest that smallholders generally avoided steep-sloping areas, maintained near-continuous vegetation cover on plots, practiced minimum tillage and broadcast planting, often inter-cropped and rotated crops, constructed trash-lines and modified terraces to control for erosion, and fallowed land to manage soil fertility
(e.g., Tukahirwa, 1995; Rugyema, 1974; Reij, 1991). Even the District Agricultural
Officer acknowledged the abilities of smallholders when he stated "the African is a good judge of his own soil" (Purseglove, 1947).
Colonial Management Approaches
Many of the traditional land management practices (e.g., modified terraces, plant-trash lines) were eventually included in the Soil and Water Conservation By- laws. It is unclear, however, whether the DAO included them by design or default
(i.e., they were similar to standard soil conservation practices recommended by soil conservation experts). In either case, the spectacular and widespread success of the
SWC program was probably directly related to the fact that many of the introduced technologies were similar to the ones traditionally employed by smallholders in the highlands. Ironically, the traditional land management practices that were employed voluntarily became compulsory with the implementation of the By-laws, and any non- compliance was met by the DAO and local chiefs with punitive action. 130
In addition to the SWC By-laws, the DAO imposed various agricultural By- laws on smallholders in the highlands. For example, the DAO introduced by-laws to control for cattle grazing, grass burning, tree planting and swamp reclamation, as well as to establish special household "famine granaries" to guard against food shortages.
As with the SWC By-laws, the agricultural regulations were compulsory and smallholders who disregarded them were punished.
The necessity for the compulsory land and agricultural management By-laws was primarily based on field observations and projected land use trends rather than empirical evidence. In fact, laboratory tests conducted under the direction of the
DAO actually indicated that some soils in the area were quite fertile and resistant to
6 erosion (Martin, 1945a, 1945b). Nonetheless, the DAO obviously believed that
strict agricultural and soil and water conservation By-laws were necessary to reduce
soil erosion, mitigate land use pressures and safeguard food production in the highlands (e.g., Erlich, 1965; Martin, 1945a; Purseglove, 1945, 1946, 1947). While population growth and other influences undoubtedly induced smallholders to alter some of the "traditional" land management practices (e.g., reduce the length of fallow periods) described by travelers, the evidence of potentially effective land management practices in the highlands makes the DAO's management approach appear unduly severe and exacting, and not entirely merited.
6 Evidence for soil and land degradation, and the results of various soil tests are further examined in Chapter Eight. 131
Ultimately, the rigid and coercive nature of the agricultural and soil and water
conservation By-laws suggests that the DAO did not fully appreciate the value of the
"native" knowledge or practices, or at best assumed that smallholders did not have the
means or the ability to adapt to the ever-increasing land use pressures without outside
assistance and strict land management interventions. Furthermore, the local
populations were excluded from participating both in the development of the original
land management regulations and the subsequent revisions of the SWC By-laws.
There is also little evidence that the DAO attempted to engender smallholders or local communities with any control over, or sense of "ownership", of the various land
management and agricultural By-laws.
Presently in Kabale and Kisoro Districts, the agricultural and soil and water
7 conservation and By-laws remain official, authorized land management regulations.
The district agricultural offices in Kabale and Kisoro have periodically revised the various regulations, although the current generation of By-laws are facsimiles of the
originals; the By-laws are also still managed in a similar top-down manner.
However, contrary to the earlier colonial "success" story, the various By-laws are now infrequently promoted and enforced by the DAO's staff, and only selectively implemented by the smallholders. The incomplete adoption of the By-laws has not provoked investigations into the reasons why the DAO no longer widely enforces them or smallholders selectively employ them, or the current nature of soil
7 The contemporary By-laws for Kabale and Kisoro Districts are provided in Appendices D and E. 132
degradation and smallholder soil resource management in the highlands.
Unfortunately, the DAO staff and outside experts continue to assert that smallholders
do not understand the severity of the environmental problems in the highlands, and
they are generally indolent, ignorant and incapable of managing their resource base. 8
Summary
The observable pattern of regional inequalities in Uganda reflect the legacy of
historical influences as well as the more contemporary situation. In the highlands of
Kigezi District, agricultural and land management systems have changed in response to a variety of internal and external factors. One of the major historical and external influences in Kigezi District was the British colonial administration. While the Kigezi
District agriculture office made considerable investments into the soil and water conservation program, the financial investments in infrastructure and agricultural development were considerably less than those of the central districts. Despite the small~and temporary-successes such as the vegetable cooperatives in Kabale, the development policies of the colonial state in Uganda were characterized by regional preferences (Ede, 1981). The efforts by the colonial state to "underdevelop" the peripheral regions to maintain labor supplies for estate-agriculture, the repressive tax systems, the agricultural and land management By-laws, and the physical constraints
imposed by distance and a highland environment, have all contributed to the
The contemporary perspectives of the DAO and smallholders concerning the SWC By-laws are further explored in Chapters Seven and Eight. 133 development of the semi-subsistence, agricultural-based economy that currently predominants in Kabale and Kisoro Districts.
Population growth resulting from both historical in-immigration and more
recent natural increases is also considered to be one of the primary agents of change in Kigezi District. Correspondingly, concerns over population growth and land use pressures provided the impetus for many of the colonial land management interventions in the region. Since the early colonial period, the rate of population growth in the highland region was mirrored in the mounting concern over land use pressures and environmental degradation within the district administration. In response to the perceptions of population-induced land pressures, risks to local food production and expanding land degradation, the Kigezi District administration implemented a number of agricultural policies, population control and land management plans. The District Agricultural Office's various agriculture and land management regulations, particularly the Soil and Water Conservation (SWC) By- laws, lead to the dramatic alteration of land use and resource management in the
highlands. While it is difficult to determine to what degree the various by-laws have helped to improve resource management, they have definitely influenced-even dictated-many of the smallholders' agricultural and land management practices in the highland region.
The agricultural and land management regulations imposed on Kigezi District were developed within a "top-down" management framework, where the problems were identified and the solutions were proposed by the DAO staff and outside experts, 134 and then imposed upon the smallholders. The SWC By-laws proscribed traditional agricultural practices (e.g., grass burning), and prescribed technologies that often required the modification or replacement of traditional agricultural practices, and were highly labor intensive. Most importantly, smallholders were not provided the opportunity to participate in the development or subsequent revisions of the very agricultural and land management program that they were expected to implement and manage. Unfortunately, the DAO appeared to make little effort to understand smallholders' soil knowledge, soil and land management practices, or support their experimental and adaptive cabilities.
In the following chapter, Chapter Six, the soil classification systems of smallholders at the three study sites are characterized and compared to the conventional USDA and FAO systems. Additionally, smallholder knowledge and perceptions of soil catenas, soil erosion and soil fertility, are presented. CHAPTER 6 SMALLHOLDER SOIL KNOWLEDGE
Scientific Soil Classification Systems
There are a great variety of soil classification systems employed around the
world that separate soils into categories that reflect the concerns and biases of each
particular system. None of the ten or more widely recognized conventional, scientific
soil classification systems is universally accepted (Buohl et al., 1989; FitzPatrick,
1983). For example, the USDA's soil taxonomy has been extensively researched and
utilized in the U.S. (e.g., Soil Survey Staff, 1975, 1993). It has not, however, been
widely applied or tested in tropics, and it is strongly biased in favor of chemical and
physical analysis, and emphasizes pedogenesis and soil morphogenesis over soil
management or land suitability considerations. Another example, the FAO soil
classification system, was developed to describe soils of the world and create a World
Soil Map (e.g., Dudal, 1968; FAO, 1993a). The FAO system generalizes about soil patterns over large regions, and as a result the map units describe the dominant soil
for a particular area and micro-level soil and landscape variations are overlooked
(FitzPatrick, 1983). The USDA and FAO systems are capable of grouping soils according to selective scientific criteria and develop "universal" soil models, but they do not specifically address the soil management or land suitability concerns of many agriculturalists and land managers.
135 136
On the other hand, smallholders in many areas of the world have developed
soil and land classification systems that both offer insights into their agricultural and
land management systems, and provide a useful vehicle for discussing related issues
with them (Pawluk et al., 1992; Tabor et al., 1990). Smallholders are often most
interested in the topsoil, or the 20 - 30 centimeters of a soil that is important for crop
production, and they often characterize their soils according to the agricultural
potential of the topsoil (e.g., Wilkens, 1987; Brouwers, 1993). Consequently, an
emphasis on scientific classification systems at the expense of the local systems can
bias investigations of traditional agricultural and land management practices, and soil
characteristics that are important to smallholders may be disregarded.
Smallholder Soil Classification Systems
Background
In Kabale and Kisoro Districts, only a limited number of conventional soil
classification surveys have been undertaken since the colonial period. The most
recent data available is quite limited in scope, however, as the soil surveys (based on the USDA and FAO systems) were only conducted at the Kachwekano (e.g.,
Tukahirwa, 1995) and Kalengyere (Ssendiwanyo, 1992) agriculture research stations in Kabale District. And while the soils on the two research stations are characteristic of soils in the highlands, they are not specifically representative of the soils that smallholders cultivate-primarily because they have been subject to different types of management. In recent decades, there is no indication that soil surveys or soil research have been conducted under smallholder agricultural conditions in the rural 137
areas. This lack of research is quite unexpected given the widespread claims of soil
and land degradation in the highlands, and overall indicative of the limited resources
that have been invested in land management in the region.
Results
An investigation of smallholders' knowledge and management of soils was
conducted at three sites in the highlands of Kabale and Kisoro Districts, in 1993 and
early 1994. Across the three study sites, smallholders revealed a wealth of
knowledge about local soils and soil characteristics. They were able systematically to
describe a wide range of soil types, as well as the agronomic potential, general
fertility, and erodibility attributes for each of the soils. The smallholder soil
classification systems for Kicumbi, Nyarurambi and Muramba parishes are presented
in Tables 6.1, 6.2 and 6.3, respectively. The soil types in the tables are presented in
a rough "catena" profile, whereby the position of the soil type in the table also
represents its relative location on a "typical" slope. (A discussion of soil catenas is
provided below).
Smallholders throughout the three sites classified soils according to a variety of
locally important criteria. These criteria included soil color, "stoniness", moisture-
holding capacity, susceptibility to drought and erosion, "till-ability" and agricultural
1 potential. A soil's agricultural potential (or land suitability) was one of the most
1 In Kicumbi parish, smallholders also included "amabare", or rocks and rock outcrops as a soil/land type. While amabare was not a soil-type per se, smallholders referred to it because it had a particular land use, i.e., it was generally suitable for tree planting.
. 1 1 1 1 1
e 2 s & 3 >5 l £ 1 g.<2 1 1 1 8 u 3 * a. sa 00 .ft S 3 > > 5 00 1 S 2
ft* ft. | c - 2 1 5 D a u . — Q H B. GO U- < > g 4 X3 o "2 2 1 i — .9 1 "I -a I I I 1 S 60 =s 2 C •c 1 5 4 o 8 v> i 9 >^ ll o a Ik. 0 B 1,* 1 i 8 1 9 § | £ ill 8 2 2 * Z a J 2 8 X o •c 3.1 8 5 2 § 75 5 2 O .2" "3 S g a 00 &L •§ ll! | « a 0 1 1 v> « P o a ?1 8 S § J2 . w •» g O u is >> A & a so a . § 8 * g.| §f o a 8 11 c r a 01 t |C 6 11 0 '35 oo I s a.S 1 - J .2 & >> 60 a ° 5 J> U It 60 <2 & i *ll- 1* I o O 2 JO a O w 3 s J1 3 3 1 a 8 8 w» *~ Q ! i §5 60 ° !C « 8 ii a b •a1 T3£ > S 2 VI I 60 * J 115 a 2 S a S o a _ =111« 2 1 > i OS 03 Q 2 U o 05 D.73 S J on
4 o < G< S I 8 2 o 7> 1 2 I S.2 o a VI I 60 - o 1 o ft. 00 00 T "f 2 a 5^ T t! h .2 T •a2 « >^ P u V if u 2~ 60 ft- a & 0,3 2 a t> y5 2 S I 8 3 s3 1! 1 u- > M 5^
go 3 o .2, -3 M S o 8 § 60 2 2 0 3 I I 2{» 00 3 I i i II i i 1 1 2 o & o 3 140
ter ter. | J "a.
bananas
/wal /wal
— <% ses ses U 3 3 wel O O wel x: 43 so 00 • 1 uce uce ity. ity. rin nn
prod 3 fertil prod fertil 1/3 ? 1
for for crops crops
Excellent Excellent i 1 cu All All £ Q S i 5 V § 09 4 0 > 3, so 2 60 a-i CJ I60 60 60 "i .£ I M ii 60 3 « — 60 C 0 j 1 I c b 11a .2 5 e -3 J £ J) o g 1 8 i j" JZ £ . el >> > o Q - 3 o US SL-g J2 S 8. S 2 > Q -a o* "o o J c/5 X 3 o ° I § u i 0 so 8 s S £ ^ J 1 3 I P s 1 * 1 U" » .2 •-a o 3 XI O 1 a 2 <£ 3- 1 ° >> |l 5 EC * -a 3 > u 1 | | o '3 11 6 I CO 03iiso CQ > a*
(*?
u o
1 .2 ii 13 8 |« a. i XI XI s i 3 on t/5
8 ~3 3i
2 II 5 5 5 | 2 § 3 C/J 1 s CQ 141
important criteria that smallholders considered, and as such it underscored the
utilitarian or functional dimensions of the local classification system. In other words,
smallholders were primarily interested in a soil's ability to produce crops, and they
managed them accordingly-preferring to cultivate certain crops on specific soil
2 types. Smallholders also described a number of subsoils and other geologic features
(e.g., exposed bedrock) in the region in terms of functional characteristics. For
example, most smallholders in the region described bedrock and rock outcrops in
terms of their use as local building or construction materials, while smallholders in
Muramba described subsoils in terms of their use as wall plaster.
Discussion
The classification schemes delineated in each table are based on the synthesis
of soil descriptions that were provided in multiple focus group sessions with
smallholders. There was often individual and group variation in the soil descriptions, but only the most common and widely accepted characteristics were incorporated into the soil classification schemes. While the descriptions of soils varied to some extent within parishes, there was limited dispute about the general classification of soil types. Some smallholders did point out, however, that there were often variations within a general soil type, i.e., there might be two or three sub-types of Orushenyi in
Kicumbi Parish. The variations of sub-types might differ according to certain
2 This "traditional" method of differentially managing soil types according to their agricultural potential, or "farming by soil", is now promoted in conventional, western agricultural systems in the U.S. -albeit with the assistance of tractors and computer technology (e.g., Roberts, 1993). 142
characteristics, e.g., varying particle sizes or levels of productivity, but they were all
3 considered an Orushenyi soil type.
Soil Catenas
Background
A soil catena is generally described as a sequence of soils (or toposequence) of
similar age, parent material and genetic features that have different characteristics due
to variations in landform relief and related drainage patterns. Soil scientists have
been aware of the correlation between soil types and physical relief in Sub-Saharan
Africa since Milne (1935, 1936) first described the relationship in the 1930' s. Milne
formulated the catena concept in recognition of the clear association between slope
forms and specific soil sequences, and numerous researchers have applied and verified
the concept in Africa and throughout the tropics (Moss, 1968: 41; Mausbasch and
Wilding, 1991; Semmel, 1994; Steila and Pond, 1989).
Results
Smallholders in Kabale and Kisoro Districts also recognized soil associations
within the highland environment, and were able to discuss the relationship between
soil types and relief. In all three parishes, smallholders provided general descriptions
of the associations between relief and soil types they observed in the field.
During group mapping exercises, smallholders were also consistently able to recreate
and map soil catenas for various landscapes in Kicumbi and Nyarurambi Parishes. A
3 Local soil names can often vary in spelling among smallholders, and parishes, e.g., ibumba vs. eibwnba. 143 diagram of a hillside profile of a typical catena in Kicumbi parish, redrawn from smallholder sketches, is presented Figure 6.1. Given the close similarity between the diagrams smallholders produced for Kicumbi and Nyarurambi parishes, a catena diagram was not reproduced for Nyarurambi. Additionally, given the minimal relief
and indistinct soil-relief associations on the volcanic plain, a catena diagram is not provided for Muramba. 4
Discussion
Smallholders were in general agreement as to the occurrence of broad soil - relief associations in their respective parishes. They also recognized variations on the general soil catena in each parish, noting that the existence and location of a specific soil depended partly on the landscape in question. There were aspects of the diagrams, however, that were consistently similar. For example, smallholders in
Kicumbi and Nyarurambi agreed upon the general landscape locations of many soil types. They noted that orucucu was found along ridges and hilltops, enombe was located along footslope areas and orufunjo occurred only in the valley bottoms.
There was also a consensus, however, that some soil types such as Orushenyi could be found in a variety of midslope and lowerslope areas. Additionally, smallholders noted that soils such as irikwiragura had relative locations but were found with specific landscape features. That is, Irikwiragura was generally located in depositional areas such as the lower reaches of large ravines or step-valley bottom
4 Smallholders in Muramba, however, were still able to discuss the concept and describe other catenas and soil associations in the general area. 144
o
,3 145 areas; these landform features were found throughout the parish at various altitudes, and correspondingly so was the Irikwiragura soil type.
Comparison of Local and Conventional Soil Classification Systems
Background
The majority of the investigations into smallholder knowledge of soils have not been supplemented with conventional, scientific soil surveys (McCall, 1996). In this study, in order to compare the smallholder soil classification systems with western, scientific classification systems, soil surveys were also conducted according to the
USDA and FAO systems. The USDA and FAO soil classification systems were utilized due to their widespread acceptance in the development and scientific communities. The USDA system, and to a lesser extent the FAO system, are primarily based on the pedogenic characteristics of soils, and specific measurements of soil chemical (e.g., Cation Exchange Capacity, or CEC) and physical properties
5 (aggregation, particle size) (e.g., Wilding et al., 1983a, 1983b). While these scientific systems do not specifically address a soil's agricultural potential or land suitability~the most important criteria among smallholders-they do provide additional data and the means to communicate with agricultural scientists on the nature of soils in the highlands.
Additionally, the local soil types were categorized according to the Fertility
Capability Classification system (FCC). This is a technical system that groups soils
5 Some of the basic physical and chemical properties of these soils as determined through laboratory analysis are provided in Appendix G. 146
according to the kinds of problems they present for agronomic management (Sanchez
et al. 1982). The FCC system emphasizes the physical and chemical properties of
soils particularly as they relate to agricultural production. While this system has been
in development for over a decade, it has not experienced widespread application.
Nonetheless, as with the smallholder system, it places particular emphasis on the
agronomic management aspects of a soil and was utilized primarily because of the applied or functional dimension of the system.
Results
For the purposes of comparing soil classification systems, pits were dug in
soils that smallholders classified according to the local system. These
"representative" soils were then characterized and tested according to the prescribed
USDA and FAO soil survey methods (e.g., Dudal, 1968; FAO, 1977; Soil Survey
Staff, 1993, 1994). The USDA and FAO classification interpretations of the local soils, as well as a brief description of the soils, are provided in Table 6.4.
Specifically, the table presents all of the soil types in each parish as described by smallholder classification system, and provides the corresponding USDA and FAO classification for each of the soil types. To conduct the FCC study, three samples of topsoil (< 30 cm.) were collected at each of the soil pits dug for the USDA and FAO soil survey. The chemical and physical properties of the topsoil samples were analyzed and each sample was grouped according to the Fertility Capability 147
drained, drained,
Soil Description
reddish-brown. black.
drained, poorly, drained, drained, drained, drained, shallow, brown. brown. brown. brown. brown. deep, deep, deep,
Imperfectly Imperfectly
drained, shallow, shallow Well Well Well Well Well dark dark black. dark dark dark dark Very very very deep, deep, very very
Classification FAO
Leptosol Leptosol Ferralsol Ferralsol Histosol Planosol Ferrasol Histosol Haplic Humic c Terric Eutric Lithic Lithic Fibric 5 » a o its u
Tropopsaniment Classification o USDA Troposaprist Tropofibrist c/5 Petroferric Hapludox Hapludox Eutrudox Eutrudox Eutrudox Humic Humic Typic Terric Typic Lithic Terric
c u Type
Local
Soil Eryakatuku Omushenyi Irikwiragura Orushenyi Orufunjo o Enombe Eibumba Orugugo S
A B B
Nyamabare Nyamabare Nyakatoma Nyakatete Nyakatete Rushongati Omurujanja Nyakatete
Kicumbi Kicumbi Kicumbi Kicumbi Kicumbi Kicumbi Kicumbi Kicumbi District: Kabale Kabale Kabale Kabale Kabale Kabale Kabale Kabale Village: Parish:
Number Profile in r~ 00 148
drained, shallow, shallow,
deep, deep,
grayish-brown. Soil Description gray. reddish-brown. reddish-brown. reddish-brown.
brown. drained, drained, drained, drained, drained. drained, drained,
brown. dark
Imperfectly moderately moderately moderately moderately
shallow, Well Well Well Well Well Well Well black. dark dark strong dark dark dark deep, deep, deep, very
Classification FAO
Arenosol Andosol Umbric Histosol Acrisol Acrisol Acrisol Acrisol Acrisol Haplic Haplic Haplic Humic Haplic Humic Terric Lithic
USDA Classification
Haplohumult Troposaprist Kandihumult
Hapludult Hapludult Hapludult Hapludult Eutrandept Andeptic Umbric Humic Arenic Typic Typic Lithic
Type
Local
Soil Irikwiragura Omusenyi
Orucucu Orufunjo Orucucu Enombe Orugugo
Insibo
a Kanyamatembe Kanyamatembe G> Nyarurambi Nyarurambi Nyarurambi Nyarurambi Nyaruranbi Nyarurambi Nyarurambi Rushongati Rushambya
Kicumbi District: Rutooga Kabale Kabale Kabale Kabale Kabale Kabale Kabale Kabale Village: Katasa Katase Parish: : ce
Number Profile ON o }2
J 149
deep, deep, deep, deep, deep, brown. brown.
Soil Description
drained,
drained, drained, drained, drained, drained, drained,
brown. dark dark
moderately moderately moderately moderately moderately
shallow, Poorly Well Well Well Well Well Well black. black. black. black. dark deep, very very
FAO Classification
Arenosol Andosol Andosol Andosol Andosol Umbric Andosol Planosol Mollic Haplic Mollic Mollic Eutric Vitric
USDA Classification
Hapludult Tropaquept Eutrandept Eutrandept Eutrandept Eutrandept Vitrandept Umbric Umbric Arenic Lithic Aerie Udic Udic
Type Local
Umukara Soil Orushenyi Onikoro Urukoro Urusenyi
Ibumba Insibo
Kanyamatembe Kanyamatembe
Nyarurambi Nyarurambi Nyarurambi
Muramba Muramba Muramba Muramba
District: Rutooga Kabale Kabale Kabale Bulere Kisoro Bulere Kisoro Bulere Kisoro Bulere Kisoro Village: Parish:
Number Profile r- 00 0\ o es 150
Table 6.5: Cross-Site Comparison of Soil Classification Systems
Parish: FCC Classification USDA FAO
Local Soil Type (Samples 1, 2 and 3) Classification Classification
1. K: Orucucu Ska (1,2,3) Typic Kandihumult Haplic Acrisol
N: Orucucu Lcka (1), Ska (2,3) Humic Hapludult Humic Acrisol M: -absent-
2. K: Eryakatuku Lkh (1,2,3) Typic Eutrodox Haplic Ferrasol N: M:
3. K: Orugugo Lkh (1,2,3) Lithic Hapludox Lithic Leptosol
N: Orugugo Sika (1), Sikh (2,3) Lithic Hapludult Lithic Haplic Acrisol M:
4. K: Orushenyi L"kh (1,2,3) Petroferric Hapludox Lithic Leptosol
N: Orushenyi Lkh (1), Lh (2,3) Arenic Hapludult Haplic Arenosol M: Urusenyi Sh Umbric Vitrandept Vitric Andosol
5. K: Enombe Lkh (1,2,3) Humic Eutrodox Humic Ferrasol N: Enombe Ckia (1,2,3) Typic Hapludult Haplic Acrisol M:
6. K: Irikwiragura Lh (1,2,3) Humic Eutrodox Humic Ferrasol N: Irikwiragura Lkh (1,2,3) Andeptic Haplohumult Humic Acrisol M:
7. K: Omushenyi Sh (1,2), Lkh (3) Typic Tropopsamment Eutric Planosol N: Omushenyi L'c'kh (1,2,3) Arenic Hapludult Haplic Arenosol M:
8. K: Eibumba Lckh, Ch, Ckh Terric Troposaprist Terric Histosol
N: Eibumba Lka (1), Lkh (2,3) Aerie Tropaquept Eutric Planosol M:
9. K: Orufunjo Oghk (1,2,3) Terric Tropofibrist Fibric Histosol N: Orufunjo Lcgka (1,2,3) Troposaprist Terric Histosol M:
10. K: N: Orukoro Sh (1,2,3) Udic Eutrandept Mollic Andosol
\ M T T 1 M: Urukoro Sh (1,2,3) Udic Eutrandept Mollic Andosol
11. K:
N: Insibo Skh (1,2), S.h (3) Umbric Eutrandept Umbric Andosol M: Insibo S"h (1), L'kh (2,3) Lithic Eutrandept Mollic Andosol
12. K: N:
M: Umukara Lh (1), Sh (2,3) Umbric Eutrandept Umbric Andosol
Note: Parishes: K - Kicumbi, N - Nyarurambi, K - Kisoro. 151
6 Classification system (e.g., Orucucu = Ska). The results of the FCC categorization
are presented in Table 6.5.
Furthermore, Table 6.5 provides the opportunity to compare the various local
and scientific classification systems within and across the three study sites. That is,
soils in different parishes with similar smallholder classifications were grouped
together, and their respective conventional classifications were also provided for
comparison purposes.
Discussion
A comparison of the three local soil classification system reveals that
smallholders across the study sites generally evaluate similar soil characteristics:
color, texture, water-holding capacity and agriculture potential. Although laboratory comparisons were not conducted, a comparison of similarly named soils in each of the
parishes also indicates a strong correlation between physical properties of the soils, particularly at a basic descriptive level. For example, there were discernible
similarities in texture, particle size and permeability among the sandy soils (i.e.,
Omushenyi) across the study sites. Additionally, where they occurred in the parishes, there was an obvious correspondence between wetland soils (i.e., Orufunjo), loam soils (i.e., Irikwiragura) and volcanic soils (i.e., Insibo, Orukoro). Furthermore, smallholders described similar uses or management preferences for soils across parishes. While the local soil classification systems were not necessarily
6 The FCC system is generally accessible to non-soil scientists and an outline of the FCC nomenclature is provided in Appendix F. Additionally, the data indicate that the FCC classification can vary among soil samples of the same general soil type. 152 interchangeable between parishes, they certainly provided the means for smallholders
in the different parishes to communicate among one another with a general degree of mutual comprehension, e.g., smallholders everywhere understood at least at a basic level that Orufunjo was a wetland soil.
The data provided in Tables 6.4 and 6.5 indicate that there is not, however, any direct correspondence between smallholder and either the USDA, FAO or FCC scientific classification systems across the three study sites. Depending on the
specific local soil type, the scientific systems exhibit varying degrees of correlation with the local system, and the degree of correlation generally increases with increasing levels of generalization. For example, with the three volcanic soils across
Nyarurambi and Muramba parishes (Table 6.5: No.'s 9, 10, and 11), there is significant correspondence between the smallholder systems and the higher levels of
classification in the USDA (i.e. , soil order level) and FAO systems. That is, the volcanic soils in the smallholder system are also classified as volcanic soils in the
USDA (i.e., Andisol) and the FAO (i.e., Andosol) systems.
A similar correlation is found with the hydromorphic, wetland soils at Kicumbi
and Nyarurambi parishes (Table 6.5: No. 8), where the local soils are classified the similarly at the higher levels of generalization in the USDA and FAO systems. That
is, the locally defined wetland soils are also classified as wetland soils in the USDA
(i.e., histosol) and FAO (i.e., histosol) systems. There are also multiple examples,
however, of little or no correlation between the various systems, regardless of the 153
in Table 6.5, there is much less correspondence between systems at the more specific
levels of the classification systems.
Unlike the USDA and FAO systems, the FCC system provides within its
nomenclature an indication of the agronomic potential and constraints characteristic of
each soil. The use of the FCC system revealed some similar classifications across
sites for the volcanic soils (Table 6.5: No. 10) and other soils like Orucucu (e.g., No.
1, ), although there were also some mixed results (e.g., No. 5) as well as significant
differences within and between sites (e.g., No.'s 7 and 8). The variations in the FCC
classification of samples from the same local soil type potentially reveal the micro-
level variations in soils, as well as some of the shortcomings of the FCC system.
While both the FCC and smallholder systems primarily evaluated the top-soil to
classify a soil, there was not a high degree of correlation between the systems. This
was possibly due to the fact that the two systems were measuring different criteria,
and/or measuring simialar criteria to different degrees of specificity (e.g., the FCC
system relied on laboratory-based soil chemical and physical analysis, while the
smallholder system was based on observable characteristics and crop responses.)
The comparison of indigenous and scientific soil classification systems can be
fraught with problems. The potential difficulties are related to the ambiguous, elastic
nature of smallholder soil classification systems (a reflection of the variability of soil knowledge among villagers), the different assumptions on which the systems are based (a risk of comparing apples and oranges) and the incorrect assumptions and bias introduced by outside researchers conducting the comparisons (Niemiejer, 1995; 154
Sikana, 1993a, 1993b). Admittedly, at each of the three study sites, soil knowledge varied among smallholders and the local soil classification systems were by no means absolute. The local systems were collectively delineated by smallholders, however, and they do provide a general framework for understanding smallholders' perceptions of their soil environments. And while the local and scientific systems are not necessarily based on similar assumptions or criteria for evaluation, they do provide unique and complementary insights into the soils and soil management practices common to the highlands. Additionally, the conventional soil surveys contribute data to an impoverished data base on soils in the highlands.
Smallholder Perceptions of the Fertility and Erodibility of Local Soils
Smallholders throughout the highlands demonstrated a basic understanding of general fertility and erodibility status of local soils, and some of the ways they are interrelated. For example, most smallholders recognized that soil erosion and the subsequent loss of topsoil-through soil loss via water transport, or downslope soil movement via hoe-cultivation~negatively affected soil fertility. The majority of smallholders were not able to evaluate the fertility or erodibility characteristics of soils in a precise way. However, smallholders were able to rank a given soil's
fertility (i.e., productivity) and erodibility (i.e., susceptibility to erosion) characteristics relative to the other local soils. These general rankings are provided in Tables 6.6 and 6.7. 155
Table 6.6. General Soil Fertility Ranking.
1. Kicumbi 2. Nyarurambi 3. Muramba
Most Fertile Irikwiragura Irikwiragura Umukara
Orushenyi Orushenyi Urukoro
Eibumba Enombe Urushenyi
Enombe Eibumba Insibo
Orufunjo Insibo Gahuma
Orucucu Umukara
Omushenyi Orufunjo
Least Fertile Omushenyi
Note: This is only a listing of the most commonly ranked soils; soils such as Eryakatuku which are rarely cultivated were not included.
Table 6.7. General Soil Erodibility Ranking.
1. Kicumbi 2. Nyarurambi 3. Muramba
Most Erodible Irikwiragura Enombe Umukara
Orucucu Orucucu Urukoro
Orugugo Irikwiragura Urusenyi
Enombe Orushenyi Insibo
Eryakatuku Insibo
Least Erodible Orushenyi Umukara
Note: Smallholders pointed out that the Eibumba and Orufunjo soil types were located on the footslopes and valley bottoms, respectively, and thus didn't experience significance erosion.
These are flexible rankings of soil characteristics, however, and smallholders articulated various opinions as to the order of the rankings, particularly concerning 156 the middle range of soils. There was often agreement, however, as to the extremes on the spectrum of fertility, or most and least fertile soils. For example, in both
Kicumbi and Nyarurambi parishes, many smallholders agreed that Irikwiragura was the most fertile soil, while they generally considered Orucucu and Omushenyi to be
7 the least fertile soils. Smallholders also pointed out that the fertility and erodibility
status of a soil often depended in part on how long it had been under cultivation, how
well it was protected by bunds and other soil conservation measures, and that the productivity of soils under continuous cultivation varied depending on the specific
soil. For example, under continuous cultivation, Irikwiragura maintained its productivity very well, while enombe experienced a marked decline after three to five years. While some smallholders indicated a need to manage the soils differently, i.e., put the Enombe soil into fallow for longer periods, land use pressures generally ensure that plots are not kept out of cultivation for extended periods of time.
A number of smallholders, particularly the elderly, also indicated that soil fertility could be evaluated given the presence of certain weed species. For example, there was general agreement among smallholders that the presence of Digitaria abyssinnica indicated poor soil fertility. The younger smallholders, however, were generally not as competent at identifying weeds, what their presence implied or their
7 Many smallholders also noted that orucucu was at one time more fertile than at present, but that it had been over-cultivated and lost its fertility in most areas. 157
8 other uses. It would appear that this local knowledge of weeds and other aspects of
9 the local agricultural systems are rapidly deteriorating.
Smallholders in all three parishes exhibited an understanding of the basic
fertility and credibility characteristics, but they also presented some uncommon perspectives-- some supernatural, others scientifically astute. For example, smallholders generally acknowledged that continuous cultivation contributed to declining soil fertility, but a number of smallholders in each parish also ascribed poor crop production to the existence of evil charms or "hexes" that an enemy had placed
10 in their plot. Smallholders in Kicumbi and Nyarurambi parishes also acknowledged that a considerable amount of soil materials had been carried into the wetlands over the years and this was generally viewed as a moderately detrimental trend. Some smallholders pointed out, however, that the upland soils appeared to enhance the productivity of the wetland soils, particularly in the footslope areas.
While this phenomenon has not been examined empirically, it is feasible that the eroded soil materials may help to increase or buffer the pH of the wetland soils—
8 Elderly smallholders were able to identify over 50 weed species at each of the study sites. Not only were the "weeds" used to indicate the fertility status of a soil, but they were also used as food during famine periods, livestock feed, human and livestock medicines, and handicraft materials.
9 Many youth are no longer interested in pursuing a rural, agricultural livelihood. Given accessible transport and the lure of urban life, many youth prefer to migrate to regional towns—even if there are few employment opportunities available.
10 Bentley (1989) has also pointed out smallholders' supernatural or alternative explanations for non-observable phenomena such as diseases or agricultural pests. Smallholders in southwest Uganda appear to explain soil micro-nutrient deficiencies in a similar manner. 158 especially after prolonged cultivation." (Additionally, alternative smallholder soil perceptions and land management practices are discussed in Chapter Seven.)
The smallholder perceptions of fertility loss and erosion indicate that smallholders recognize these processes, but often conceptualize their causes and impacts differently from the conventional sciences. This does not imply that they are
always ignorant of a process, or the consequences of their actions. Rather, it indicates that they have varied ways of looking at their environment, some of which are contradictory to conventional sciences, others which are complementary and others which can expand upon them.
Summary
This chapter has described a wealth of knowledge about soils and soil processes among smallholders in the highlands of southwest Uganda. The local soil classification systems are based on soil characteristics that are important to the
smallholders. Given the agricultural basis of the rural economy, it is was not surprising to learn that smallholders classified soils primarily according to their
agronomic potential. Smallholders at all three study sites revealed a well-developed soil classification system, and a basic understanding of some of the soil processes
relating to soil erosion and soil fertility. While smallholder soil knowledge was quite
comprehensive, it was not uniformly represented among all smallholders, nor was it equally complex or precise for all soil characteristics. For example, there was variation in the depth of soil knowledge among smallholders, and in each parish
11 The pH of the local soil types are provided in Appendix G. 159
smallholders acknowledged these differences when they identified some individuals
(e.g., "soil experts") as being more knowledgeable than others. Additionally, while
smallholders often recognized that soil fertility might be declining in agricultural
plots, they ascribed the reasons to a variety of causes ranging from the agronomic
(e.g., overcultivation) to the cosmological or supernatural (e.g., witchcraft).
Some commonalities were found between the local soil classification systems across the three study sites. The functional and descriptive nature of the smallholder classification systems provided the basis for the general correspondence, and the
means for smallholders throughout the region to communicate with a basic level of mutual understanding. There were some commonalities identified between the
smallholder and scientific soil classification systems, but multiple differences between the local and scientific systems were also evident--and there were differences among the scientific systems as well. The incongruities between systems were basically a product of the different criteria the various systems used to differentiate between soils, and the different levels of generalization within the classification systems that
was being compared. These differences indicate that there is no one system for
classifying soils or evaluating soil erosion and soil fertility status, just as there is no single way of conceptualizing human-environment relationships. Local ("indigenous") and scientific knowledge represent different perspectives on a continuum of knowledge and experience, and are not static and mutually exclusive ways of viewing the world. Soil classification systems, local or scientific, represent attempts to simplify the complexity and continuum of the real world into more understandable, 160 discreet classes (Tabor et al., 1990). While every classification system creates artificial distinctions among soil types, they also provide a common means for discussing soils among the users of a specific system.
If regional resource management initiatives in the highlands of southwest
Uganda aim to improve soil and land management, then enhanced communication among smallholders, and between smallholders and agricultural and development professionals will be imperative. The utilization of the local soil classification system as a common language can facilitate communication between the involved parties.
Indeed, it would be easier for a handful of DAO staff and outside experts to learn the smallholder soil classification system and local perspectives on resource degradation
and management, than it would be for the entire population of smallholders in the highland region to learn the USDA, FAO or other systems.
Even if conventional science eventually determined that the local classification systems were ambiguous or completely inaccurate, an awareness of such shortcomings or incongruities might provide insights into the perspectives of smallholders, as well as opportunities to discuss management alternatives from the vantage point of smallholders. Rather than contradictory, the different systems should be considered complementary, and both valued for their abilities to provide different and unique insights into the nature of highland soils, as well as potential management recommendations for them.
In Chapter Seven, which follows, the conventional perceptions of soil and land degradation in the highlands are presented, and then compared to the analogous 161 and unconventional perceptions of smallholders. The various smallholder soil erosion control and soil fertility management practices, including the implementation of the soil and water conservation by-laws and the use of organic and inorganic inputs, are examined. Finally, smallholder-cited agricultural and land management concerns and priorities are examined. CHAPTER 7 SOIL DEGRADATION AND SOIL RESOURCE MANAGEMENT
Introduction
The highlands of Kabale and Kisoro Districts are comprised of a diversity of landscapes including hilly terrain with steep slopes, volcanoes, wetlands, and minimal surface hydrology. The region has experienced considerable anthropogenic modification over the last 50 to 100 years, and is currently characterized by dense populations, widespread deforestation, and extensive cultivation. The agricultural
areas activity is concentrated on the hillslopes, although the majority of the wetland have also been reclaimed for agricultural purposes in the last three to four decades.
The highlands are commonly considered to be a region beset with land use pressures and a diversity of environmental problems. The problems cited include deforestation, wetlands destruction, stream and lake siltation, soil erosion, soil fertility decline and general land degradation; these problems are generally considered to be the result of overpopulation, intensive agricultural production, land shortages and farm
fragmentation (e.g., Bagoora, 1988; GOU, 1994b; Ngabirano, 1993; World Bank,
1993).
162 Soil Degradation '
Soils are an essential resource in productive agricultural systems (e.g., Ewel,
1986; Young, 1989). Although temperature, rainfall and other climatic factors are as
important as soil for agricultural production, soils are the easiest component of the
2 agroecosystem for smallholders to manage (Nicholaides and Moran, 1995). Soils
can be managed to maintain or increase agricultural production, but they also can be
easily mismanaged and degraded to the point where they are no longer productive.
Soil degradation encompasses both soil erosion (e.g., via natural and human-induced
land disturbance) and soil fertility declines (e.g., via overcultivation and the net-
export of micro-nutrients through crop production and harvest).
Soil erosion is a function of rainfall erosivity, soil credibility, topography,
crop management and soil conservation (Hudson, 1976; Tukahirwa, 1995). Rainfall
erosivity is the force that causes soil detachment and particle transport. Erosivity is
related to the kinetic energy of both raindrop impact and water runoff; it can be
determined by measuring the amount and intensity of rainfall (Lai, 1988b). Soil
erodibility is a measure of the resistance of soil to particles attachment. It is related
to soil clay and organic matter content (Lai, 1988b), particle dispersion (Middleton,
1930) and soil aggregation (Bryan, 1968). In terms of topography, slope length and
1 This section on soil degradation is in part a summary of a review undertaken by Bellows (1992).
2 For the purposes of this dissertation, "soil resource management" (SRM) refers to the complex of management practices that smallholders employ to control soil erosion and maintain soil fertility.
163 164 steepness affect the overland flow of water that is responsible for sheet and rill erosion. Particle detachment by raindrop splash and overland flow velocities are greater on steep slopes than shallow slopes (Lindsey and Gumbs, 1982), while longer slopes also allow for the increased velocity of surface run-off (Hudson, 1976). Crop management and mulching also affect soil coverage. Proper plant canopy cover can
impact, protect soils from erosion by absorbing the kinetic energy of the raindrop while mulches can also mitigate erosion by reducing the velocity of the rainfall runoff
(Wischmeier, 1975). Soil conservation measures can both increase soil cover (e.g., agroforestry) and reduce the amount and velocity of overland flow (e.g., hedgerows, check-dams).
Erosion degrades soil resources through the erosion of topsoil and organic matter. On steeply sloping lands with shallow topsoils and poorly developed subsoils,
(Lai, soil losses due to erosion may lead to an irreversible decline in soil productivity
during 1985). Organic matter, due to its low bulk density, is preferentially removed erosion. Short fallows and weed and agricultural residue burning can also deplete soil
nutrients and organic matter (Hudson, 1976; Lai, 1987). The loss of organic matter
results in the chemical, physical and biological deterioration of the soil. More
specifically, the loss of organic matter causes decreased aggregate stability, decreased
water infiltration, decreased water-holding capacity, a breakdown of microbially
mediated nutrient cycles, and an increase in bulk density (Lai, 1987). Soil erosion
generally induces soil fertility declines by removing topsoil, and hence micro-nutrients
and organic matter, and thus detrimentally affecting crop yields. 165
Popular Perceptions of Soil and Land Degradation
During the colonial period in Kigezi District (1913-1962), the district administration staff characterized the highlands as a region of high population densities and acute land use pressures (Purseglove, 1946; Tothill, 1938; Turyagenda,
1964). The DAO eventually concluded that continuous cultivation and a concomitant reduction in fallow periods were inducing a loss of soil fertility and general soil deterioration that without management interventions would lead to serious soil degradation and food shortages. Consequently, given the high population densities at the time and continued population growth projections, the DAO considered it imperative that agricultural and land management practices in the highlands were changed. The district agricultural officer at the time (i.e., Purseglove) was primarily
responsible for introducing the Soil and Water Conservation By-laws into Kigezi
3 District (e.g., Martin, 1945a; Purseglove, 1945, 1946).
A variety of studies have continued to underscore the detrimental impact of
population pressures on the environment in the highlands; they describe the physical
environment as one with low to medium soil fertility, moderate to high susceptibility
to soil erosion, and widespread soil and land degradation (e.g., Dorsey et al., 1990;
Grenzebach, 1981; Jagannthan, 1990; Langlands, 1974a; Oilier et al., 1969; Tothill,
1938; Tukahirwa, J., 1992; Were et al., 1992; Zerihun, 1994). More specifically, a
soil productivity map described the region as one with soils of low (3) and good (5)
levels of productivity (scale: 1 = nil to 6 = high productivity) (Langlands, 1974a).
3 The by-laws are described in detail in Chapter Five, and in Appendices D and E. 166
Additionally, a soil erosion hazard map of Uganda described the southwestern highlands as a region of high to very high erosion hazards (UNEP, 1988), while maps produced for the African Mountains Association also characterized the highlands as a region of steep slopes, shallow soils and multiple soil constraints (Grosjean and
Messerli, 1988). Specific soil-related problems cited for the highlands include sheet
soil and rill erosion, gullying, mass wasting (e.g., soil slumping, land slides), acidification, nutrient leaching, low cation exchange capacity (CEC), and low levels of soil organic matter and soil organisms (e.g., Bagoora, 1988; Getahun, 1991;
Grosjean and Messerli, 1988; GOU, 1994b; Peden and Kakuru, 1993; Tukahirwa, E.,
1992; Tukahirwa, 1988; UNEP, 1988; Zake, 1991).
As the aforementioned studies reveal, the widespread concern over soil and land degradation in the highland region that originated during the colonial era has not by any means abated. A recent (May 26, 1993) centerpiece headline in a national newspaper explicitly underscores the prevailing view of the state of the environment in the highlands: "Land Degradation on Kabale Slopes" (Ngabirano, 1993). The author described numerous soil-related problems in the highlands and suggested that poor land management would eventually lead to landslides, catastrophic floods, declining soil fertility and eventual food shortages, and climatic change. Indeed, district-level officials, agricultural researchers, rural development workers and
environmental conservationists alike continue to assert that both the viability of the
few protected natural forest and wetland systems, and the sustainability of the 167 highland agricultural systems are threatened (Cunningham, 1992; Cunningham et al.,
4 1993; Peden and Kakuru, 1993).
While the wetland and forest environments have obviously and indisputably experienced extensive clearance in recent decades, the status of soil resources in the
highlands is less clear. Despite the widespread proclamations of serious soil and land degradation in Kabale and Kisoro Districts, the various claims are only marginally supported with empirical, scientific evidence (e.g., Bagoora, 1988). Rather, they are based almost entirely on qualitative observation, the extrapolation of data from macro- level soil and geological surveys and related soil-map projects, the tendency to reiterate popular perceptions about the nature of the highland environment and to perpetuate simplistic interpretations of land use without critical examination, and an overall misunderstanding of local resource management and land use practices. To
date, there has been very little basic soil science research conducted, and there is virtually no information available indicating the degree, extent or frequency of soil- related problems in the highland region. Furthermore, very little research has been undertaken to better understand how smallholders perceive of soil resource management problems, conceptualize soil erosion and soil fertility processes, or actually manage their soil and land resources.
4 In the highlands of Kabale and Kisoro Districts, Mgahinga and Bwindi Impenetrable National Parks, as well as Echuya and Mafuga Forest Reserves, have all experienced agricultural encroachment problems in recent decades. 168
Scientific Evidence of Soil and Land Degradation
Although it has been assumed that dense populations, local land use practices and environmental conditions would combine to induce the collapse of the highland agricultural systems in southwest Uganda, the demise that has been considered inevitable for decades has not occurred. The colonial land management interventions, particularly the Soil and Water Conservation By-laws, might be partly responsible for averting the expected agroecological collapse. There is also evidence, however, to indicate that some soils in the highlands region are resistant to erosion~or at least much less erodible than previously assumed.
Paradoxically, the first data that indicated that soils in the region were not highly erodible were collected in 1945, when the Kigezi District Agricultural Office
(DAO) undertook a field and laboratory study of "soils in the overpopulated areas" in
5 the Kabale Town area. The soils examined were the heavily cultivated soils along the hillslopes and described as reddish-brown loams. Laboratory results of 24 soil samples revealed a mean pH of 6.7, indicating that the soils were well supplied with
bases (i.e., macronutrients). The study also revealed that the soils were well- aggregated with good, stable structures-although the soils in plots that had been intensively cultivated for a number of years did exhibit signs of some deterioration
(Martin, 1945b). Finally, the study reported that "in the Kabale area the soils are
5 In 1943, the population density reported for the Kabale region of Kigezi District 2 2 was 139 people/km , while one smaller area of Kabale supported 277 people/km ; the specific locations or land areas were not reported (Purseglove, 1946). 169
very permeable to water and the rainfall intensity is low so that very steep slopes can
be cultivated without much danger" (Martin, 1945a, 1945b; Purseglove, 1946).
Recent soil erosion research conducted on soils from the Kachwekano
agricultural research station in Kabale District corroborate the findings of the earlier
DAO study. One laboratory study found that clay-based soils (i.e., Ultisols) collected
from hillslopes on the research station outside of Kabale Town were well aggregated,
with stable structures and a good resistance to erosion (Magunda, 1992). The stable
nature of the Kachwekano clay soil (USDA classification: Typic Palehumult) was
attributed to the high alumina and iron (hydr)oxides content, the presence of hydroxy-
6 layered vermiculites and high levels of organic matter.
A soil erosion trial experiment conducted on the same research station also
found comparable results; an analysis of the field data indicated minimal soil erosion
under traditional cropping systems (i.e., sole-crops of sorghum and bush beans)
(Tukahirwa, 1995). This study described the most conspicuous features of the soil
environment as follows: very low interrill and rill erosion rates; relatively stable soils
with a well developed sub-angular blocky structure, high organic matter (5-10%),
7 and highly aggregated kaoline clay. The study found very high rainfall infiltration
6 In a lab-based comparison test, this soil was also found to be more resistant to erosion than a Kabanyolo clay soil (USDA classification: Mollic Kandiudalf) collected from the undulating lowlands of central Uganda (Mugunda, 1992: xi-xii). Martin (1945b) also reported that soils in Kabale were better aggregated with more stable structure than soils in Buganda, in the central districts.
7 Generally speaking, the basic chemical and physical properties of top-soil samples collected during the soil survey for this study reveal similar traits (e.g., moderate levels micro-nutrients and organic matter); these data are presented in Appendix G. 170 rates (1637 millimeters/hour (mm/h) to 2021 mm/h), gentle rainfall intensity, relatively few intensive storms (Tukahirwa, 1995), low runoff and minimal soil
8 erosion. (On slopes of 10%, 25% and 45%, the runoff and soil erosion rates were:
63,000, 270,000 and 138,000 1/ha/yr, and 1.4, 38.0 and 9.3 t/ha/yr, respectively.)
The study found the major factors that influenced the runoff and soil erosion were rainfall intensity, slope steepness and surface cover. While the two recent erosion studies were collected from a limited number of sites on one research station (as opposed to field-plots under smallholder conditions), Tukahirwa' s (1995) study was
managed according to basic smallholder practices and still revealed minimal erosion.
The results from the colonial study, the two contemporary studies and field observations indicate a resistance to erosion among soils in the highlands that has not been previously recognized or widely acknowledged. The studies demonstrate that contrary to the popular perception among DAO staff and various outside experts, some of the soils are not highly easily eroded and correspondingly, that the macro- level generalizations about soil hazards in the highlands are inaccurate. The research data cannot be extrapolated to the diversity of soils in Kabale and Kisoro Districts, but they do emphasize the inaccurate assumptions and the abundance of mis- information that characterize the current knowledge of soils in the highlands; they also underscore the overall paucity of scientific data for soils in the region and the need
Correspondingly, field observations made during the course of the study also indicate that severe soil erosion and soil slumping were infrequent and occurred only after high-intensity rainfall events. 171
9 for additional soil surveys and soil investigations. Finally, while crop-cover
management of the field-plot and the steepness of the slope cultivated also influence
erosion, there is evidence that many smallholders in the highlands utilize a variety of
effective soil resource management practices, and that they have modified many of
their agricultural and land management practices in response to changing conditions of
the highland agroecosystem.
Smallholder Perceptions of Soil Degradation
Soil erosion is a natural process that occurs regardless of human interference
in an environment, although human modification of the environment can also induce
or exacerbate the process (e.g., Millington et al., 1989). In the highlands of
southwestern Uganda there is some evidence of human-induced soil and land
degradation, and it is largely related to agricultural production. During the course of
this study, a number of types of soil degradation were observed in the field at each of
the study sites. They included sheet and rill erosion in agricultural plots, soil
slumping on steep slopes, physical erosion on cinder cones, and gullying along
10 footpaths, cattle trails and roadways. The observations of such events, however,
9 There have been no scientific studies or soil surveys conducted on the volcanic soils in the southwestern area of Kisoro District; field observations made during this study as well as smallholder observations reveal that many of the volcanic soils are highly permeable and resistant to water erosion. Conversely, many of the volcanic soils have large particle sizes and are quite friable; hence those volcanic soils located on slopes
(e.g. , cinder cones) are quite susceptible to physical erosion, particularly that induced by cultivation or grazing.
10 Sheet erosion - the removal of a thin layer of soil from the land surface by rainfall runoff; rill erosion - the formation of small erosion channels several centimeters in depth, often found in cultivated soils; gully erosion - similar to rill erosion but occurring on a 172 provided no indication of the extent or severity of the phenomena. While both natural and human-induced forms of soil and land degradation occur in the highlands,
there is little understanding of their extent or how smallholders manage for the problems.
The majority of smallholders in the study sites acknowledged that some form of soil erosion occurred within their respective parishes, and also within many of the plots that they cultivated." Smallholders were also able to identify a number of different types of erosion and mass wasting processes, including sheet, rill, gully and wind erosion, and soil slumping and landslides. Correspondingly, smallholders in
Kicumbi, Nyarurambi and Muramba Parishes were asked to identify the types of erosion they observed in their agricultural plots; their responses are provided in Table
7.1. 12
larger scale; they can be many meters in depth and width; soil slumping - the downslope movement of soil, often induced by water saturation and the subsequent reduction in the shear strength of a soil and/or the disturbance of a shallow soil lying on bedrock; and landslides - similar to soil slumping but occurs on a larger scale. The latter two forms of mass wasting occur naturally, also can be caused by cultivation and overgrazing.
11 Smallholders in neighboring Rwanda also characterized as an overpopulated highland region with extensive land degradation, have also demonstrated an awareness of a number of environmental problems as well as an ability to adapt to some of them (Ndiaye and Sofranko, 1994).
12 The high number of "Don't Know" responses in Muramba Parish is attributed more to the minimal relief in the parish and the general reticence of smallholders to participate in the survey than actual unawareness of the phenomenon itself. The discrepancies between the size of the population sample in each parish and the actual number of responses is due to missing or incomplete responses. 173
The majority of the smallholders surveyed recognized some form of soil erosion within their plots. A sizeable portion of smallholders in each parish identified
rill erosion as a common feature in their plots, although sheet erosion was not as
Table 7.1. Smallholder Perceptions of Soil Degradation.
Smallholder Responses Do you have on 1. Kicumbi 2. Nyarurambi 3. Muramba your field-plots? (n = 67) (n = 95) (n = 185)
1. sheet erosion: Yes: 4 ( 6%) 19 (20%) 3 ( 2%) No: 52 (78%) 74 (79%) 81 (49%) Don't Know: 11 (16%) 1(1%) 81 (49%)
/O 1 Of \ ^ 1 (HC Of \ /TQ /A'} Of \ 2. rill erosion: Yes: 54 (ol%) 11 (/Otd)
No: 2 ( 3%) 22 (23%) 16 (10%) Don't Know: 11 (16%) 1(1%) 80 (48%)
3. gully erosion: Yes: 10 (15%) 45 (48%) 24 (15%) No: 46 (69%) 48 (51%) 61 (37%) Don't Know: 11 (16%) 1(1%) 80 (48%)
4. wind erosion: Yes: 6 ( 9%) 48 (52%) 22 (13%) No: 50 (75%) 45 (48%) 63 (38%) Don't Know: 11 (16%) 80 (48%)
5. soil slumping: Yes: 29 (44%) 54 (57%) 6 ( 4%) No: 27 (40%) 39 (42%) 78 (47%) Don't Know: 11 (16%) 1(1%) 81 (49%)
6. landslides: Yes: 0 ( 0%) 1 ( 1%) 4 ( 2%) No: 56 (84%) 92 (98%) 80 (49%)
Don't Know: 11 (16%) 1 ( 1%) 81 (49%)
widely reported. The minor reporting of sheet erosion is likely due to its
inconspicuous nature, while rill erosion is more readily observable in the field. Gully erosion was identified as a problem by approximately half the respondents in
Nyarurambi, while it concerned but a small percentage (15%) of respondents in 174
Kicumbi and Muramba Parishes. Soil slumping was identified by roughly half of the respondents in Kicumbi and Nyarurambi Parishes, while smallholders in Muramba reported very low occurrences of soil slumping and gully erosion, most likely a result of the minimal relief in the parish. Very few respondents in any of the study sites reported incidences of landslides.
Smallholders in the highlands indicated a general awareness of soil degradation problems, but they also revealed some uncommon perspectives on soil processes and soil resource management. For example, in both Kicumbi and Nyarurambi Parishes, a number of smallholders pointed out that during erosion events top-soil and nutrients
were not "lost" per se, but rather "transferred" elsewhere in the landscape, i.e.,
although soils "erode" and are transported downslope, the soil movement is essentially from plot to plot. A few smallholders described this scenario: given that most smallholders had plots scattered throughout the landscape, they generally both lost and gained soil during specific erosion events and in a cumulative way throughout the course of a year.
While smallholders did not necessarily suggest that any parity was achieved with the plot-to-plot movement of soil, (e.g., they noted that ridge-top plots experienced a net loss of soil and have become increasingly less productive), neither did they simply conceptualize soil erosion as a destructive process that must be halted at all costs. Furthermore, for many smallholders, reducing erosion was not one of 175 their management priorities, even though they generally acknowledged that it did
13 often negatively impact on agricultural production.
Smallholder Soil Resource Management Practices
As discussed in Chapter Five, a variety of historical studies and travel accounts suggest that smallholders traditionally utilized a number of effective soil and land management practices. For example, they generally avoided steep-sloping areas, maintained near-continuous vegetation cover on plots, practiced minimum tillage and broadcast planting, often inter-cropped and rotated crops, and carefully scheduled land preparation and agricultural production activities in accordance with the rainy and dry seasons. Smallholders also used a variety of soil conservation measures to manage their land resource, many which are currently recommended within the scientific community. These included plant-trash lines and modified terraces to control for erosion, raised beds to avoid flooding, and fallows to manage soil fertility (e.g.,
Tukahirwa, 1995; Rugyema, 1974; Reij, 1991). Smallholders continue to employ
many of these practices-some of which are currently prescribed in the By-laws- although the different practices are not equally popular, and the use of any given
practice is by no means uniform within or across the three study sites.
Soil and Water Conservation By-Laws
The Soil and Water Conservation By-laws (SWC) introduced during the
colonial period remain legal statutes in Kabale and Kisoro Districts. Indeed, the
13 The agricultural and resource management concerns and priorities of smallholders are discussed further in Chapter Eight. 176 entirety of the exacting land management regulations delineated in the original SWC program are included in the contemporary SWC By-laws in Kabale and Kisoro
Districts, while the program has also remained top-down in management approach.
However, since independence (1962), and particularly with the advent of Idi Amin's rule (1972), the soil and water conservation program have experienced limited implementation. Reviews of colonial reports and discussions with DAO staff and smallholders indicate that the use of the SWC By-laws has decreased throughout the highlands in recent decades; the By-laws are now infrequently promoted or enforced by the DAO staff and parish chiefs, and adopted only selectively by smallholders.
To investigate the status of the SWC By-laws in Kicumbi, Nyarurambi and
Muramba Parishes, smallholders were asked whether they were aware of the soil and water conservation By-laws, and whether they were enforced. Additionally,
regardless of their responses to the first two questions, smallholders were also asked whether they utilized specific soil conservation measures; their responses are presented in Table 7.2. In Nyarurambi and Kicumbi Parishes, the majority of responding smallholders, (99% and 75% respectively), stated that they were aware of district-level Soil and Water Conservation By-laws, while only a minority (20%) of smallholders in Muramba Parish recognized the existence of By-laws. In Kicumbi
Parish, only 33% of the respondents who reported an awareness of the By-laws stated that they were actually enforced by authorities, while in Nyarurambi and Muramba
Parishes all of the respondents who indicated an awareness of the regulations stated 177
that they were enforced. (However, the number of respondents reporting an awareness of the By-laws in Muramba was only 24 of a possible 185.)
Table 7.2. Smallholder Perceptions and Use of the SWC By-laws.
Smallholder Responses
Soil and Water Conservation Measures 1. Kicumbi 2. Nyarurambi 3. Muramba (n = 61\ (n = 185)
1 Arp thprt* Hi«trirt-lf*vpl ^oil and watpr conservation by-laws? Yes: 47 (75%) 88 (99%) 29 (20%)
No: 16 (25%) 1 ( 1%) 115 (80%)
2. Are the by-laws enforced? Yes: 15 (33%) 89 (100%) 24 (100%) No: 31 (67%) 0 0
3. Do you use any of these practices to control for soil erosion in you plots?
a) Grass bunds: Yes: 8 (13%) 74 (79%) 21 (13%) No: 53 (87%) 20 (21%) 140 (87%)
b) Hedge rows: Yes: 10(16%) 16 (17%) 14(9%) No: 84 (84%) 76 (83%) 146 (91%)
c) Contour Planting: Yes: 47 (72%) 77 (83%) 93 (58%) No: 18 (28%) 16 (17%) 68 (42%)
d) Trash Lines: Yes: 19 (31%) 80 (85%) 17 (11%) No: 43 (69%) 14 (15%) 142 (89%)
e) Mulching: Yes: 5 ( 8%) 9 (10%) 35 (22%) No: 57 (82%) 82 (90%) 124 (78%)
0 Run-off: Yes: 19 (31%) 31 (33%) 44 (28%) Channels No: 42 (69%) 62 (67%) 116 (72%)
g) Soak-way pits: Yes: 2 ( 3%) 5 ( 6%) 2(1%) No: 58 (97%) 86 (94%) 157 (99%)
h) Check-dams: Yes: 1 (2%) 3 ( 3%) 1 ( 1%) No: 58 (97%) 86 (96%) 156 (99%)
Among the three parishes, the survey results indicate that the smallholders in
Nyarurambi Parish were the most aware of, and the greatest implemented of, the 178 various SWC measures. Comparatively speaking, these results were reflected to a lesser degree in Kicumbi, while smallholders in Muramba exhibited the least awareness and use of the prescribed soil conservation measures. In nearly every category describing the use of specific soil conservation measures (Table 7.2: 3a - h), respondents in Nyarurambi parish indicated the highest use, followed by smallholders in Kicumbi and Muramba parishes, respectively. This pattern was altered only in the category of "mulching" (Table 7.2: 3e), where respondents in Muramba reported the greatest use (22%), followed by Nyarurambi (10%) and Kicumbi (8%).
This general trend among parishes was unexpected given that the lowest
population density and correspondingly the most land available is in Nyarurambi (322
2 2 people/Km ), followed by Kicumbi (406 people/Km ) and finally Muramba (621
2 people/Km ) parishes. Generally speaking, the results imply an increasing awareness and use of the By-laws with decreasing parish-level population densities. This trend runs contrary to what might be expected given a Boserupian-type interpretation of the situation: on one end of the spectrum the smallholders in Nyarurambi would be expected to have less impetus to employ the By-laws due to the lower population densities, while the land shortages in Muramba would be expected to induce
smallholders to widely employ some of these soil conservation measures.
While this trend might simply be an anomaly, there might be a number of other explanations for the results. First, the results might be attributed to the differential degrees that the DAO and parish chiefs proselytize and enforce the By- laws in each parish. Second, the smallholders might value the benefits of the soil 179 conservation measures to different degrees depending on their tradition of use in each
parish. As discussed previously, there is some evidence to indicate that many of the
mandated SWC measures also have traditional origins and it is difficult to differentiate between the two impetuses. Furthermore, the practices that some younger smallholders may have learned from their elders and now consider "traditional" might actually have their origin in an earlier era of By-laws, when the DAO actively
14 enforced by the regulations. Third, it is conceivable that some smallholders, particularly in Muramba, did not want to acknowledge awareness of the By-laws for fear that they would be fined for not abiding by them. Fourth, the inter-parish differences in by-law implementation might also underscore the environmental variation in the region. For example, of the three parishes, Muramba is characterized by the least relief, and some of the most permeable and fertile soils. Consequently, smallholders in Muramba might have determined that the soil conservation measures prescribed in the By-laws were inappropriate given their particular management and/or environmental situations, or did not provide worthwhile returns for their time and labor investments. 15
Overall, it appears that the SWC By-laws are not employed to the degree that they were in the colonial and early independence days; many of the prescribed soil
14 For example, interviews with elders in Kisoro indicate that the raised-bed agricultural system was introduced by the DAO in the 1940's. However, because younger smallholders learned the practices from their elders, they identify the raised-beds as a local technology.
15 Paradoxically, as described later in this chapter, smallholders in Muramba listed soil erosion second only to land shortage as a agricultural and land management problem. 180 and water conservation practices have been abandoned (e.g., check dams and soak- way pits) or are poorly managed (e.g., grass bunds), while prohibited practices like grass burning are fairly common. The survey results did reveal, however, that a number of soil and water conservation practices mandated in the By-laws are still utilized in each of the study sites. Of the various practices, the biological measures were more popular than the physical structures (i.e., "engineering technologies").
Smallholders indicated that this was largely because the biological measures were multi-functional (e.g., hedgerows provide firewood and demarcate plots) and required less labor for construction and maintenance than the physical ones. Of the biological measures, the majority of the respondents in all three parishes utilized contour planting, while a small percentage of smallholders used trash lines and grass bunds.
Of the physical constructions, only run-off channels were employed to any significant degree, and only then by about one-third of the smallholders in each parish.
Paradoxically, smallholder- managed run-off channels-which are still prescribed in the district-level By-laws-are quite often characterized as erosion gullies by DAO staff, local research staff and outsider experts.
Other SWC measures that remain widely used among smallholders include
raised-beds in Muramba, and grass bunds in Kicumbi and Nyarurambi Parishes. For
example, raised soil-bed agriculture was introduced by the DAO during the colonial
era and is still widely employed throughout Kisoro District in Muramba Parish.
Smallholders have pointed out that the raised beds are very effective in protecting
crops from flooding after heavy rains. Additionally, they deposit agricultural detritus 181 and household refuse between the beds, rotate the beds on to the detritus at the end of the season, and thus continually incorporate new organic matter into their plots. The widespread adoption of raised beds and their continued use underscore the willingness of smallholders to adopt and utilize exogenous technologies, particularly when they require minimal capital or labor investment and the benefits can be readily perceived.
Another introduced soil conservation measure, grass bunds planted along the hillslope contour, also remain a prominent landscape feature in the highlands- although relatively few smallholders actually reported using them in the survey.
Nonetheless, of those smallholders in Kicumbi and Nyarurambi Parishes who utilize grass bunds, few have invested much effort in bund maintenance, or attempted to
maintain the bund-width (1 meter) prescribed in the By-laws. While many smallholders recognized the erosion control benefits of the grass bunds (i.e., they effectively trapped eroded topsoil), they also noted that the bunds contributed to other management problems and they give a number of reasons for their reluctance to improve their bund management.
First, the "down-slope" smallholder is often tempted to undermine the upper
bund during land preparation, thereby inducing its collapse. The individual
downslope is then able to "harvest" the fertile soil that was being conserved in the bund, and thus reap the benefits of someone else's management efforts. Secondly, wide grass bunds also attract grazing livestock that often raid crops and also cause bunds to collapse. Furthermore, there is also a strong gender-based dimension to the conflict between crop production and land management, and uncontrolled livestock 182 grazing. Women are the primary agricultural producers, and men are the principal livestock owners and caretakers, as well as landowners. Correspondingly, while women have access to land and some related resources (e.g., compost, manure), men generally possess absolute control of the resources, thus the two often have different resource management concerns. For example, while women often cite uncontrolled cattle grazing as one of their primary agricultural problems, most men deny that cattle
and small livestock present any land management problems, and in the end little is done to address the issue.
Given these various problems related to grass bunds, few smallholders were willing to invest in intensive bund management or follow the specific guidelines of the
By-laws. However, smallholders also rarely destroyed or completely abandoned the bunds either. While smallholders generally recognize the soil conservation benefits of the grass bund, the bunds have not maintained their prominence in the agricultural system simply because smallholders perceived the soil conservation benefits to be greater than other costs (e.g., many smallholders also noted that bunds occupied potential crop-land). Rather, smallholders pointed out that the bunds comprised an integral part of the local land tenure system and served to demarcate individual plots boundaries. Presently, the grass bunds are now more common than the native trees and hedgerows that were traditionally used to mark boundaries.
As the previous sections have revealed, there is irregular implementation of the SWC By-laws among smallholders in the three study sites in Kabale and Kisoro
Districts. Interviews with DAO staff and smallholders reveal a variety of additional 183
reasons that account for the general deterioration of the once widely-lauded land
management program. Within the DAO in Kabale and Kisoro, the reasons for the
irregular extension of the By-laws include the overall lack of funding, poor salaries,
minimal transport and per diem funds to support extension work, institutional inertia,
and the lack of motivation among staff members. Among smallholders, the reasons
for their irregular adoption of the By-laws include a disinterest in and/or disagreement
over the necessity of SWC measures, the lack of DAO extension assistance, the high
labor inputs required to construct and maintain SWC structures, uncontrolled cattle
grazing (which can cause crop loss as well as the destruction of SWC structures),
insecure land tenure, and gender-based inequities in terms of control of and access to
resources. Furthermore, many smallholders perceive of the externally-managed
agricultural and land management programs as intrusive, manipulative and
coercive. 16 Unfortunately, the general ineffectiveness of the SWC By-laws and the
divergence of opinion between the smallholders and DAO and various outside experts
has not stimulated additional investigation on soils and soil resource management in
the highlands.
16 For example, in 1992-93 smallholders in southwest Kisoro became quite hostile with development and conservation organizations when the implementation of a new
management program for Mgahinga National Park (i.e., Virunga Volcanoes) also involved a re-demarcation of the park boundaries. As a result, many smallholders were evicted from homesteads and/or prohibited from access to agricultural plots on the footslopes of Mgahinga volcano. While the new park management plan largely re- established old boundaries, the park had been poorly managed for decades and the local people had considered it their right to use the land, especially given the land shortages. Similar cases of land alienation have been described elsewhere in Kabale and Kisoro Districts (e.g., Mugisha, 1992). 184
Input Utilization
The concern over soil management and agricultural production in the southwest highlands has not been limited to district officials and outside experts.
Many smallholders have also expressed a concern over soil erosion and declines in
soil fertility and agricultural productivity. While soil erosion is often directly linked to the loss of topsoil and contributes directly to the deterioration of soil fertility, many smallholders attributed much of the decline in crop productivity to continuous cultivation with insufficient soil amendments. Many of smallholders' soil erosion control practices were discussed in the previous section. To better understand how smallholders are specifically managing for soil fertility, they were asked about their use of organic and inorganic inputs. The responses are presented in Table 7.3.
In Kabale District, one World Bank funded study found that out of 180 households, 34% had purchased some type of agro-chemicals and 53% had purchased seed (World Bank, 1989), while another study reported that 11.5% of respondents (n
= 79) had used fertilizers (CARE, 1994). However, the survey results presented here reveal an even lower level of input use among smallholders in the highlands.
The survey also corroborates exactly what smallholders expressed in the focus groups and individual interviews: that the vast majority of smallholders in both Kabale and
Kisoro Districts do not commonly use purchased inputs (e.g., fertilizers, pesticides, herbicides, fungicides, improved seeds) in their agricultural production systems. 185
Table 7.3. Smallholder Use of Agricultural Inputs.
Smallholder Responses Do you use the following 1. Kicumbi 2. Nyarurambi 3. Muramba inputs? (n = 67) (n = 95) (n = 185)
103 1. Manure: Yes: 48 (87%) 66 (76%) (64%) No: 7 (13%) 21 (24%) 59 (36%)
2. Compost: Yes: 53 (91%) 84 (97%) 155 (95%)
No: 5 ( 9%) 3 ( 3%) 8(5%)
127 3. Plant Trash: Yes: 20 (38%) 69 (79%) (78%) No: 33 (62%) 18 (21%) 35 (22%)
4. Ashes: Yes: 2 ( 4%) 4 ( 5%) 33 (21%) No: 50 (96%) 80 (95%) 128 (79%)
5. Fertilizers: Yes: 0 0 0 No: 52 (100%) 84 (100%) 160 (100%)
6. Pesticides, etc.: Yes: 1 ( 2%) 6 ( 7%) 1 ( 1%) No: 51 (98%) 78 (93%) 159 (99%)
7. Improved Seeds: Yes: 1 ( 2%) 8 ( 9%) 0 No: 51 (98%) 78 (91%) 159 (100%)
In Kicumbi, Nyarurambi and Muramba Parishes, none of the respondents used
fertilizers, while less than 10% of respondents in each parish used other agro- chemical products or improved seeds. The few smallholders who did use pesticides utilized them on cash crops. For example, in Nyarurambi Parish, the agro-chemical inputs used were fungicides, and they were applied to Irish potatoes that were destined for the markets in Kampala. Smallholders also expressed an interest in purchasing improved seed and agro-chemical products if they were available and reasonably priced. Unfortunately, while a very limited array of agro-chemical products were available in Kabale and Kisoro towns, they were prohibitively agro- 186 chemical products available for sale in the periodic markets in the rural areas that smallholders normally frequent.
The vast majority of smallholders across the three study sites did not utilize purchased, agro-chemical inputs. The majority of them did, however, use locally available organic amendments in their agricultural plots. Over 90% of the respondents in Kicumbi, Nyarurambi and Kisoro Parishes used compost, while a
17 sizeable majority (> 64%) in each parish used manure. In Nyarurambi and
Muramba Parishes, over three-quarters of the respondents used plant trash or agricultural detritus in their plots; only in Kicumbi did a minority of smallholders
(38%) use plant trash. Ashes were the only locally available organic input that
smallholders did not widely use, although it was used more in Muramba Parish (21%) than the other two parishes. Unlike with the implementation of the SWC measures,
there is no obvious difference in input use between the three study sites.
Although input management was not specifically addressed in the survey questionnaire, smallholders did reveal some of their practices in focus groups and individual interviews. For example, many smallholders across the study sites
18 indicated that organic inputs, particularly manure, were often in short supply.
17 Compost here refers to household refuse and post-harvest crop detritus. It is generally transported to the field "green" or partially decomposed and then incorporated into the soil; it is only rarely completely processed in a pit and rendered true compost.
18 While many smallholders raise some small livestock (e.g., sheep, goats), only those individuals with cattle generally have access to reliable sources and ample supplies of manure-although a few people are able to purchase or trade for it. Additionally, it is generally the "wealthy" individuals who own cattle, while the overall cattle population 187
Smallholders also indicated that the distance to the plot as well as the specific soil
19 type in a plot partly determined whether or not inputs were applied. The more distant plots generally received fewer amendments, although smallholders did indicate
that if they cultivated few plots they were generally more willing to transport the amendments greater distances to enhance crop production.
Regarding soil factors, many smallholders noted that they could generally apply moderate amounts of amendments on productive soils (e.g., Enombe) and achieve visible increases in crop production, while only application of significant amounts of inputs could affect an increase in crop production on the poorer soils
(e.g., Orucueu). Smallholders generally did not apply amendments to the less productive soil types unless, again, they had few relatively plots. Consequently, smallholders often reserved their organic inputs for plots closer to the household compound, and for plots with at least moderately productive soils.
Additionally, much of the burden for managing compost and manure and transporting them to the field fell on women. The large number of responsibilities that women in the highlands also shouldered (e.g., child-rearing, water and fuelwood
in the highlands is quite small.
19 Smallholders generally cultivated multiple plots, and the average number of plots per respondent in Kicumbi, Nyarurambi and Muramba parishes was 5.6, 9.3 and 3.0, respectively. (The nature of the fragmented landholding system is further examined in Chapter Eight.) The distance to the various plots of the survey respondents was not systematical measured, although many distances were evaluated during the course of the fieldwork. While many plots were located near the household compound, some smallholders also cultivated plots that were over 5 kilometers, or a two+ hour walk, from their home. 188 gathering, cooking) meant that the frequency of input applications was often dependent on the availability of both the input-which was often quite irregular-and
the time to transport and apply it. Consequently, given the considerable distances to some plots, the variations in soil types among plots, and the multiple household and agricultural responsibilities of women, the fertility management of different plots was often quite variable. Furthermore, there was often a gender-based dimension to the management of organic inputs: men often assumed the priority-rights to use manure, and occasionally compost, on their cash crops. Thus women only had access to the remains of organic inputs, which could occasionally amount to nothing at all.
Finally, for many smallholders, soil erosion and soil fertility declines involved a variety of processes with both positive and negative results. For example, a number of smallholders in Kicumbi Parish not only recognized the erosion process, but they
in part manipulated it in attempt to manage soil fertility on their plots. The smallholders recognized the existence of a fertility gradient in most hillside plots that generally runs from the fertile soil located "downslope" behind the grass bund, to the back of the plot upslope, where the soils are generally exposed, inherently infertile sub-soils. This fertility gradient has evolved as a result of erosion caused by run-off and erosion induced by cultivation, whereby soil is moved downslope during land
20 preparation. Smallholders also noted that crop production was lowest at the back
20 Smallholders generally face into the slope when they cultivate a plot; tilling the land with a hoe, they normally pull the soil downslope. This type of human-induced
"mechanical erosion" was considered to accelerate soil loss on steep slopes in neighboring Rwanda (Lewis, 1992). 189 or "up-slope" portions of a plot, and that area generally required the heaviest application of organic amendments.
Consequently, they applied compost or manure only to the backslope portion of the plots. They did this not only because it was the most nutrient deficient portion of the plot, but because they recognized that the erosion processes would eventually transport or "distribute" the manure and other nutrients down-slope over the remainder of the plot. While this practice might not be the most effective means of
managing manure or maintaining fertility, it does underscore smallholders' basic
understanding of some aspects of the soil erosion and soil fertility processes in the
highlands. While smallholders did not necessarily understand all aspects of the
processes, they do possess their own unique perspectives.
Agricultural and Resource Management Concerns
In Kicumbi, Nyarurambi and Muramba Parishes, smallholders were concerned
about a number of agricultural and resource management issues in addition to soil
erosion and soil fertility declines. Some of the issues that smallholders raised
included land and labor shortages, soil degradation, tool shortages, uncontrolled
grazing, disease and pest problems, declining crop yields, and distance to field plots,
expensive for the majority of smallholders. Furthermore, there were virtually no
Smallholders were asked to identify their leading concerns and rank them in order of
importance from 1 to 5, where 1 was the greatest concern, and 5 was the least
pressing of the concerns. The major agricultural and land management issues are
listed in Table 7.4. Furthermore, the number of times a problem was ranked first is 190
Table 7.4. Smallholder Agricultural and Land Management Concerns
Results from a Ranking Exercise (1 - 5): Frequency that
Problems were Cited First: 1) and Cited in Total: 2).
Management Concerns 1. Kicumbi 2. Nyarurambi 3. Muramba
1X. L^BLU\JlT flhor JI1UIShortao^l«Lt 1) 14 1) 20 1) 9 2) 32 2) 32 2) 37
£>• LallU JHUI UlLV 33 38 110 42 52 137
^ I1 QfiH111 Fra ompn raft r*n j • i i agiiiciiuiiiuu 0 1 2 6 3 35
d. T^ictanpf* to Pirate 3 4 20 13 30 37
S Tool Shorts of* 2 11 17 30 47 35
T Tnpnntrr^l Ip/i (""rraTino 5 2 4 22 15 62
7 i~ji O . L/C^l llllllg V_^i UU 1 lCllXo 3 1 2 28 31 48 07 * OftrlinpX,/ Cv Uof1 V_-("Von1 V_f LI I\UUIRotation*;1 1XJX1S 0 0 8 4 lint* XV*10 OSoilV 1 i XF*»rtilitvCI 1111LY 1_/CV1X11COre* 2 2 0 47 45 19 1 1 Sr\i 1 Prr»cinn 0 11 18 11 53 98 1 iz. ucciinc oi rauows 1 6 20 13 13. Lack of Fertilizers 0 0 1 3 16 36 14. Lack of Pesticides 0 4 1 3 27 35 IS. Lack of Markets 0 1 0 3 18 10 16. Lack of Improved Seeds 4 4 0 9 25 29 191 indicated, as is the total number of times the problem was cited irrespective of its actual ranking. In terms of how smallholders prioritized their concerns, the issue that was cited as the number one concern the most often in each parish was land shortages. The priority concern that was cited the second most often in Kicumbi and Nyarurambi parishes was labor shortage, while in Muramba it was the distance to field plots. The priority concern that was cited the third most often was uncontrolled grazing in Kicumbi Parish, tool shortages and soil erosion in Nyarurambi Parish, and soil erosion in Muramba Parish. In terms of how often a concern was cited in total by smallholders, irrespective of its actual ranking, a slightly different picture was presented. In Kicumbi Parish, the most cited problem was soil fertility decline, followed by land shortage and labor shortage. In Nyarurambi Parish, soil erosion was cited the most often, followed by land shortage and tool shortage, while in Muramba Parish, land shortage was cited the most often, followed by soil erosion and uncontrolled grazing. In Muramba Parish, it was unexpected that such a high number of respondents would cite soil erosion as a management problem, primarily because the local soils are quite permeable to rainfall and there is minimal slope. The high number of responses might be ascribed in part to the sporadic yet destructive torrents that course off the volcanoes through standing-gully s during the rainy season, the active implementation the 21 of SWC By-laws by the DAO , as well as the long-term 21 The Kisoro DAO was quite active in implementing the SWC By-laws, particularly when compared with the Kabale DAO. 192 inculcation of smallholders with the belief that soil erosion was pervasive in the region. While the array and ranking of problems were slightly different in each parish, a number of cross-parish themes do appear. First, land shortage was the major agricultural and land management problem cited by the respondents in all of the study sites. Thus it appears that regardless of the population density of the respective parishes, smallholders are concerned with land availability. Second, soil-related problems, including soil erosion and soil fertility declines, were clearly recognized as a problem by respondents in all of the parishes. Most probably related to soil problems, a number of smallholders in each parish also expressed a concern over declining crop productivity. There were also some differences between parishes: in Kicumbi and Nyarurambi, labor shortages were a major concern, while in Muramba the distance to field-plots was an important concern. There is some indication that these differences were related to population density. For example, respondents in the least densely populated parishes (i.e., Kicumbi and Nyarurambi) indicated a concern over labor shortages. On the other hand, in the most densely populated parish of Muramba, while labor was not in short supply, the land shortages and land fragmentation meant that smallholders cultivated plots quite distant from their homes. Finally, although they were only intimated in the survey, during many focus groups smallholders also identified a number of social and economic issues that affected their agricultural and land management practices. These included the low annual incomes (Uganda: GNP = $163, GDP = $142) (e.g., shortage of capital to 193 purchase tools, seed), land tenure conflicts and their implications for land management (e.g., grass bunds management), gender-dimensions of agricultural production and land management problems (e.g., uncontrolled cattle grazing), gender- based control of and access to resources (e.g., manure), and timely and equal access to markets (e.g., for cash crops like Irish potatoes). Summary In Kabale and Kisoro Districts, in the highlands of southwest Uganda, it has been assumed since the early colonial period that soil resources are being degraded. The soil and land degradation is considered to be the result of a combination of a variety of physical and human factors: an environment characterized by steep hillslopes and moderately shallow to shallow soils along the hillslopes and ridgetops, and agricultural production systems characterized by overpopulation, land shortages, intensive land use and poor resource management. While these environmental and social conditions can lead to resource degradation, it is not necessarily automatic. For example, the limited scientific data available indicate that some soils in the highlands are more fertile and less susceptible to erosion than previously assumed. Correspondingly, these data emphasize the danger of generalizing about soils given the environmental and soil diversity in the region, and the attendant hazards of developing Soil and Water Conservation By-laws or other land management policies based solely on conjecture and superficially examined assumptions. They also underscore the need to initiate additional soil surveys and erosion studies to better understand soils in the highlands. 194 Soil and land degradation undoubtedly occurrs in the highlands, and smallholders themselves have identified soil erosion and soil fertility declines as two of their primary agricultural and land management concerns. And while very little is understood about the scale, frequency, magnitude and causes of soil related problems in Kabale and Kisoro Districts, there has been less known about the soil and agricultural knowledge and management practices of smallholders. This study has demonstrated that many smallholders in Kabale and Kisoro Districts utilize a variety of soil resource management practices that are of both indigenous and exogenous origins, and many of the practices that smallholders use are also prescribed in the district-level SWC By-laws. Not all smallholders fully comply with the SWC By-laws, however, and there are a variety of reasons behind smallholders' selective use of the prescribed soil and water conservation measures. For one, some smallholders in each parish have stated that the SWC regulations are reminiscent of the colonial presence in the region. While it is but one of many factors that influence the land management decisions of smallholders, this perspective underscores the persistence of negative colonial influences and top-down management approaches, and the importance of understanding the historical dimensions of land use and resource management in the highlands. Additionally, smallholders did not perceive the benefits of all the By-laws, nor did they necessarily have the means (e.g., time, labor) to implement all of the recommended measures when they did. This is particularly true for women, who were burdened with the responsibilities of household management in addition to 195 agricultural production and land management; even where men assisted women with the field work, there was not always sufficient labor or interest present. Furthermore, in addition to the individual differences, there were also inter- parish differences among smallholders in the awareness and implementation of the By- laws. The basis for these differences cannot be fully determined, but some influences include the different levels of activism among district- and parish-level extension personnel, motivations and labor supplies among smallholders, and parish-level environmental conditions and constraints. In total, the implementation of the By-laws can be best characterized as follows: they are periodically revised by the District Agricultural Office (DAO), occasionally enforced by the DAO staff and village chiefs and selectively utilized by the smallholders. In the highlands of southwest Uganda, the use of purchased agricultural inputs among smallholders is quite limited. On the whole, the use of purchased inputs such as fertilizers, pesticides and improved seeds were utilized by 0-5 % of smallholders surveyed in the region. Unfortunately, even if smallholders were willing, able and economically situated to expand their use of purchased inputs, there remains serious supply and distribution problems in most of the highlands. There is evidence from other agricultural areas in Africa where with improved infrastructure and market opportunities, smallholders adopted fertilizers and increased crop production (Goldman, 1993b). If fertilizers and other agro-chemical inputs were made available to smallholders in the highlands, a similar change might take place: the agricultural conditions in the region indicate that there is now probably an opportunity for such an 196 investment. Smallholders in Kabale and Kisoro Districts do utilize a variety of locally available organic inputs, including manure, compost and agricultural detritus (i.e., plant trash), to help manage soil fertility. The use of the organic inputs, however, is not uniform among smallholders or across plots, and it is influenced by the availability of organic materials, distance to plots, soil types and the gender- dimensions of labor availability and inputs. Overall, the majority of smallholders in the highlands were interested in improving soil resource management within the agricultural production systems. While many smallholders considered soil erosion and soil fertility declines in part a "cost" of agricultural production, most smallholders also stated that soil resource management could be improved given the appropriate resources and the mitigation of other technological and social constraints. In addition to soil-related problems, some of the primary smallholder agricultural and land management concerns were land, labor and tool shortages, uncontrolled grazing, few market opportunities for cash crops, land tenure conflicts and the gender-based dimensions of agricultural production and resource management. In the following chapter, Chapter Eight, the ecological, agricultural and land use types of change in the highland region are examined. The various factors that have influenced the changes, as well as their different manifestations across the three study sites are also explored. Finally, the abilities of many smallholders to adapt in various ways to the changing conditions are examined and the dynamic nature of the highland agricultural production systems is underscored. CHAPTER 8 LAND USE AND AGRICULTURAL CHANGE Environmental Change in Southwest Uganda The highlands of southwest Uganda have a long history of environmental change, although the catalysts for the change and the specific time frame involved remain unclear (e.g., Hamilton, 1982; Morrison, 1968; Morrison and Hamilton, 1974; Webster, 1979). The historical record of vegetational change in the southwestern highlands has been difficult to determine precisely, but the pollen data collected from highland swamp cores in Kabale District provide evidence for forest clearance in the region from roughly 2,200 years to as early as 4,800 years B.P. (Livingston, 1984; Taylor, 1990). Similar evidence of ecological change to the south in the highlands of Rwanda around this general period has led researchers to posit that the forest disturbance in southwest Uganda was most likely the result of human endeavor and probably for the purposes of cultivation (Hamilton et al., 1986, 1989; Taylor, 1990). Additionally, there is linguistic and archaeological evidence to indicate that the establishment and expansion of, and interactions between, horticultural and pastoral production systems precipitated much of the environmental change (i.e., deforestation) in the highland region at least as far back as 1,000 years B.P.--if not much earlier (Schoenbrun, 1990; 1993). 197 198 Although it is difficult to precisely determine all of the causes, the various scientific data suggest that human modification of the environment has been a constant feature of the highland region for at least 1 ,000 years. It also has been difficult to ascertain the specific rate and extent of ecological transformation over the last millennium, or whether the highland agricultural and pastoral production systems experienced fluctuations or even collapse during earlier times. It is speculated, however, that ecological change, particularly in the form of forest and wetland clearance, has occurred at an extremely rapid rate over the last 100 years (e.g., Hamilton, 1969, 1984; Karani, 1982; Ngologoza, 1969). Land Use Change The various environmental and cultural-historical records indicate that change- climatic, ecological, population and agricultural~has long been a predominant feature of the ecosystems in the highlands of southwest Uganda. Over the past 50 to 100 years, smallholders have expanded cultivation into increasingly marginal environments, including areas of steep-slopes and poor soils, as well as wetland protected forest areas. 1 Correspondingly, many colonial and government officials and outside experts have warned for decades that population pressures and land shortages would imminently induce a collapse of the ecological and agricultural systems in the highlands. 1 Encroachment into Mgahinga and Bwindi Impenetrable National Parks, and Mafuga and Echuya Forest Reserves, all located in southwestern Uganda, continue to pose major conservation management concerns (e.g., CARE, 1994; Cunningham, 1992; Cunningham etal., 1993). 199 Although the highland region is currently characterized by land use pressures and beset by a variety of agricultural and resource management problems, the predicted decline or collapse of the production systems has not occurred. Furthermore, while smallholders have indisputably expanded agricultural production in recent decades, largely in response to population pressures and concomitant land shortages, there is little known about how they have intensified agricultural production or modified other aspects of their agricultural and land use systems in response to these problems. A number of smallholder responses to land shortages and other resource management problems are explored here. Recent Vegetational Change Throughout much of Uganda, and particularly in the southwestern highlands, there has been widespread deforestation and wetland destruction in recent decades- primarily a result of the quest for agricultural land (e.g., Hamilton, 1984; Karani, 1982; Struhsaker, 1987; Tukahirwa and Veit, 1992). For example, a recent biomass survey of the greater Kabale Town area, (including the Kicumbi Parish study site) indicates that "subsistence farming dominates the entire land use activity, occupying about 75% of the total area" (GOU, 1993b: 162). The extensive forest clearance has implications for the availability of fuelwood and charcoal resources throughout the highlands. In Kabale District, the vast majority of households-98. 5% --reported firewood as their primary fuel source (GOU, 1992a), while in Kisoro District, 98.8% of households surveyed reported firewood as the primary fuel source (GOU, 1992b). 200 A survey of the biomass in Kabale District revealed that fuelwood is in short supply, and that many people often have to travel considerable distances to procure it (GOU, 1993b). While some charcoal is produced locally (e.g., often from Black Wattle (Acacia mearnsiij), it is prohibitively expensive for most smallholders, and moreover it's generally marketed for cash (GOU, 1993b). Women, and to a lesser degree children, are responsible for collecting fuelwood. For many smallholders, many types of biomass that can be burned are used as fuel, including stover from cereal crops and other agricultural detritus. Many women collect stover, plant materials, weeds and brush to burn during trips to and from their plots. Consequently, much of the biomass/organic materials that could be composted or re- incorporated into the soil are often used as fuel. Contrary to popular opinion, many smallholders in the highlands plant trees on their landholdings. For example, in Kabale District, a multi-village transect walk conducted in 1995 (Lindblade et al., 1996) actually revealed increased tree cover (9.9% of land area) over the same transect-area in 1945 (4.2% land area) (Purseglove, 1946). In Table 8.1, the tree planting practices of smallholders in the three study sites are described. In the highlands, trees are planted as boundary markers, and for fuelwood and timber resources; many of the trees are planted in woodlots near the homestead. Trees planted across the agricultural landscape are predominated by exotic species such as various eucalyptus, acacia mearnsi (Black Wattle), and cupressus lusitania (GOU, 1993b; Peden et al., 1991). The eucalyptus, introduced during the 201 Table 8.1. Tree Planting Practices. Smallholder Responses Tree Planting 1. Kicumbi 2. Nyarurambi 3 rVfuramha 1 . Do you plant trees? Yes: 12 (19%) 60 (65%) 63 (39%) No: 50 (81%) 32 (35%) 99 (61%) 2. What are the benefits of planting trees? a. Demarcation 2 5 39 b. Reduce erosion 7 30 15 c. Keauce lanosnaes 1 2 2 d. Produce fodder/firewood 2 22 5 3. What are the drawbacks of planting trees? a. Uncontrolled Grazing 1 9 39 b. Labor Demanding 3 7 3 c. Easily Destroyed 7 11 0 d. Thievery 0 24 3 e. Don't Know 1 0 0 colonial period, are raised in woodlots, or scattered around homesteads, for use as fuelwood, charcoal, timber and building poles, or fencing materials. 2 The acacia were originally introduced for tannin production, but are now used for charcoal making or fencing materials. Other trees of minor popularity included the euphorbia species (popular as fences or boundary markers), markhamia platycatyx and ficus natalensis (ICRAF, 1987, 1993; Okorio et al., 1988). The fast-growing and multi- purpose eucalyptus are the most popular tree species popular among smallholders, and 2 The eucalyptus species are very popular among smallholders due to their rapid growth habits, ability to grow in a diversity of environments-including areas of poor soil and wetlands-and multiple uses. Ironically, they remain an unpopular tree species among many agroforesters and development workers, and are not regularly incorporated into local resource management programs. 202 while some native species such as euphorbia remain common, many of the indigenous trees are rapidly disappearing. Tree planting at all three study sites is generally controlled by men, although women participate in their management as well. Many men plant trees as a cash crop. Men have complete ownership rights and control of the land, and local land tenure and cultural traditions bar many women from planting trees or hedgerows without the prior approval of a husband or other male family member. This is another example of the gender-based constraints to improved resource management in the highlands that confront women, who are responsible for managing land resources on a daily basis yet are often unable individually to address many resource management problems. The majority of the wetlands in the highlands have also been exploited in recent decades for the purposes of smallholder cultivation, vegetable production and dairy farms. Swamp reclamation started as far back as 1929, when the colonial administration encouraged the drainage of wetlands around Kabale town for the purposes of vegetable cultivation and tree plots (Gibb and Partners, 1967; Ngologoza, 1969; Martin, 1945b). Presently in the former wetlands west and south of Kabale town, most of the reclaimed swamps have been converted to dairy farm pastures, although some smallholder cultivation does exist (e.g., Murindwa, 1991). In the more distant wetland valleys, there are fewer or no dairy operations, more smallholder cultivation and some remaining native vegetation. For example in Kigeyo Swamp, which is partially located in Nyarurambi Parish, approximately 25% - 50% 203 of the area remains under native vegetation (e.g., papyrus-wetland complex). In both Kabale and Kisoro Districts, many smallholders have reported that the widespread cultivation of the wetlands have resulted in the loss of biomass for fodder and compost, as well as other plant materials traditionally used for the production of household wares such as weirs, baskets and mats. During the course of the fieldwork many smallholders reported that annual rainfall totals have decreased in recent decades, although the meteorological records reveal no remarkable trends (re: Appendix B). Smallholders have also reported that the morning mists are not as common or heavy as they once were, and that temperatures are generally warmer. While their observations cannot be confirmed, the phenomena they describe might be the result of overall climatic change in the region, and/or the extensive deforestation and the drainage of the wetland environments. The emergence and rising numbers of malaria cases in the highlands (a former non-malarial zone) also suggests either that the climate is changing, and/or that the transformation of the vegetation in the region has had other wide ranging effects. Given the extensive human-modification of the environment, it is conceivable that climatic change has occurred over the last 100-odd years in the highlands; there is, however, little known about former or current levels of micro-climatic variation in the region. Land Fragmentation In the highlands of southwestern Uganda, the intensive agricultural systems have been characterized by widespread land fragmentation since colonial times. 204 Indeed, overpopulation and land fragmentation are often described as two of the primary causal factors responsible for environmental degradation in the region (e.g., Byagagaire and Lawrence, 1957; Kururagire, 1969; Purseglove, 1950, 1951; Turyagenda, 1964). Given the dispersed nature of the plots and the various forms of tenure claims on them, it is difficult to determine the actual size of smallholder landholdings. Recent surveys by the Government of Uganda and the World Bank have provided some indication of the small size of landholdings in the highlands, but the findings are somewhat varied. In one study, 62% of the rural households in the regions reported landholdings of less than one hectare each, while 23% had somewhere between 1 and 2 Ha each. In total, 85% of the rural households produce crops and raise livestock on holdings that average less than two hectares each (GOU, 1994b: 19). A second study reported that 99% of a sample of 180 households in Kabale cultivated less than one hectare (World Bank, 1992b), while a third study in Kabale reported cultivable cropland of 2.4 Ha per household (Grisley and Mwesigwa, 1994). Finally, in Kabale District, the per capita available cultivable area has been variably reported as 0.39 Ha (GOU, 1994b) and 0.286 Ha (World Bank, 1993). While there is some variation in the studies, the estimates of per capita cultivable land area represent some of the lowest in the country. While smallholders in the highlands generally cultivated a small land area in the form of scattered holdings, they did own the majority of their plots. During focus groups discussions with smallholders in the study sites, the majority reported owning 205 most of their plots. While an active land rental market has been described for Kabale District (e.g., Grisley and Mwesigwa, 1994), the great majority of smallholders in the 3 study sites reported very little rental of agricultural plots. As Table 8.2 indicates, only 4 smallholders in Kicumbi and Muramba Parishes, and 7 in Nyarurambi Parish reported renting plots; of these renters, the majority rented only one or two plots. Similarly, only a small percentage of smallholders in each site reported borrowing plots. Table 8.2. Renting, Leasing or Borrowing of Agricultural Plots Smallholder Responses Renting, Leasing or Borrowing 1. Kicumbi 2. Nyarurambi 3. Muramba How many plots do you rent or lease? 0. 63 87 161 1. 2 2 2 2. 1 3 2 3. 1 4. 1 5. 1 How many plots do you borrow? 0. 59 87 154 1. 5 2 10 1 2. 2 3 3. 1 1 4. 1 5. 3 Even in the study that described the active rental market, 88% of households surveyed (n= 85) owned all of their plots. 206 There are also variations in reports on the number of plots per households. One study reported 6.7 plots per household in Kabale (World Bank, 1992), while another reported an average of 7.0 plots per household (Grisley and Mwesigwa, 1994). In Table 8.3, the number of plots per smallholder in Kicumbi, Nyarurambi and Muramba Parishes are presented. As the data in Table 8.3 reveal, fragmented landholdings are also characteristic of the three study sites. There was an average of 5.6 plots per person in Kicumbi, 9.35 plots in Nyarurambi and 3.0 plots in Muramba. The average number of plots ranged from 3.0 plots/person in the most densely populated parish (re: Muramba), to 5.6 plots/person in the parish with the medium population density (re: Kicumbi), to 9.35 plots/person in the parish of lowest population density. (The range in the number of plots people owned was also greatest in Nyarurambi, followed by Kicumbi and finally Muramba.) Counter-intuitively, the number of landless followed the opposite trend, with the greatest number of landless people in Nyarurambi and the 4 least number of landless people in Muramba. One explanation for this latter trend might be the availability of alternative employment options for people in the less densely populated parishes. For example, Kicumbi Parish is close (12 km) to employment opportunities in Kabale Town, and Nyarurambi is bisected by a major transportation route and supports a number of 4 It is not entirely clear how landless people pursue their livelihoods, although some individuals work as agricultural laborers while others are involved in trade and marketing. 207 Table 8.3. Fragmented Landholdings. Smallholders with "x" Number of Plots Nyarurambi. 3. Muramba. Number of Plots 1. Kicumbi. 2. 0. 11 18 6 1. 2 1 20 2. 2 - 45 3. 10 2 46 4. 11 2 19 5. 6 4 16 6. 2 6 4 7. 2 4 2 8. 3 7 3 9- 4 6 " 10. 3 10 4 11. 2 2 12. 3 6 13. 1 3 14. 2 - 15. - 7 16. - 1 17. 1 2 18. 2 3 19. 20. 4 21. 22. 3 23. 24. - 25. 1 26. 27. 1 28. 30. 1 Average plots/ Average plots/ Average plots/ person = 5.6 person = 9.35 person = 3.0 Median plots/ Median plots/ Median plots/ person = 4 person = 9 person = 3.0 208 small trading centers and shops. Muramba, on the other hand, is the most remote of the three parishes and despite the proximity to Kisoro Town, there are few trading or other employment opportunities available to people. (While Muramba had been a thriving trading center in the 1960's, the economic opportunities decreased drastically with the onset of the political and economic misfortunes in Uganda in the 1970's and 1980's, and the more recent troubles in neighboring Rwanda and Zaire.) Land fragmentation, also known as plot or field scattering, is the pattern of land ownership consists of numerous discrete parcels, often scattered over a wide area (Bentley, 1987). The critics of land fragmentation assume that fragmentation is detrimental to farming in that it lowers land productivity, and that it is costly because smallholders waste a lot of time moving from plot to plot. Studies of land fragmentation in Ghana, Rwanda and Kenya, however, suggest that it does not have an adverse affect on land productivity (Blarel et al., 1992; Place and Hazell, 1993). Small farms and fragmented landholdings are often efficient and agriculturally productive (Ruthenberg, 1980; Netting, 1993). Land fragmentation or field scattering has also been described as effective agricultural risk management, where smallholders can capture micro-environmental (e.g., soil) variations and disperse losses (Bentley, 1987; Goland, 1993). Land fragmentation is a correlative of intensive agriculture and individual tenure, and is mischaracterized as a significant problem (Bentley, 1987; Igbozurike, 1970). In fact, a recent study in Machakos, Kenya found land fragmentation to be one of the factors that induced agricultural intensification and eventually improved agricultural production and resource management (Mortimore 209 and Tiffen, 1994). These studies also cast doubt on the need for ambitious land registration, titling and/or consolidation programs that are often suggested as a means of mitigating environmental problems. In Kabale and Kisoro Districts, smallholders noted that their landholdings have become increasingly fragmented and dispersed over the last 4 to 5 decades. The growing population and the inheritance patterns, where landholdings are divided among sons, are primarily responsible for this phenomenon. Smallholders also recognized that the fragmented nature of their landholdings can impede efforts to efficiently manage their crops and land resources, i.e., a number of smallholders did complain of the great distance to some of their plots. They also noted, however, that multiple and scattered plots provide opportunities to diversify crop production and mitigate risk. The majority of smallholders reported cultivating plots scattered across the hillsides, and in Kicumbi and Nyarurambi many smallholders also cultivated land in the valley bottoms. 5 Table 8.4 presents smallholder assessments of the advantages and disadvantages of multiple plot-holdings. Over 94% of respondents in each parish reported that there were advantages to cultivating more than one plot. In terms of advantages, the majority of respondents were most appreciative of the different soil types of multiple holdings. The multitude of soil types generally allowed for the diversification of crop production and reduce the chances that the smallholder would 5 For example, in a recent survey in Kabale, 76% of households reported owning land in both valley bottoms and hillsides (Grisley and Mwesigwa, 1994). 210 Table 8.4. Advantages and Disadvantages of Multiple Plots Smallholder Responses Advantages and Disadvantages 1. Kicumbi 2. Nyarurambi 3. Muramba Are there advantages to cultivating more than one plot? Yes. 61 (98%) 92 (99%) 146 (94%) No. 1 (2%) 1 (1%) 9 (6%) What are the advantages ? 1. Soil variations Yes: 56 (89%) 66 (71%) 116 (79%) No: 7 (11%) 27 (29%) 31 (21%) 2. Micro-climate variations 32 (51%) 56 (60%) 59 (40%) 31 (49%) 37 (40%) 88 (60%) 33 (52%) 63 (68%) 103 (70%) (e.g., heavy rains, hail) 30(48%) 30 (32%) 44 (30%) 4. Minimize losses to livestock 2 (3%) 19 (20%) 42 (29%) 1 AC 1 fjf \ 61 (97%) 74 (80%) 105 (71 %) 1 /I (ff \ "20 /OO QL \ 5. Minimize losses to thievery I (3 70) 3U \5Zto) 51 \LLto) 61 (97%) 63 (68%) 115 (78%) 6. Average (maximize) yields U U 63 (100%) 93 (100%) 89 (60%) Are there disadvantages to cultivating multiple plots? Yes. 36 (57%) 53 (58%) 113 (82%) No. 27 (43%) 39 (42%) 25 (18%) What are the disadvantages ? 1 . Losses to livestock Yes: 30 (81%) 48 (90%) 80 (69%) *7 / 4 r\ ft* \ O £. 1 fif \ No: 7 (19%) 5 (10%) 36 (31 %) 2. Losses to thievery 26 (70%) 47 (89%) 90 (77%) 11 (30%) 6 (11%) 26 (23%) 3. Shortage of labor 19 (51%) 14 (26%) 33 (28%) 18 (49%) 39 (74%) 83 (72%) 4. Distance to plot 15 (40%) 17 (32%) 27 (23%) 22 (60%) 36 (68%) 89 (77%) 5. Shortage of inputs 6 (16%) 8 (15%) 22 (19%) 31 (84%) 45 (85%) 94 (81%) 211 Orucucu). To a lesser degree, multiple be limited to cultivating poor soil types (e.g., micro-climatic (e.g., rainfall, dew) plots also meant that smallholders could capture smallholders the means to cultivate a variations. While multiple plots provided also provided a means of minimizing multitude of soil and micro-climatic types, they plots if a smallholder with multiple losses due to climatic hazards. For example, good particular plot of bush beans, chances are were to experience a hailstorm on a equally damaged. the other plots would not all be (Kicumbi - 57%, Nyarurambi - 58% A smaller percentage of the respondents were disadvantages of cultivating and Muramba - 82%) also reported that there sites cited disadvantages to cultivating multiple plots. Smallholders in all three study to closely manage the plots and more than one plot, and they included the inability well as the shortage of labor and distance mitigate losses to livestock and thieves, as to plots. expressed a fear of Throughout focus group meetings, however, smallholders consolidate landholdings in the any attempts by the government or other outsiders to two basic reasons: 1) they didn't believe region. They were wary of consolidation for they of land after transfers took place, and 2) that they would receive an equal amount poor or more distant land. The vast feared that they would be assigned parcels of "status quo" landholding system, majority of smallholders stated a preference for the 6 over which they at least had some control. exchange plots with one 6 Smallholders indicated an interest in and a willingness to or the overall nature of the local another; however, the degree that they exchange plots, land market in the region was not examined in this study. 212 The opportunity to diversify crop production by cultivating multiple soil types and to minimize crop loss due to location-specific nature of many climatic hazards and thievery losses are hallmarks of the smallholder's ability to manage for agricultural risk in systems characterized by fragmented landholdings. Fallow Change and Intercroppin g Field fallowing, where an agricultural plot is allowed to rest for one or more cropping seasons, is a soil conservation and soil enhancement technology used in a variety of arable agricultural systems around the world. Fallow-field systems occur where one arable crop follows another, and where established fields are clearly separated from one another (Ruthenberg, 1980). Under the assumption of static production technologies, increasing population densities can either promote or discourage the use of fallow technologies to manage soil fertility. Fallow practices may be reduced or even eliminated to address demands for food, or conversely, as soil fertility deteriorates from intensive use, households may be forced to resort to more, not less, fallowing to maintain productive soils (Grisley and Mwesigwa, 1994: 82). The highland landscape of southwest Uganda is characterized by a patchwork of plots~a pattern representative of the dense populations, intensive agricultural production, traditional inheritance practices and colonial soil conservation programs common to the region. In recent years, some observers have suggested that in many areas of the highlands, there has been a shift in agricultural practices from field- fallowing systems towards continuous cultivation (e.g., Webb et al., 1994). It is > 213 posited that land shortages have precipitated the decline of the traditional fallow system, both in terms of frequency of fallow and length of fallow period, and that this trend is leading to declining soil fertility, and in some areas to soil erosion and increased weed problems. In an attempt to accommodate land shortages, many smallholders have expanded cultivation into increasingly marginal environments, including areas of steep slopes and poor soils, as well as wetlands and protected forest zones. There is also evidence that smallholders continue to employ the fallow system to manage soil erosion and soil fertility. As the population densities have increased in the southwestern highlands, some aspects of the fallow system have changed. For example, many smallholders have stated that fallow periods have grown shorter in length over the last few decades primarily because of increasing land shortages. However, smallholders have also not abandoned the fallow practices entirely, and grass-weed fallows continue to be employed by many individuals. While many smallholders no longer fallow for extended periods, many do use the practice for one to two seasons to maintain fertility and reduce topsoil erosion. In Table 8.5, smallholders in the three study sites were questioned as to their fallow practices. Of the respondents, 33 in Kicumbi, 54 in Nyarurambi and 7 in Muramba reported employing fallow at the time of the survey. The vast majority ( 90%) reported using fallow to restore soil fertility, while virtually none of the respondents (with the exception of two individuals in Nyarurambi) reported that plots 214 were fallow due to labor shortages. There have also been a number of studies conducted in recent years that corroborate much of what smallholders indicate. Table 8.5. Smallholder Fallow Practices Smallholder Responses Fallow Practices 1. Kicumbi 2. Nyarurambi 3. Muramba Are any of your plots in fallow? Yes: 33 54 7 No: 34 39 158 Why are they in fallow? 1. To restore soil fertility 32 49 6 2. Lack of labor to cultivate 0 2 0 In a study conducted in Kabale District under the auspices of the World Bank, 14% of cultivated area was under fallow (World Bank, 1992b), while in another study in the region, 76% of farms reported some fallowing, and 26% of all cropland was reported fallowed (Grisley and Mwesigwa, 1994). The latter study found that even as population densities increased, households engaged in more land- (and labor-) intensive technologies, especially intercropping, while continuing to rely on fallows to manage soil erosion and soil fertility (Grisley and Mwesigwa, 1994: 82). Another study in Kabale has also found, contrary to popular expectations, that fallowing is a widespread practice among many smallholders (Lindblade et al., 1996). Compared with a similar study conducted in 1945 (Purseglove, 1946), this study found an overall increase in fallow: 19.7% of the land was under fallow in 1945, while 33.8% of the land was fallowed in 1995. Following a typical hillside profile, 215 the study described the ratio of cultivation to fallow as follows: one-to-one on the ridge summits and shoulders; more cultivation than fallow on the footslopes and backslopes; and one-to-one on the toe slopes and swamp areas. Most fallow occurs for less than a year (i.e., one season), while some plots are fallowed for one-to-two years. Fallow lengths also generally increased as one moved up the hillside (Lindbladeetal., 1996). However, as the data in Table 8.5 indicate, the smallholders surveyed for this study also revealed a reduction in fallow with increased population density. While the difference in fallow practice among respondents in Nyarurambi (59%, with the lowest population density) and Kicumbi (49% - with the medium population density) is slight, the use of fallow drops off considerably among respondents in Muramba (4% - with the highest population density). This data implies that there might actually be a point where population density and land shortage is so great that fallowing is no longer a viable land management option. In addition to increased fallowing, the study also found that intercropping had increased among smallholders. Double-cropping (i.e., two or more crops interplanted) was found to have increased by about 10% since 1945, while it is also hypothesized that triple and quadruple cropping have also become more common (Lindblade et al., 1996). Smallholders chose to intensify land use through intercropping rather than further reduce or eliminate fallow practices for a number of reasons: a) given the land shortages, a variety of food crops can be produced from one plot, b) it reduces risk of food shortages in the case of a particular crop failure, 216 c) it establishes a good ground cover, thereby reducing weed competition and erosion hazards, and d) it makes efficient use of a limited labor supply. On the negative side, smallholders also noted that intercropping generally resulted in lower yields per crop. At the present time, a number of technologies that improve intercropping production and enhance fallows (i.e., leguminous intercrops, managed fallows with green manures) might be explored. Intercropping technologies that allow for a greater degree of crop species and or varieties, input selection and allocation, and management methods could allow household's to continue current SRM strategies for the immediate future. A number of alternative technologies might also be investigated for the long term, although they each have drawbacks. They include fertilizers (which are generally costly and unavailable), terraces (which are generally unpopular and labor-intensive) and agroforestry technologies (which invoke gender issues related to management of trees and land tenure conflicts). Weed Management Another aspect of agricultural production and fallow systems is weed management. A large number of smallholders in each of the study sites stated that weed infestations were becoming increasingly problematic. It is difficult to discern, however, whether the increased weed problem is due to decreasing soil fertility or 7 management problems. In Kabale District, the perennial weeds which are a problem include Cyperus rotundus, Cyperus esculentus, Cynodon dactylon, Imperata cylindrica 7 For example, although the species is not present in the highlands, Striga hermonthica is regarded as a symptom of deteriorating soil fertility in northern Uganda as well as within the scientific community (e.g., Albert and Runge-Metzger, 1995)). 217 and Digitaria abyssinnica; the latter was found to be the most serious weed problem (Webb et al., 1994). A number of annual broadleaf species were also important, including Bidens pilosa, G. parviflora and Commelina spp. (mainly C.benghalensis) (Webb et al., 1994). In the study sites of Kicumbi, Nyarurambi and Muramba, smallholders-particularly the elderly, also identified the majority of these weed 8 species within their crop communities. A number of elderly smallholders also indicated that soil fertility could be evaluated using certain weed species as indicators (cf., Fairhead, 1990). The younger smallholders, however, were generally not as competent at identifying weeds, what their presence implied or their other uses. The majority of the weeding was conducted using manual methods, rather than mechanical or chemical techniques. (Few herbicides or other pesticides are available in the region.) With most crops, smallholders were aware of the need for early weeding. Nevertheless, many smallholders considered that they were unable to carry out most of their weeding in a timely manner, and their crop yields suffered as a result. Inadequate labor was perceived by most smallholders as the primary constraint for crop production, and even when labor could be hired it was generally in short supply. Labor was especially difficult to hire in the bottleneck periods of March-April, August and December. Peak labor demands for women were in March- April (digging, planting, weeding) and August (harvesting), while for men they were Elderly smallholders were able to identify over 50 weed species at each of the study sites. Not only were the "weeds" used to indicate the fertility status of a soil, but they were also used as food during famine periods, livestock feed, human and livestock medicines, and handicraft materials. 218 August (i.e., harvesting) and December-January (land preparation). The majority of the households hired labor and, as with family labor, much of that was provided by women. Weed management is perceived by many smallholders to be an increasing problem in all three study sites. As with many aspects of agricultural production in the highlands, the gender-division of labor also affects weeding. The problem is exacerbated by the fact that it remains primarily a women's task, and crop yields have declined partly as a result of untimely and overall inadequate weed management. Weeding takes considerable time and hard work and has a social cost: in most areas, this burden and cost is borne by women (e.g., Webb et al., 1994). Croppin g System Changes In response to changing environmental, social and economic conditions, and in an attempting to maintain or increase agricultural productivity, smallholders have continually modified their cropping practices and profiles over the last four to five 9 decades. For example, as described above, intercropping has become increasingly common in the highlands, and within the intercrops smallholders also continue to 10 experiment with their crop mixtures, as well as with specific crop varieties. Furthermore, the cultivation of "traditional" crops such as peas and millet have 9 A number of excellent overviews of agricultural change in Uganda are available (e.g., Amann et al., 1972; Masefield, 1962; McMaster, 1966). 10 Smallholders in each parish, again particularly the elderly, were able to list and identify multiple varieties of many crops. For example, some women were able to identify over twenty varieties of local bush beans. 219 decreased in recent decades due to the high labor demands and poor yields, while the cultivation of more recently introduced crops like Irish potatoes and wheat have increased primarily due to increased market opportunities. On the whole, cropping profile modifications have occurred as a result of a variety of influences, including population pressures, social interactions, declining yields, changing cultural food preferences and market influences. Table 8.6. Cropping Profile Changes General Responses (Non-Ranked) From Focus Group Meetings Food and Cash Crops 1. Kicumbi 2. Nyarurambi 3. Muramba Twenty-Five years ago, which crops were most Field Peas Field Peas Field Peas important for: Millet Millet Millet Sorghum Sorghum Sorghum Food? Bush Beans Bush Beans Bush Beans Sweet Potatoes Cash Income? Bush Beans Sorghum Millet Field Peas Bush Beans Bush Beans Tobacco Millet Field Peas Coffee Tobacco Tobacco Bananas Trees Today, which crops are most important for: Bush Beans Bush Beans Sorghum Sweet Potatoes Sweet Potatoes Maize Food? Sorghum Irish Potatoes Climbing Beans Maize Maize Irish Potatoes Irish Potatoes Sorghum Cash Income? Sorghum Sorghum Climbing Beans Bush Beans Bush Beans Maize Sweet Potatoes Vegetables Bananas Sweet Potatoes Tobacco 220 Among smallholders and the DAO staff and outside experts, there is general agreement that production levels for a variety of crops (e.g., millet, peas) have declined in recent decades. It is difficult to ascertain, however, whether changes in the production of certain crops are a result of declines in soil fertility, crop disease, cultural choice, or a combination of factors. Smallholders also acknowledged that land shortages often compelled them to plant crops on poorly-suited, non-preferred soils. For example, they might plant culturally important crops like sorghum (for beer production), or market-oriented cash-crops like Irish potatoes on an unsuitable soil if the soil type they preferred was unavailable. In a variety of mixed-sex focus groups meetings, elderly smallholders (age > 60 years) were asked about various aspects of crop change in their respective parishes. The general trends that smallholders are reported are presented in Table 8.6. In each of the parishes, smallholders reported a decline in millet and field peas for cash and especially food crop production. The participants stated that the declines were primarily due to a decrease in yields for both crops, and the labor intensive nature of millet production, which was difficult to weed and traditionally a "women's" crop, so men provided no assistance at all. Other causes for decline which were cited less frequently were changing food preferences and land shortages. Additionally, smallholders observed that pumpkins (re: squash, gourds) and cash crops such as pyrethrum and coffee had also declined due to changing food preferences for the former and poor market opportunities for the latter. 221 Presently, a number of new crops have seen expanded cultivation within the highland agricultural systems. In terms of food crops, both maize and tubers, i.e., Irish potatoes and sweet potatoes, have gained in prominence. Their dramatic increases have been due to changing food preferences and increased cultivation of the wetlands (and hence, ease of cultivation of sweet potatoes). 11 However, one of the primary reasons that Irish potatoes have expanded in production is due to the increased market opportunities. The Irish potatoes are one of the few crops in the highlands that are consistently oriented towards market production, and buyers come directly from Kampala to purchase the harvests. Sorghum remains a staple cash crop and a famine food crop for smallholders throughout the highlands. It is primarily sold locally and used to produce sorghum beer (i.e., muramba) and a non-alcoholic drink (i.e., obushera). Bush beans and climbing beans also remain viable cash crop for many smallholders, with traditional 12 markets in the highland region, Kampala, Rwanda and Zaire. In each of the parishes, smallholders report a general trend towards a decrease in the production of pulses and an increase in the production of tubers. While this trend has certain economic advantages, it also implies certain nutritional and dietary disadvantages if it continues. Many households earn money from crop sales, mainly sorghum and Irish 11 In Kicumbi and Nyarurambi Parishes, most of the maize and tubers are grown in the wetlands areas. 12 With bean production, smallholders generally endure less risk--at least compared with tubers-as the harvest stores relatively well and it can be consumed within the household. Other crops which have seen expanded production, primarily with a market orientation, include wheat and vegetables (e.g., cabbages). 222 potatoes. While men and women market their own crops, men generally control all of the money produced-underscoring another aspect of the gender biases inherent in the agricultural system. In Muramba Parish, tubers have not increased in production to any great degree. On the other hand, the climbing bean has been widely adopted in recent years-even though it requires staking material and is thus more labor-intensive than bush beans. Nonetheless, many smallholders make the effort to find staking materials, with many people intercropping with local shrubs and cultivating elephant grass to produce stakes. Smallholders in both Kicumbi and Nyarurambi parishes have not, however, adopted many climbing bean varieties, claiming that they are susceptible to bird-losses and that staking materials are unavailable. It appears that the high population densities and land shortages in Muramba Parish have induced smallholders to make investments in the intensification of bean production through the adoption of high yielding climbing beans, despite their labor and material costs. This may be interpreted as an illustration of Boserup's model of increased labor intensity to acieve greater output with higher population densities. Summary Highland agricultural systems are dynamic and subject to continual adjustment, regardless of the level of social organization, ecological knowledge and adaptations available (e.g., Jodha, 1990; Jodha et al., 1992). In the highlands of southwest Uganda, stresses on the natural resource base, environmental (e.g., climatic) or human-induced, have been common features of the highland agricultural production 223 systems and often have been major catalysts for change and adaptation. Population pressures have been one of the primary forces influencing land use in the region since the turn of the century, but they cannot be regarded apart from other social, economic and political factors. Population growth has undoubtedly precipitated widespread environmental change in the highlands of southwest Uganda, but the human modification of the landscape for agricultural purposes has often been unjustifiably and simplistically interpreted as environmental degradation. Despite predictions of the collapse of the highland agricultural systems and the inevitable food shortages, the smallholder production systems have endured. The agricultural system has not evolved undamaged, however, and there has been a widespread loss of natural forest and wetland ecosystems. Smallholders have also undoubtedly employed unsustainable land management practices and expanded cultivation into marginal environments~but they have also adapted to land use pressures and other varied influences in ways heretofore not well understood. To date, there has been little documentation of the adaptive practices or strategies that smallholders have employed to address local land use pressures, as well as other influences internal and external to the highland production systems. This study has identified a number of "positive" smallholder responses to land use pressures and other forces external to the region. They include a modest increase in tree planting, the utilization of a system of fragmented landholdings to manage for agricultural risk, an increase in the use of short term fallows in the less densely 224 populated parishes--an adaptation that has existed for some time in the more densely populated Muramba Parish, the expansion of intercropping practices, and the adaptation of cropping profiles to meet changing food preferences, capture cash crop market opportunities and intensify food production. In Muramba Parish, there has also been an expansion of production of labor-intensive crops such as climbing beans. Smallholder adoption of specific soil and water conservation measures were also outlined in previous chapters. Economic and market forces have also influenced the cropping choices of smallholders, and they continue to wield an influence today. Some of the economic and political factors that have impacted on land use in the region have been both external and intermittent in nature, and the highlands have periodically experienced integration into and disengagement from the larger socio-economic system of Uganda and East Africa (e.g., Rwanda and Zaire). Agricultural production and land use have been affected by both the incorporation of the region into the national market system (e.g., vegetable production schemes and cultivation of wetlands) and subsequent separation from that larger system (e.g., income loss and conversion of wetlands to large scale dairy farms) (cf., Chazan, 1988; Chazan and Shaw, 1988). Labor constraints in some aspects of production (e.g., weed management) have also had negative effects, as have gender biases throughout the highland agricultural systems- which have increased demands on women's labor, but continued to restrict their access to resources. 225 While not all smallholders have responded to influences in the region in a useful manner, many have demonstrated a good deal of flexibility and adaptability. Smallholders in the highlands have selected and adapted new ideas from multiple sources, and the technological change that has occurred has been largely an endogenous process (e.g., Mortimore and Tiffen, 1994). More specifically, it has been a process whereby smallholders have selected a number of exogenous elements, including new crops and land management techniques, to incorporate into their agricultural system. Even where smallholders were initially coerced to adopt a crop or land management practice (e.g., grass bunds), they continued to use them only after they determined that they provided an advantage over previous crops or practices. In the following chapter, Chapter Nine, a summary of the research results and the conclusions of the study are presented. The recommendations for future research in the highlands of southwest Uganda are also discussed. CHAPTER 9 SUMMARY AND CONCLUSIONS This study has attempted to expand the understanding of the rich and diverse nature of ecological and agricultural knowledge among smallholders in general, and among those smallholders in the highlands of southwest Uganda in particular. It has also attempted to elucidate some of the forms of interactions between dense populations, agricultural production, and soil resource management, and the processes that influence land use change in highland environments. Summary Since the turn of the century, numerous examples of land degradation and food shortages have been reported and documented for East Africa (e.g., Anderson, 1984; Stockdale, 1937; Wayland and Brasnett, 1938). Soil erosion and declining soil productivity continue to be cited as major threats to sustainable agricultural production in the region by scientists and smallholders alike (e.g., Lai, 1987, 1988a; Peden and Kakuru, 1993). Many of the current efforts to address declining agricultural productivity and to mitigate environmental degradation in the region are warranted, but just as the western popular media's numerous accounts of famine have resulted in the perception of food shortages for all of Sub-Saharan Africa, reports of environmental crises have also been generalized for large regions of the continent. 226 227 The propensity to universalize environmental problems in part can be traced to the media's fixation on crises as opposed to successes, and to a widespread reliance on macro-level data to diagnose and characterize environmental problems. Soil and land degradation undeniably occurs throughout East Africa, but the type and severity of many environmental problems are over-generalized for many regions and the resulting "homogenization" often overshadows the micro-level variations in the environmental problems, as well as the differing responses of smallholders to them. The region is most accurately characterized by a variety of environmental management problems that occur at varying degrees of frequency, magnitude and scale and that are addressed by the state, institutions, projects and individuals with variable degrees of success. In southwest Uganda, the Districts of Kabale and Kisoro exemplify a densely populated highland environment with a history of land use pressures and intensive agricultural production systems. It has been posited that both physiography and intense land use pressures leave soils in the region highly susceptible to degradation, and a common prognosis is that current land-use systems will not be capable of providing the growing population with food (e.g., Ngabirano, 1993; Grosjean and Messerli, 1988). Contrary to popular perceptions, this investigation into soil resource management and land use change has actually revealed a diversity of agroecological knowledge and land management practices among smallholders in the highlands. It has also identified a variety of internal and external factors that have influenced land 228 use change in the region over the last fifty years, as well as the varied, adaptive responses of smallholders to those influences. Smallholder Soil Knowledge In the highlands of southwest Uganda, despite the widespread claims of environmental degradation, neither the various causes of degradation, nor the differing abilities of smallholders to manage for them have been well understood. There is little known about the soil and agricultural knowledge and management practices of local smallholders, or the adaptive strategies they utilize in response to various stresses. During the course of this study, smallholders in Kabale and Kisoro Districts indicated their use of a soil classification system that was practical and functional in nature, and oriented towards the agricultural management of the local soils. Not only was the local soil classification system complex, but it reflected local environmental conditions and smallholders' agricultural and land management concerns. The study also demonstrated that smallholders possessed a basic understanding of the fertility and erodibility status of local soils, as well as an awareness of the relationship between soil erosion, fertility loss and declining crop production. The majority of smallholders in the highlands are capable and conscientious soil resource managers, although they also possess differing management motivations, abilities and constraints. While the knowledge of soils varied among smallholders within and across the sites, the level of general knowledge among smallholders is sufficient to allow for discussion and the exchange of ideas concerning improved agricultural production and resource management in the highlands. 229 Soil Degradation and Soil Resource Management Little is known about the scale, frequency, magnitude and causes of soil related problems in Kabale and Kisoro Districts, and this study has not attempted to establish that soils in the southwestern highlands of Uganda are less susceptible to erosion or more fertile than previously supposed. It does emphasize, however, that conventional scientific knowledge of soils in the region remains altogether incomplete, while what is "known" about soils has often been both incorrect and inappropriately generalized for the entire highland region. The widespread claims of environmental degradation in the highlands of southwest Uganda have, however, influenced land management in the region for more than fifty years. They have provided the justification for the compulsory land management programs, fostered an official disdain for smallholder management practices, and restricted the smallholders' options to address their agricultural and resource problems. The continued efforts to depict extensive soil and land degradation in the highlands, and to assess culpability for land management problems solely to smallholders have perpetuated these trends. Furthermore, the common depiction of smallholders as poor land managers has meant that they have seldom, if ever, participated in the development or refinement of district-level agricultural and resource management initiatives. Soil and land degradation certainly occur in the highlands, and not all smallholders employ the recommended, district-level soil conservation technologies. For many smallholders, soil resource management is not an end unto itself, but a means to the end of crop 230 production. They also have other problems to address that compete for their time and resources-e.g., water and fuel collection. Both smallholders and district officials indicate an interest in improved soil resource management and agricultural production in the highlands, but they also often have differing opinions about the severity of, various causes of, and solutions to, soil and land degradation in the region. For the DAO and various outside experts, smallholders in the highlands remain short-sighted perpetrators of deforestation and land degradation who could benefit from the strict enforcement of the various district- level resource management By-laws. The DAO message has been effectively inculcated among some smallholders, who correspondingly do blame their own mismanagement for soil degradation in the region. Other smallholders, however, cite the exacting and repressive nature of the SWC By-laws themselves, the lack of agricultural support, and of adequate inputs and markets, the poor economy, and other economic and social inequities for exacerbating their agricultural and resource management problems. Given the scarcity of scientific data for soils and the incomplete adoption of the soil and water conservation program, it is remarkable-although not atypical-that the DAO staff, outside experts and agricultural researchers have not previously investigated the agro-ecological knowledge and resource management experience and abilities of smallholders in the highland region. As this study has demonstrated, smallholders in the region utilize a variety of effective soil and land management practices that are both indigenous and exogenous in origin. The knowledge and 231 technical skills of smallholders are certainly not a panacea for soil and land management problems in the region, but they have the potential to both expand scientific technical knowledge and further empower the smallholders who own and use the knowledge (e.g., McCall, 1996; Pawluk et al., 1992). This last point is particularly important given that smallholders have rarely been provided the opportunity to participate in the dialogue on resource management and land use change in the highlands. Even if all the smallholders in the region were poor land managers, their exclusion from participation in regional resource management initiatives would not be warranted. That this exclusion has occurred for over 50 years is made all the more egregious given the wealth of knowledge and the diversity of innovative abilities smallholders have exhibited. Finally, the gender- dimension of agricultural production and land management cannot be overlooked: women undertake the vast majority of agricultural work, and correspondingly most of the soil management. Yet women are also marginalized: they have little control over or access to resources, and they have the least power in the local social structures. Consequently, increased agricultural production and improved land management will depend largely on the expanded empowerment of local women. Land Use Change and Smallholder Adaptations There are a variety of demographic, social, economic, political and historical influences that have impacted on land use and land use change in the highlands of southwest Uganda. 232 Colonial influences The role of the State cannot be overlooked in any study of land use change, and the colonial and post-colonial government policies in Uganda have been important in directly shaping land use changes in the southwestern highlands over the last five to six decades. These policies have included swamp reclamation, tree planting, agricultural (e.g., grazing, famine-granary) By-laws, and soil and water conservation By-laws. The coercive nature of many of the colonial interventions in the highlands also account in part for the unwillingness of smallholders to abide by many of the contemporary land management regulations or utilize recommended soil technologies. The prejudices towards potentially beneficial soil conservation measures underscores the persistence of negative colonial influences and the importance of understanding the historical dimensions of land use and resource management in the highland region. The history of the colonial soil conservation experience in southwestern Uganda also reveals the limitations of a "top-down" approach, where exogenous technologies were introduced without consultation with smallholders, and local soil knowledge and practices were largely ignored or disregarded. Unfortunately, many of the reasons behind the general failure of post-colonial, externally-developed and directed soil conservation programs have been overlooked and few lessons have been learned from the experience. At minimum, the failures and shortcomings of these experiences underscore the need to consider alternative approaches to improved soil resource management in the highlands. And while the colonial land management interventions obviously shaped many of the land use 233 practices in the region, many smallholders have also adopted and adapted those recommended land management practices that they perceived as useful. Change and adaptation While numerous population and large-scale system collapses have occurred in Africa in the last century, almost all have been caused by extreme social and/or biophysical perturbations; there is no evidence that these collapses have been significantly influenced by land degradation or land use pressures (Goldman, 1995, 1996). Ecological change has occurred in southwest Uganda for well over a millennium, and change remains a predominant feature of the contemporary highland agricultural systems. In Kabale and Kisoro Districts, smallholder agricultural and land management practices have also undergone widespread change over the last 50 to 60 years. While land use pressures in the region have lead to deforestation and land degradation, as well as the disappearance or alteration of traditional land management practices (e.g., extended fallows), many smallholders have also often adapted to internal and external pressures without completely destroying their production system. Many smallholders have demonstrated varied abilities to adapt to a variety of internal and external influences such as factor scarcities (e.g., land) and marketing opportunities (e.g., cash crop production) (e.g., Goldman, 1993b). Smallholders in the highlands have responded in a positive manner to land use pressures and other forces external to the region. Their responses include a modest increase in tree planting, the utilization of a system of fragmented landholdings to manage for agricultural risk, an increase in the use of short term fallows in the less densely 234 populated parishes, the use of organic inputs, the expansion of intercropping practices, the selected adoption of soil and water conservation measures, and the adaptation of cropping profiles to meet changing food preferences, capture cash crop market opportunities and intensify food production. The flexibility and resiliency that are inherent to these agricultural production systems are traits that should be supported rather than suppressed or narrowly standardized. Future initiatives to improve resource management in the highlands will benefit not only from an improved understanding of the knowledge and experience of smallholders, but also from better understanding the varying abilities of smallholders to adapt to changes and hence the overall diverse and dynamic nature of the highland agro-ecosystem. Conclusions Theoretical Implications Rapid population growth is often considered to be incompatible with the sustainable management of the environment, particularly in the densely populated regions of Africa. A number of recent studies suggest, however, that population growth can induce increased crop and tree production, and improved resource management (Holmgren et al., 1994; Lindblade et al., 1996; Mortimore, 1993). For example, a case study from a dryland area around Machakos, Kenya, has demonstrated that improved resource management and agricultural production can occur with increasing population growth (Mortimore and Tiffen, 1994; Tiffen et al., 1994). In Machakos District, where colonial staff once expressed concerns over land 235 degradation, the driving forces behind the transformation were land scarcity, fragmented landholdings, and the private appropriation of grazing land. Over time, as the size of the holdings shrank, the cultivated area in the district rose, leaving less land for grazing awhile the cultivated area per person stagnated. These changes lead to the integration of livestock and crop enterprises within the household production system, and to the overall intensification of agricultural production system (Mortimore and Tiffen, 1994; Tiffen et al., 1994). The highlands of southwest Uganda currently exhibit some of the characteristics of Machakos prior to its agricultural transformation: southwest Uganda is a densely populated region with small and fragmented landholdings, and it is widely assumed to be experiencing severe environmental degradation. Indeed, many observers suggest that the growing population and land use pressures indicate that a collapse of the system is imminent, and that alternative approaches to agricultural production and land management will be required in the highlands, if they are not already imperative. According to a Boserupian viewpoint, however, impressions or evidence of declining productivity are not simply indicators of a system collapse, but a necessary prelude to the intensification of land and labor use. Following Boserup, the preconditions for agricultural intensification are: increasing population density, peace and security for trade and investment, and equitable marketing and tenure systems (Mortimore, 1993). In the highlands of southwestern Uganda, many of these preconditions now exist. Additionally, as the rural youth population searches outside of the agricultural 236 sector for employment and moves to urban areas (e.g., Bryceson, 1996), the labor demands will increase for many smallholders, particularly women, and thus create additional pressures whereby smallholders might consider alternative measures to maintain or increase agricultural production. Some researchers posit that the labor constraints and land use pressures like those characteristic of the highlands will induce smallholders to intensify agricultural production by further increasing labor inputs, and/or expanding the use of purchased inputs and other technologies that might become available (e.g., Pingali et al., 1987). For smallholders involved in intensive agricultural production in the highlands, purchased inputs will most likely become increasingly attractive; there will be few long-term alternatives for soil fertility enhancement other than nutrient imports such as chemical fertilizers. Applied Implications The expanded use of purchased inputs might contribute to an improvement in agricultural production in the highlands, but it is unlikely that they will make an immediate or dramatic contribution towards the improvement of the agricultural production and land management systems in the highlands of Kabale and Kisoro Districts in the immediate future. Theoretically, the conditions might be such that smallholders in Kabale and Kisoro Districts are prepared to adopt intermediate technologies or purchased inputs, but these resources remain prohibitively expensive and are not yet readily available in the region. Until the appropriate infrastructure and market incentives are present to facilitate the adoption of these technologies, alternatives to purchased inputs need to be explored. Even if purchased inputs do 237 become widely utilized in the region, alternative technologies can complement the advantages of inputs (e.g., chemical fertilizers provide a concentrated supply of nutrients otherwise generally unavailable in the system), as well as provide multiple benefits of their own (e.g., green manures can fix biological nitrogen in the soil, but also provide organic matter to the soil and create a ground cover and thus reduce erosion). Alternative technologies to improve agricultural production and resource management in the highland agricultural systems might include "green-green" revolution technologies such as agroforestry, green manures, and improved biological nitrogen fixation. Many of these alternative technologies, however, also have their drawbacks: some technologies such as agroforestry have been introduced, but have not yet seen widespread adoption. Additionally, green manures and organic inputs will be able to address only some of the nutrient requirements in the agricultural system, and they are often labor-intensive management options. In the highland region, the rugged and hilly terrain, extensive wetland areas and fragmented landholdings reduces the likelihood that the agricultural systems will ever be widely mechanized or even cultivated using animal traction: hoe-based agriculture will undoubtedly be commonplace in the region for some years to come. The shortage of purchased inputs for the foreseeable future and the enduring importance of mixed subsistence and semi-commercial suggests that researchers might also consider practices already utilized by smallholders (e.g., from Muramba Parish: raised-bed agriculture, cultivation of staking material for climbing bean production) 238 for further promotion in the region. They should also include the refinement and/or extension of existing smallholder management practices-i.e., advocate the expanded use and improved management of composting, manure use, short-term fallows practices. To achieve successful adoption of agricultural and land management practices, smallholders must be able to implement and manage new technologies at relatively low cost and low risk. (Many available "on the shelf soil conservation technologies recommended for highland areas (e.g., bench terraces) are inappropriate for the fragmented landholdings and heterogeneous environments characteristic of Kabale and Kisoro Districts.) A reliance upon local knowledge, experience and resources also has the potential to minimize vulnerability for local smallholders, as well as reduce dependence on externally supplied inputs and experts. Given the diversity of environments and the variety of smallholders in the region, a range of adaptable land management technologies and strategies will need to be developed, tested and promoted. In short, menus of technical and management options will need to be created in collaboration with the prospective users: smallholders. Finally, smallholder knowledge and management systems continually evolve, a result of the dynamic and ever-changing nature of interactions between humans and their environment, and their efforts to solve existing problems and address new ones (Amanor, 1994a, 1994b; Brouwers, 1993). An improved understanding of the dynamic aspects of the highland agricultural production systems and the constraints that smallholders face can help create more multi-faceted and flexible alternatives. 239 Future research efforts should not, however, simply attempt to develop a menu of management practices, but also directly support the adaptive, innovative and experimental capacities of smallholders. Methodological Implications The success or failure of future resource management efforts in the highlands will depend on the participation of smallholders~the very people who are currently blamed for environmental mis-management in the highlands (e.g., Chambers, 1983, 1994c; Pretty, 1995). Smallholders will need to become more involved in agricultural and land management efforts~not simply as passive observers or "implementers" of technologies, but as active participants in the research and management processes. The expanded participation of smallholders in future soil resource management initiatives can facilitate the development of a range of technologies and recommendations which reflect local conditions. It can also begin to "invite" smallholders to feel greater "ownership" of the technologies, and thus improve the chance for adoption. To this end, it will be important to strengthen local social institutions to address land management problems and resource conflicts, initiate a multidisciplinary approach to soil resource management, and expand the role of smallholders in future programs. Smallholders, development workers and agricultural researchers all have an incomplete knowledge of the various social, economic, political and bio-physical elements of the highland agricultural systems. They each possess different yet valuable knowledge, experiences and skills that should be considered complementary, 240 rather than contradictory (Walker and Wortmann, 1993). The innovative capacities and experimental skills of smallholders merit expanded consideration and full-fledged support. Correspondingly, smallholders should not only be encouraged to participate in on-station and on-farm research, but their own adaptive traits and experimental skills should be supported. Lastly, collaboration between agricultural researchers, development workers and smallholders can assist in the development of participatory research methodologies. Smallholders' research priorities and management constraints need to be better understood, and the collaborative development of research methods can help to improve the flow of information and ideas between smallholders, researchers, development workers and policy makers. A strong collaboration among the various participating groups can also facilitate the development of successful soil technologies and resource management strategies which reflect local needs, resources and environments in the highland region. Recommendations for Future Research In terms of theoretical topics, there are a number of problems concerning the nature of agricultural and land use adaptation and change in the highlands of southwest Uganda that could be addressed. If the region is in fact posed for further agricultural intensification or a more dramatic and profound transformation, then opportunities will be available to investigate the nature of changes within the agricultural system, particularly the processes whereby smallholders adopt new technologies such as fertilizers, or adapt present land management practices (e.g., 241 modify fallow management to include green manures, etc.) It would also be useful to investigate whether smallholder knowledge-e.g., soil knowledge-changed with other changing aspects of the agricultural system, and if so how and under what conditions. Furthermore, the differentiation of agro-ecological knowledge along gender, age or economic lines might be explored. In terms of applied topics, research efforts in the highland region might be be directed towards the diverse plot management practices of smallholders. More specifically, investigations might address how the use of soil resource management practices vary between plots according to social differences among smallholders, as well as variations in soil types, distance to plot and the location of plots on the landscape. Another topic might involve a more in-depth examination of the gender- based dimensions of resource management problems in the region; the specific role of women within the highland agricultural system has not yet been fully explored. Finally, an investigation might be undertaken into the location-specific dimensions of environmental degradation, land use and resource management, i.e., evaluate the types, extent and causes of related variations in the highland region. Future endeavors to develop sustainable agroecosystems in the highlands will of southwest Uganda will depend on a better understanding of smallholder agroecological knowledge and resource management, the variety of factors influencing land-use systems, and the dynamic relationship between culture, agriculture and resource management. APPENDIX A MEAN MONTHLY RAINFALL (MM) FOR KABALE, UGANDA - 1918 TO 1993 APPENDIX B ANNUAL RAINFALL (MM) FOR KABALE, UGANDA - 1918 TO 1993 1992 1986 1980 C 1973 1967 : 1961 : 1955 : i 1949 1942 1936 1930 1924 1918 i 1 1 1 ; i — i mm 0 200 400 600 800 1000 1200 1400 1600 Source : DAO, 1994. 243 APPENDIX C QUESTIONNAIRE SURVEY I. IDENTIFICATION OF SAMPLE HOUSEHOLD 1. Parish: Single Response Only (SING) 1. Kicumbi 2. Nyarurambi 3. Muramba 2. Village Code: 3. Name of Household Head: II. PARTICULARS OF FIELD OPERATIONS 4. Interviewer: 5. Date of Survey: 6. Field Supervisor: 7. Date of Field Survey Inspection: 8. Date of Office Inspection: 9. Household Survey Code: (SING) 1 . Original Household Surveyed 2. Substitute Household Surveyed 3. Not Surveyed 10. Informant Code: (SING) 1 . Household Head 2. Other Member of Household 7. Other 244 245 11. Type of Informant Code: (SING) 1 . Cooperative and Capable 2. Cooperative but not Capable 3. Informant is Occupied 4. Not Prepared to Cooperate 5. Member is Away 7. Other III. SOIL RESOURCES 12. Which Local Soil Types Do You Cultivate? Multiple Response Possible (MULT) 1. Orushenyi 10. Urukoro 2. Eibumba 11. Insibo 3. Eryenombe 12. Umukara 4. Oruchuchu 13. Urugugo 5. Eryorugugo 14. Ingwa 6. Eryomushenyi 15. Orowondo 7. Eryorufunjo 16. Gahuma 8. Eryakatuku 17. Amabare 9. Eririkwiragura 77. Other 13. What Criteria Do You Use to Describe Soil Types? (MULT) 1 . Vegetation Type 13. Soil Taste 2. Vegetation Age 14. Soil Smell 3. Vegetation Vigor 15. Top Soil Depth 4. Soil Color 16. Natural Soil Fertility 5. Organic Matter Content 17. Elevation 6. Particle Size 18. Slope 7. Aggregate Nature 19. Direction of Slope-face 8. Soil Stickiness (clay?) 20. Position on Slope 9. Soil Texture 21. Erodibility 10. Stoniness 22. Permeability 11. Soil Tilth (Till-ability) 23. Moisture Holding Capacity 12. Soil Touch 77. Other 246 14. Evaluate Each Soil Type That You Cultivate: TABLE A.A D. c D. Fertility Erodibility Most 2nd-Most List Soil Type Rating: Rating: Favored Favored 1 - 5 1 - 5 Cron Cron 1. 2. 3. 4. 5. 6. 7. 8. A. FERTILITY RATING: 1. excellent fertility, 2. good fertility, 3. medium fertility, 4. poor fertility, 5. infertile. B. ERODIBILITY RATING: 1. Minimum erodibility, 2. low erodibility, 3. medium erodibility, 4. high erodibility, 5. highest erodibility. C/D. CROP LIST: 1. Sorghum, 2. Millet, 3. Maize, 4. Wheat, 5. Irish Potatoes, 6. Sweet Potatoes, 7. Bush Beans, 8. Climbing Beans, 9. Peas, 10. Vegetables, 11. Bananas, 12. Tobacco, 13. Pyrethrum, 14. Pasture, 15. Coffee, 16. Trees, 17. All Crops. 77. Local Soil Types: 1. Orushenyi 10. Urukoro 2. Eibumba 11. Insibo 3. Eryenombe 12. Umukara 4. Oruchuchu 13. Urugugo 5. Eryorugugo 14. Ingwa 6. Eryomushenyi 15. Orwondo 7. Eryorufunjo 16. Gahuma 8. Eryakatuku 17. Amabare 9. Eririkwiragura 18. All Soils 77. Other 247 IV. SOIL RESOURCE MANAGEMENT 15. Are There District-Level Soil and Water Conservation By-Laws? 1. Yes a) Are Any of the By-laws Practicle or Useful? I. Yes. a) Which By-Laws/Practices? b) Why? (MULT) -a- -b- 1. : 2. : 1. Reduces Soil Erosion/Landslides 2. Controls Run-off 3. Improves Soil Fertility 4. Enforced (Fines & Imprisonment) 5. Traditional 7. Other 2. No. Why Not? (MULT.) 1. Ineffective 2. Not Interested 3. High Labor Demands/Time Consuming 4. High Labor Costs 2. No. (Go to Question #19) 8. Don't Know 16. Are the By-Laws Enforced? 1. Yes a. By Whom? (MULT) 1. District Agricultural Office Staff (DAO) 2. Parish Chief 3. Local RC's 7. Other b. What are Penalties for Ignoring the By-Laws? (MULT) 1. Fines 2. Imprisonment 3. None 7. Other 2. No 8. Don't Know 248 17. Do You Have Soil Erosion on Your Plots? 1. Yes. a) What Type? (MULT) 1. Rill Erosion 2. Gully Erosion 3. Wind Erosion 4. Soil Slumping 5. Landslides 7. Other b) What are the Major Causes of Soil Erosion in Your Plots? 2. No 8. Don't Know 18. Do You Use Any of the Following Practices to Control Soil Erosion in Your Plots? A. B. C. D. E. Practices Used? Created Managed Pluses Minuses By? By? Tree Planting Grass Bunds Hedgerows/Shrubs Contour Planting Trash Lines Mulching Run-off Channels Soak-way Pits Check-Dams Other: A. Use: 1. Yes 2. No 8. Don't Know B. Initial Implementation: 1. Men 2. Women 3. Both 8. Don't Know C. Maintenance: 1. Men 2. Women 3. Both 8. Don't Know 249 D. Advantages E. Disadvantages 1. Demarcation/boundary 1. Uncontrolled Grazing 2. Reduces Soil Erosion 2. Heavy Labor Demands 3. Reduces Landslides 3. Easily Undermined 4. Reduces Flooding 4. Thievery 5. Provides Fodder/Fuelwood 5. Ineffective 7. Other 8. Don't Know 7. Other 8. Other 19. Do You Use Any Inputs In Your Plots? 1. Yes (If yes, continue to la.) 2. No. Why not? (MULT) 1 . Lack of inputs. 4. Lack of Labor 2. Cost of inputs. 7. Other 3. Cost of Labor 8. Don't Know la. Describe Your Input Utilization: A. B. C. D. E. F. Applied For ? For ? Inputs Use? By Whom Source Means Crops Soils Manure Compost Plant Trash Ashes Fertilizers Pesticides Improved Seeds Other: A. Use of Inputs: 1. Yes 2. No 8. Don't Know B. Input Management (Applied By ?): 1. Men 2. Women 3. Both 8. Don't Know C. Obtained Where: 1. Household 2. Store 3. Market 4. DAO 5. Relative/Neighbor 7. Other 8. Don't Know 250 D. Obtained How: 1. Self-supplied 2. Purchased 3. Credit 4. Loan 5. Given/Donated 6. Exchanged 7. Other 8. Don't Know E. Crop Types: 1. Sorghum 5. Irish Potatoes 9. Peas 13. Pyrethrum 2. Millet 6. Sweet Potatoes 10. Vegetables 14. Pasture 3. Maize 7. Bush Beans 11. Bananas 15. Coffee 4. Wheat 8. Climbing Beans 12. Tobacco 16. Trees 17. All Crops 77. Other 88. Don't Know V. LAND RESOURCES AND MANAGEMENT 20. How Many Plots Will You Manage/Have Use of This Year? 21. How Many of These Plots Do You? 1. Own. How Many of These Plots Are? (MULT) 1. Inherited 2. Marriage Gift 3. Purchased 4. Donation/Gift 5. Exchange 7. Other 2. Rent or Lease 3. Borrow 7. Other ~ 8. Not Known 22. How Many of These Plots are Located in: (MULT) 1. Swampland (Valley bottom) 2. Lower-slope (Foot-slope) 3. Mid-slope 4. Upper-slope 5. Ridge-top 6. Upper Valleys 7. Other 251 23. How Many of These Plots are _?_ From Home (Distance) (MULT.) 1 At Household . 2. Less than 1/2 mile from Household 3. More than 1/2 to 1 mile from Household 4. More than 1 to 3 miles from Household 5. More than 3 to 5 miles from Household 6. More than 5 miles from Household " 7. Other 24. How Many Plots Did You Cultivate Last Season (long rains/September-December)? 25. How Many Plots Will You to Cultivate This Season (short rains/March-May)? 26. For a Given Season, Are There Advantages to Cultivating More Than One Plot? 1. Yes. What Advantages? (MULT.) 1. Capture Soil Variation/Optimize Crop Prod. 2. Capture Micro-climatic Variation 3. Minimize Climatic Hazards (e.g., hailstones) 4. Reduce/Disperse Uncontrolled Grazing Losses 5. Reduce/Disperse Thievery Losses 6. Average Good and Poor Yields 7. Other 8. Don't Know 2. No 27. For a Given Season, Are There Disadvantages to Cultivating More Than One Plot? 1. Yes. What Disadvantages? (MULT.) 1. Uncontrolled Grazing Losses 2. Thievery Losses 3. Shortage of Labor 4. Travel Time or Distance to Plot 5. Shortage of Inputs (Compost, Manure) 6. General Management Problems 7. Other 8. Don't Know 2. No 252 28. Has Your Fallow-Management System Changed in the Last Twenty Years? 1. Yes. How? (MULT) 1 . Fallow Periods Have Shortened 2. Fallow Periods Have Lengthened 3. Fallow Periods Occur Less Often 4. Fallow Periods Required More Often 2. No 8. Don't Know 29. Are Any of Your Plots Currently in Fallow (Resting)? 1. Yes. How Many of Your Plots ? Why? 1. To restore soil fertility. 2. Lack of labor to cultivate all plots. 7. Other 2. No. Why Not? 1. Land Shortage 2. Not Necessary 3. Unaware 7. Other 30. Do the Soil Types You Cultivate Require Fallow Periods of Different Lengths to Restore Their Fertility? 1. Yes a. Which soils require the longest fallow periods? Longest: 2nd. 3rd. b. Which soils require the shortest fallow periods? Shortest: 2nd. 3rd. 2. No 31. Do You Currently Practice Crop Rotation? (MULT) 1. Yes. Why? 1. Changing Seasons/Rainfall Period 2. Given Crop Has Exhausted Soil Type 3. Break Disease/Pest Cycles 7. Other 2. No. Why Not? 1 . Land Shortage 2. Limited number of soil types 3. Lack of Labor 4. Don't Know Importance 7. Other 32. Do You Currently Practice Intercropping? (MULT) 1. Yes. Why? 1. Land Shortage 2. Diversify Food Production 3. Increase Food Production 4. Traditional Practice 7. Other 2. No. Why Not? 1. Infertile Soils 2. Climate Changes 3. Pest Problems 4. Lack of Labor 7. Other 33. What Do You Do With Plant Residues After Harvest? (MULT) 1 . Roof Thatching Material 2. Cooking Fuel 3. Burnt in the Plot 4. Buried in Plot 5. Dried and Turned into the Soil 6. Nothing 7. Other 8. Don't Know 34. What Do You Do With Weed Materials After Weeding? (MULT) 1. Burnt in the Plot 2. Buried in Plot 3. Dried and Turned into the Soil 4. Nothing 7. Other 8. Don't Know 35. What Are Your Major Crop Production Problems This Season? (Rank Top Five Problems, 1 - 5) 1. Labor Shortage 10. Soil Fertility Loss 2. Land Shortage 11. Soil Erosion 3. Land Fragmentation 12. Fallow Decline 4. Distance/Time to Plot 13. Lack of Fertilizers 5. Tool Shortage 14. Lack of Pesticides 6. Uncontrolled Grazing 15. Lack of Markets 7. Disease/Pest Damage 16. Lack of Improved Seeds 8. Decline of Crop Yields 17. Changing Climate 9. Decline of Crop Rotation 77. Other: VI. AGRICULTURAL LABOR 36. Do You Plan to Work as a Hired Agricultural Laborer This Year? 1. Yes. How Often? (Average) (SING.) 1. At Least Once a Week 2. At Least Once a Month 3. At Least Once a Cropping Season 7. Other 2. No 255 37. Do You Plan to Hire Agricultural Labor This Year? (MULT) 1. Yes. For Which Tasks? 1. Cultivation 2. Planting 3. Weeding 4. Harvesting 2. No 38. Do You Belong to Any Labor Sharing or Work Groups? 1. Yes. Name 2. No VII. PERSONAL CHARACTERISTICS OF RESPONDENT 40. Respondent's Name: 41. Respondent's Sex: 1. Male 2. Female 42. What is Your Age ? 43. What is Your Marital Status? (SING) 1. Married 2. Single/Never Married 3. Separated/Divorced 4. Widow(er) 44. What is Your Profession? (MULT) 1. Farmer 2. Civil Servant (Teacher, Health Worker, etc.) 3. Shopowner 4. Trader/Market Seller 5. Artisan/Craftsman 6. Administration Official/Resistance Council Member 7. Other 45. What is the Highest Level of Education You Completed? (SING) 1 . Not attended 2. Primary (Year 1 - 4) 3. Primary (Year 5 - 7) 4. Secondary 5. Post-Secondary 7. Other 46. Were You Born in This Parish? 1. Yes 2. No a. Where did you come from? (Parish/County/District) b. When (Year) ? c. Why Did You Leave Your Parish of Birth? (MULT) 1 . To Marry 2. Shortage of Land (No Land Available) 3. No Employment Available 4. Political Problems 7. Other d. Why Did You Come Here? (MULT) 1. To Marry 2. Purchased Land (Land Available) 3. Inherited Land 4. Family Resides Here 7. Other 47. How Many People Live in the Household ? APPENDIX D KABALE DISTRICT SOIL AND WATER CONSERVATION BY-LAWS This law shall be called the Kabale soil and water conservation law (1990). The following rules to prevent soil degradation shall be followed: 1. All land under cultivation or cleared for cultivation, or land planted to trees shall be provided with bunds of earth/ stone across the slopes and parallel to the counter at intervals not exceeding 16 yards apart on [land of moderate slope] and 12 yards on hilly areas. 2. Bunds shall have a minimum width of 3 feet and a height of not less than one foot. Each bund shall in cases be planted with grass (the downward facing slope inclusive), such as Kikuyu grass. 3. On sloping land planted to annual crops, trash lines consisting of dead vegetation shall be laid parallel to, and half-way between the existing bunds. 4. When crops are planted in lines, the lines shall be across the slope and parallel to the contours. In the case of sweet potatoes, the ridges shall be used and planting shall be done on soil ridges across the slope and parallel to the contour. In such a case lateral bunding shall be practiced. 5. No fields or lots shall be demarcated by furrows or gullies. Live hedges or shrubs or stone walls shall be used for demarcation. 6. No annual crops shall be cultivated within 9 feet of any perennial or seasonal water course or any maintained roads. 7. All paths, cattle tracks, ditches and access roads shall be protected against erosion by run-off channels, soak-way pits, or stakes to prevent erosion. 8. Paths or tracks may be closed by a Gombolola (Sub-County) Chief with the assistance of RC's [Resistance Council Members] to prevent erosion; alternative routes [shall be] provided. 257 258 9. Cultivators on any steep sided slopes vulnerable to erosion/landslides shall be stopped by the Gombolola Chiefs, with the help of the RC's, on the advice of the District Agricultural Officer (DAO) or his field assistants, as and when deemed necessary to prevent soil erosion. Alternative vegetation (e.g., Black Wattle, Cypress, Leucaena, Sesbania, and Erythrina abysinnia), shall be planted on such land. 10. Any form of grass burning, be it trash or on hillsides, is prohibited. Noxious weeds like couch grass (lumbugu) and comelina shall be heaped together to decay. 11. All house compounds, except the winnowing area and compounds around buildings, shall be grassed over with a low-growing grass. Stone-filled trenches shall be constructed under food caves and the trenches shall discharge into soakway pits. 12. In the case of reclaimed swamps, the land holder shall ensure that the water channels bordering on or passing through the land allocated to him/her, are kept clear of earth, rubbish and other obstructions for smooth water flow and prevention of flooding. 13. Land planted to bananas or coffee shall be covered with mulch where possible. 14. All Black wattle [and] any other trees, whether in lines or plantations, shall be thinned in the year of germination to 4.5 feet apart and at the age of three years to a minimum of 9 feet apart. 15. Any person disobeying the provisions of this law shall be guilty of an offence and shall on first conviction be liable to a fine not exceeding Uganda shillings 3,000/ = 1 or imprisonment for 15 days, or both, and shall on subsequent conviction be liable to a fine not exceeding 5,000/= or to imprisonment, [whichever is deemed] more effective. The Kabale District Soil and Water Conservation Law, Considered and Passed by the Kabale District Resistance Council, May, 1990. Source : DAO, 1992. 1 1990 Exchange Rate (Approximate): US$ 1 = Ugandan shillings 950. 259 Note: At the time of the field research for this project, the most recent revision of the by-laws had been undertaken in 1990. Many of the by-laws are remain relatively vague, nor are the terms (e.g., soak-way pits) are not clearly defined. This ambiguity contributes to the confusion over how to implement the SWC by-laws among smallholders. The rough exchange rate in 1990 was $1 U.S. to 950/= Shillings Ugandan. , APPENDIX E PROPOSED SOIL AND WATER CONSERVATION BY-LAWS FOR KISORO DISTRICT Draft Prepared by the Department of Agriculture (DAO), Kisoro District, 1993) Summary: 1 . This law shall be called "Kisoro District By-law" for soil and water conservation. 2. A steady soil erosion and water loss has been going on for many years. Therefore it has been found imperative to find a measure that would facilitate reduction of such loss especially where man is the main agent. Hence the drafting of the by-laws governing water and soil loss through erosion which must be revived. Erosion: For purposes of this document, erosion shall mean the loss of soil caused by water, wind, fire or animals. Therefore, any person who causes soil loss by or through: (a) Running water on cultivated bare fields. (b) Concentrating cattle in/on path resulting into gullies. (c) Over-grazing land or: by destroying soil conservation structures, or wind breaks or by burning bushes or by setting fire on bushes, etc. without authority is guilty and liable to fine or imprisonment. The By-Laws: 1. This law shall be called the Kisoro District Soil and Water Conservation law (1993). 2. The following rules to prevent soil degradation shall be followed. 260 261 3. All land under cultivation or cleared for cultivation or land planted with trees, shall be provided with stop-wash bunds of grass/shrub coverage, across the slopes and parallel to each other at intervals (horizontal interval) of NOT more than 15 meter a part on flat land and 10 meters on hilly surfaces/ slopes (7-10 meters). 4. A bund shall have a minimum width of 1 meter (feet) and a height of not less than 0.6 [meters] (2 feet). Each bund shall be planted with Kikuyu grass (umucyacya) or Elephant grass (urubingo) or any other suitable grasses. The same bunds shall be made more stable by planting two rows of trees/shrubs such as Vanomia spp (imibirizi), Cassuarina, Leuceana, Sesbania and Erythrina abysinica (igiko). 5. On slopes which are more than 25%, [the] Hinga-Laza method is to be recommended. 6. When crops are planted in lines, the lines shall be across the slopes and parallel to the contour lines (bunds). 7. No fields or plots shall be demarcated by furrows; large trees such as eucalyptus, posts, live hedges, shrubs and stonewalls shall be used for demarcation of [an] individual's fields or land. 8. No crops shall be cultivated within 3 meters (9 feet) of any perennial or seasonal waterway or any maintained Feeder Road. 9. All paths (burungi bwansi), cattle tracks, ditches and access roads shall be protected against soil erosion by run-off channels, soak-way pits, etc. 10. Cultivation on any steep sided-slopes vulnerable to erosion/land slides shall be stopped by the chiefs with the help of the RC's on the advice of the District Agricultural Officer or his representative as and when deemed necessary to prevent soil erosion. Available trees shall be planted (in order) on such land. 11. Any form of grass burning be it trash (crop waste) on hill sides is prohibited. Noxious weeds like couch grass (urwire) and comelina shall be heaped together to decay over a time. 12. All house compounds: except [for] drying/winnowing area, all compound around buildings shall be grassed over with a low-growing grass. Stone-filled trenches shall be grassed over with a low-growing grass. Stone-filled trenches shall be constructed under roof eaves and the trenches shall discharge into soak-way pits. 262 13. In case of reclaimed swamps, the land holder shall ensure that the water channels bordering on or passing through the land allocated to him/her are kept clear of earth, rubbish and other obstructions for smooth water flow and prevention of flooding. 14. Land planted to bananas, coffee fruit trees, etc. shall be covered with mulch where possible. In absence of mulch, a cover crop is recommended, cut-off drains or soak pits shall be constructed in the plantation accordingly. 15. Agro-forestry is to be introduced as soon as possible. 16. All recommended trees, whether in lines or plantations, shall be thinned to l spacing or 4 A feet (1.4 meters) within a year of planting and 9 feet (3 meters) after 3 years. 17. All large trees shall be planted in a distance of not less than 15 feet (5 meters) away from-crop-fields. 18. Any person disobeying the provisions of this law shall be guilty of an offence and shall on first conviction: (a) be liable to a fine not less than shs. 3,000/= and not exceeding 10,000/ = or imprisonment for 15 days or both and shall on any subsequent conviction be liable to a fine not less than 5,000/= and not exceeding 20,000/= or to imprisonment for one-month or both with the exception of fine. (b) To fine not less than 10,000/= and not exceeding 30,000/= or imprisonment for (six months it he or she fails to prevent any fire law rully lit by him on land occupied or owned by him or lit with his authority or consent, from extending on to the land of any other person, or form causing damage to the property of any other person, shall be guilty of [a] misdemeanor. (c) To fine not less than 10,000/= or imprisonment for three years it he or she wilfully and unlawfully sets fire on: i. a crop or hay or ley under cultivated land. ii. standing trees, saplings or shrubs whether indigenous or not, under cultivation, is guilty of a felony. Source : DAO, 1993. 263 Note: The proposed by-laws were not yet formerly adopted in mid- 1994, although the DAO expected this final version would be passed. The by-laws are adapted directly from the Kabale Soil and Water Conservation By-laws. As with the Kabale by-laws, many of the regulations are ambiguous and terms are not defined. The rough exchange rate in 1993 was $1 U.S. to 1,000/ = Shillings Ugandan. APPENDIX F FERTILITY CAPABILITY CLASSIFICATION SYSTEM The Fertility Capability Classification system FCC) is a technical system for grouping soils according to the kinds of problems they present for agronomic management of their chemical and physical properties. INTERPRETATION OF FCC NOMENCLATURE Type: Texture of plow-layer or surface 20 cm, whichever is shallower; S = Sandy topsoils: Loamy sands and sands (by USDA definitions); L = Loamy topsoils: < 35% clay but not loamy sand or sand; C = Clayey topsoils: > 35% clay; O = Organic Soils: > 30% O.M. to a depth of 50 cm or more. Substrata type (texture of subsoil). This designation is used only if there is a marked textural change from the surface, or if a hard root/restricting layer is encountered within 50 cm: S - Texture as in type; L = Texture as in type; C = Texture as in type; R = Rock or other hard root - restricting layer. 264 265 Modifiers: g = (gley); soil or mottles < 2 chroma within 60 cm of the soil surface and below all A horizons, or soils saturated with water for > 60 days in most years. a = (aluminum - toxicity): > 60% Al-saturation of the effective CEC within 50 cm of the soil surface, or > 67% acidity saturation of CEC by sum of cations at Ph 7 within 50 cm of the soil surface or pH < in 4.6 1:1 H20 within 50 cm except in organic soils where pH must be less than 4.7; h = (acid); 10-60% Al-saturation of the effective CEC within 50 cm of soil surface, - or Ph in 1:1 H2 0 between 4.6 6.0; i = (high P-fixation by iron); % free Fe^ 1 clay > 0. 15 and more than 35% clay, or hues of 5YR or redder and granular structure. This modifier is used on in clay (C) types; it applies only to plow-layer or surface 20 cm of soil surface, whichever is shallower. k = (low K reserves): < 10% weatherable minerals in silt and sand fraction within 50 cm of the soil surface, or exchangeable K < 0. 15 meq per 100 g. = (gravel); a prime (') denotes 15-35% gravel or coarser (> 2 mm) particles by volume to any type or substrata type texture; two rime marks (") denote more than 35% gravel or coarser particles (> 2 mm) by volume in any type or substrata type. Interpretation of types and substrata types: S : high rate of infiltration, low water - holding capacity. L : medium infiltration rate, good water - holding capacity. C : low infiltration rates, good water - holding capacity, potential high runoff if sloping, difficult to till; when i modifier is present, these (Ci) soils are easy to till, have high infiltration rates and low water - holding capacity. artificial O : drainage is needed and subsidence will occur; possible micronutrient deficiencies. 266 LC: susceptible to severe soil degradation from erosion exposing undesirable subsoil; high priority should be given to erosion control. Interpretation of Modifiers: g : denitritication frequently occurs in anaerobe subsoil; tillage operations and certain crops may be adversely affected by excess rain unless drainage is improved by tilling or other drainage procedures; good soil moisture regime for rice production. a : plants sensitive to Al-toxicity will be affected unless lime is applied; extraction of soil water below depth of lime incorporation will be restricted; lime requirements are high; this modifier is desirable for rapid dissolution of phosphate rocks. Mn-toxicity may occur on some of these soils. h : low to medium soil acidity; requires liming for Al-sensitive crops such as cotton. i : high P-fixation capacity; requires high levels of P fertilizer or special P management practices; sources and method of P fertilizer application should be considered carefully; with C texture, these soils have granular soil structure. k : low ability to supply K; availability of K should be monitored and K fertilizers may be required frequently; potential K-Mg-Ca imbalances. Sample interpretation of FCC unit: sh: high rate of infiltration, low water - holding capacity low to medium soil acidity; requires liming for Al-sensitive crops such as cotton. Lkh: medium infiltration rate, good water - holding capacity; low ability to supply K; availability of K should be monitored and K fertilizers may be required frequently. Source: Sanchez, Cuoto and Buohl, 1982. APPENDIX G CHEMICAL AND PHYSICAL PROPERTIES OF TOPSOIL SAMPLES H • 268 lO o >n 00 os CM OS Tt •* m o OS so m 00 00 CO m <* 3 Tt Tt in CM m in so ci ci CM Tt Tt Tt *i o Tt 1/1 Tt Os r» so - m CM CM oo cm CM CM CM CM i DO CM Tt Tt CM 00 SO 00 CM 00 o 00 SO so CM Tt O Tt O CM C> o CM CM CM >n CM CM CM CM ci CM CM CM I 1 1 cm Na mg/lOOg — - CM — cm >n O in Tt Cl ci cm CM - - a. CM 5 Tt r» mg/lOOg m CM 00 00 t» r- K m »— m m - m m in 00 2 00 00 in 00 m OS OS •0 00 Os 00 Os in Ca mg/lOOg m m CM c Organic o Matter D q q °) ON Tt OS Os os in Tt SO q ci CM Os s % cm CM CM r~ cm so Tt ci ro CM ci Tt Tt Tt Tt Tt o '8 OS 00 cm Tt SO m vo Tt — Os r~ m r- 00 00 Tt m SO ° a,** CO Tt in m in in so SO SO so SO m Tt Tt Tt m in >n o 1 1 Type 0 'Pi Soil i cd § O c ^£ X9 3 i 1 1 §> Local b X s 5 B s m s s O X o B O a tt Soil rj cn Cl Csj Sample N CM Cl CM Cl CM ci cm cm ci m' Tt Tt Tt in in in sd sd SD a 5 Z 1 269 00 c> ov en ci r- Ov * 1 r- r- Ov Ov Ov r- m pi m r- 00 00 m m CI tS (N (N in m t t- Ov cv. VO VO 1 m CI fS o ts ts CM (N fS (N 9 • 00 \o 00 00 1 00 VO VO r- r- Ov m in r~ r- *D fS fS ts ci ci - Na mg/lOOg SO 00 00 CI m vo VO VO vo Ov vo 00 00 VO 00 r~ tS r*» in CI mg/lOOg m 00 00 Ov r- 00 00 00 in 00 in K (N ri m m »-« - Ov in vo in 00 00 00 00 Ov 00 Ca mg/lOOg •n o o o o o 00 r~ m r» r- r-» a a. 00 d o ts in vo 00 00 00 CI tS o r- r- d in m CI t (S ts Organic Matter in vo vo Ov 00 Ov ts tT VO VO ts in «n o in vo VO r~ 00 % rn ri fj vd Ov iri vo' in in vd 00 oo' r~ IT) in 3 — a DC ts "1 "1 Ov r-; © ts ts in rn vo ts - Type Soil o 3 s f 1 3y o a o Local o S 2 5 a 5 o s X O • S * W s o # Soil ts CI (N ci Sample -o On o ON ON * 1 t en i r- On 1 m CN no no v© t- On cn Cn *o in 1 • i >n ON (N cn CN CN CN CN CN (N >> IT) • NO in ON Na mg/lOOg •t 00 NO 00 O o (N (N CN - 00 00 r- NO - NO CI K mg/lOOg (N CM ts 00 00 m 00 en ON in m ON 00 00 •* 00 — Ca mg/lOOg m o 00 en c> >/> 8 o m CN cs CN CN m cn tn CN cn ON ON a a- o. * o o o o r- CN CN 00 NO CS On p» CN o CN CN o >r> Q. en * Organic Matter 00 © in co On ci 00 00 >n On in 00 NO NO en en d in cn ri cn © no' no' r-' cn % cm ts ~ X vD NO in C-- r-; CI o <* CN no 00 "* © co no no NO* NO* NO NO <* Tf in U"i no' Type '"""> Soil 0 o to 00 3 -| 60 1 a Local 1 | g X> > s s B B B o 1 1 1 UJ • o c s # Soil CN cn CN m cn en CN en •—1 Sample CN en 271 -o p- in in p- p~ en 55 in i in en CO oo 00 oo oo NO NO NO NO en NO NO p~ P- p- - <** t 1 i O t T en ci fS NO en en NO - - On P» i ON en - en m in U OO * CI in 00 00 00 00 Na mg/lOOg o r- ON p- NO CI C-) tN c> CI m NO o 00 ts 00 <*N o t 00 C-4 00 in m m 00 O Ca mg/lOOg O t*» On 8 in en NO On en P~ p-j p~ en "* en tN * •* m tN § en S 00 a- d. o Tt" NO o 00 fS ON 00 NO in 00 O i/i o pi fS CI ts tN 8 in a. p- IN p~ 00 Organic Matter p- o On NO NO m Type Soil S 2 '>» e o o Local 1 1 1 1 si O * a M p B a # Soil en CM ci en en Sample ON ON On © d 6 Abel, N. 1988. The Impact of AIDS on Food Production Systems in East and Central Africa. In The Global Impact ofAIDS, eds. M. Carballo and J. Mann, pp. 36-53. New York: Alan R. Liss. Adams, W. 1989. Indigenous Farmer Managed Irrigation in Sonjo, Tanzania. The Geographical Journal 160:17-33. Adams, W. and Anderson, D. 1988. Irrigation Before Development: Indigenous and Induced Change in Agricultural Water Management in East Africa. African Affairs 87:519-536. Admassie, Y. 1992. The Catchment Approach to Soil Conservation in Kenya. Nairobi: Regional Soil Conservation Unit/Swedish International Development Authority. Agrawal, A. 1995. Dismantling the Divide Between Indigenous and Scientific Knowledge. Development and Change 26:413-439. Albert, H. and Runge-Metzer, A. 1995. Management of Natural Resources: Systems Approaches to Striga Control in Sub-Saharan Africa. Journal of Farming Systems Research-Extension. 5:91-106. Alexandratos, N. 1988. World Agriculture: Toward 2000. An FAO Study. London: Belhaven Press. Allan, W. 1965. 77a? African Husbandman. New York: Barnes & Noble. Allan, N., Knapp, G. and Stadel, C. eds. 1988. Human Impact on Mountains. Totowa, NJ: Rowman & Littlefield. Ali, A. 1995. Population Pressure, Environmental Constraints and Agricultural Change in Bangladesh: Examples From Three Agroecosystems. Agriculture, Ecosystems and Environment 55:95-109. 272 - Almy, S., Ateh, C, Woldetatios, T., Mboussi, M., Poubon, C. and Besong, M. 1991. Farmer's Approaches to Soil Fertility Maintenance Under Reduced Fallows in the Southwest Province Of Cameroon. Journal for Farming Systems Research Extension 2:1-18. Altieri, M. 1991. Traditional Farming in Latin America. The Ecolosist 21-93- 96. Amann, V., Belshaw, D. and Stanfield, J. eds. 1972. Nutrition and Food in an African Economy. Vol. 1. Kampala: Makerere University. Amanor, K. 1991. Managing the Fallow: Weeding Technology and Environmental Knowledge in the Krobo District, Ghana. Agriculture and Human Values VIII:5-13. 1994a. • The New Frontier. Farmer Responses to Land Degradation: A West African Study. London: Zed Books. 1994b. • Ecological Knowledge and the Regional Economy: Environmental Management in the Asesewa District of Ghana. Development and Change 25:41-67. Anderson, D. 1984. Depression, Dust Bowl, Demography and Drought: The Colonial State and Soil Conservation in Africa. African Affairs 83 321 343. Anderson, J. 1990. Soil Conservation in Developing Countries. Washington DC: World Bank. Anderson, R., May, R., Boily, M., Garnett, G. and Rowley, J.. 1991. The Spread of HIV-1 in Africa: Sexual Contact Patterns and the Predicted Demographic Impact of AIDS. Nature 352:581-589. Andriesse, J. 1988. Nature and Management of Tropical Peat Soils. FAO Soils Bulletin No. 59. Rome: Food and Agriculture Organization (FAO) of the United Nations. Anusionwu, E. 1977. A Framework for a Regional Study of African Agricultural Rural Economy: A Case Study ofKigezi District in Uganda. Ph.D. Dissertation, Cornell University. Arizpe, L., Stone, P. and Major, D. eds. 1994. Population and the Environment: Rethinking the Debate. Boulder, CO: Westview Press. Arnould, E. 1990. Changing the Terms of Rural Development: Collaborative Research in Cultural Ecology in the Sahel. Human Organization 49:339-354. Arrow, K., Bolin, B., Costanza, R. Dasgupta, P., Folke, C, Holling, C.S., Jansson, B., Levin, S., Maler, K., Perrings, C. and Pimentel, D. 1995. Economic Growth, Carrying Capacity, and the Environment. Ecological Economics 15:91-95. Attems, M. 1968. Permanent Cropping in the Usambara Mountains: the Relevancy of the Minimum Benefit Thesis. In Smallholder Farming and Smallholder Development in Tanzania, ed. H. Ruthenberg, pp. 137-174. Munich: Weltforum Verlag. Austen, A. and Hendrick, D. 1983. The Role of Technology in the African Past. African Studies Review 26:162-184. Bagoora, F. 1988. Soil Erosion and Mass Wasting in the Highland Area of Uganda. Mountain Research and Development 8:173-182. Baker, R. 1984. Protecting the Environment against the Poor. The Ecologist 14:53-60. Barnett, T. 1994. The Effect of HIV/AIDS on Farming Systems and Rural Livelihoods in Uganda, Tanzania and Zambia. Rome: Food and Agricultural Organization of the United Nations. Barnett, T. and Blaikie, P. 1992. AIDS in Africa: Its Present and Future Impact. London: Belhaven Press. Barnett, T., Tumushabe, J., Banteyba, G. Ssebuliba, R. Ngasongwa, J., Kapinga, D., Ndelike, M., Drinkwater, M. Mitti, G. and Haslwimmer, M. 1995. The Social and Economic Impact of HIV/AIDS on Farming Systems and Livelihoods in Rural Africa: Some Experience and Lessons from Uganda, Tanzania and Zambia. Journal of International Development 7:163-176. Barney, G. ed. 1988. 77a? Global 2000 Report. Washington: Seven Locks Press. Barrows, R. and Roth, M. 1990. Land Tenure and Investment in African Agriculture: Theory and Evidence. The Journal of Modern African Studies 28:265-297. 275 Bassett, T. 1988. The Political Ecology of Peasant-herder Conflicts in the Northern Ivory Coast. Annals of the Association ofAmerican Geographers 78:453-472. Bates, R. 1981. Markets and States in Tropical Africa: The Political Basis of Agricultural Policies. Berkeley, CA: University of California Press. 1989. • Beyond the Miracle of the Market: The Political Economy of Agrarian Development in Kenya. Cambridge: Cambridge University Press. Baxter, P. 1959. The Kiga. In East African Chiefs, ed. A. Richards, pp. 278- 310. London: Faber and Faber. Bebbington, A. 1993. Modernization from Below: An Alternative Indigenous Development? Economic Geography 69:274-292. Becker, C. 1990. The Demo-Economic Impact of the AIDS Pandemic in Sub- Saharan Africa. World Development 18:1599-1619. Behrens, C. 1989. The Scientific Basis for Shipibo Soil Classification and Land Use: Changes in Soil-Plant Relationships with Cash-Cropping. American Anthropologist 91:83-100. Beinhart, W. 1984. Soil Erosion, Conservationism and Ideas about Development: A Southern African Exploration. Journal of Southern African Studies 11:52-83. 1989. Introduction: • The Politics of Colonial Conservation. Journal of Southern African Studies 15:143-162. Bell, M. and Roberts, N. 1991. The Political Ecology of Dambo Soil and Water Resources in Zimbabwe. Transactions of the Institute of British Geographers 16:301-318. Bellon, M. 1995. Farmers' Knowledge and Sustainable Agroecosystem Management: An Operational Definition and an Example from Chiapas, Mexico. Human Organization 54:263-272. Bellon, M. and Taylor, J.E. 1993. "Folk" Soil Taxonomy and the Partial Adoption of New Seed Varieties. Economic Development and Cultural Change 41:763-786. Bellows, B. 1992. Sustainability of Steep Land Bean (Phaseoulus Vulgaris L.) Farming in Costa Rica: An Agronomic and Socio-Economic Assessment. Ph.D. Dissertation, University of Florida. Belshaw, D. 1980. Taking Indigenous Knowledge Seriously: The Case of Inter-cropping Techniques in East Africa. In Indigenous Knowledge Systems and Development, eds. D. Brokensha, D. Warren, and O. Werner, pp. 197-204. Lanham, MD: University Press of America. Bennett, J. W. 1969. Northern Plainsmen. Arlington Heights, IL: AHM. 1976. • The Ecological Transition: Cultural Anthropology and Human Adaptation. New York: Pergamon Press. Bentley, J. 1987. Economic and Ecological Approaches to Land Fragmentation: In Defense of a Much Maligned Phenomenon. Annual Review of Anthropology 16:31-67. 1989. • What Farmers Don't Know Can't Help Them: The Strengths and Weaknesses of Indigenous Technical Knowledge in Honduras. Agriculture and Human Values VL25-31. Berkes, F. ed. 1989. Common Property Resources: Ecology and Community Based Sustainable Development. London: Belhaven Press. Bernard, F. and Thorn, D. 1981. Population Pressure and Human Carrying Capacity in Selected Locations in Machakos and Kitui Districts. Journal of Developing Areas 15:381-406. Bernard, F., Campbell, D. and Thorn, D. 1989. Carrying Capacity of the Eastern Ecological Gradient of Kenya. National Geographic Research 5:399-421. Bernard, H.R. 1994. Research Methods in Anthropology: Qualitative and Quantitative Approaches. Thousand Oaks, CA: Sage Publications. Berry, L., Lewis, L. and Williams, C. 1990. East African Highlands. In The Earth as Transformed by Human Action, eds. B.L. Turner II, W. Clark, R. Kates, J. Richards, J. Mathews and W. Meyer, pp. 533-542. Cambridge: Cambridge University Press and Clark University. Berry, S. 1984. The Food Crisis and Agrarian Change in Africa: A Review Essay. African Studies Review 26:59-112. 277 • 1989a. Access, Control and Use of Resources in African Agriculture: An Introduction. Africa 59:1-5. • 1989b. Social Institutions and Access to Resources. Africa 59:41-55. • 1993. No Condition is Permanent: The Social Dynamics ofAgrarian Change in Sub-Saharan Africa. Madison, WI: University of Wisconsin Press. Bertrand, J., Bauni, E., Lesthaeghe, R., Montgomery, M., Tambashe, O. and Wawer, M. 1993. Factors Affecting Contraceptive Use in Sub-Saharan Africa. Washington, DC: National Academy Press. Beyer, J. 1980. Africa. In World Systems of Traditional Resource Management, ed. G. Klee, pp. 5-37. New York: John Wiley & Sons. Biggs, S. and Farrington, J. 1991. Agricultural Research and the Rural Poor. Ottawa: International Development Research Centre (IDRC). Bilsborrow, R. and Geores, M. 1994. In Population and the Environment: Rethinking the Debate, eds. L. Arizpe, Stone and D. Major, pp. 171- 207. Boulder, CO: Westview Press. Bilsborrow, R. and Ogendo, 0. 1992. Population-driven Changes in Land Use in Developing Countries. Ambio 21:37-45. Biwas, M. 1994. Agriculture and Environment: A Review, 1972 - 1992 Ambio 23:192-197. Black, R. 1990. Regional Political Ecology in Theory and Practice: A Case Study from Northern Portugal. Transitions of the Institute of British Geographers 15:35-47. Blaikie, P. 1985. The Political Economy of Soil Erosion in Developing Countries. London: Longman. 1988. Land • Degradation in Nepal. In Deforestation: Social Dynamics in Watersheds and Mountain Ecosystems, eds. J.D. Ives and D.C. Pitt. London: Routledge. !989a. • Explanation and Policy in Land Degradation and Rehabilitation for Developing Countries. Land Degradation and Rehabilitation 1:23- 37. 1989b. • Environment and Access to Resources in Africa. Africa 59-18- 40. Blaikie, P. and Brookfield, H. 1987a. Land Degradation and Society. London: Metheun. 1987b. Defining • and Debating the Problem. In Land Degradation and Society, eds. Blaikie, P. and Brookfield, H., pp. 1-26. London: Metheun. 1987c. • Approaches to the Study of Land Degradation. In Land Degradation and Society, eds. Blaikie, P. and Brookfield, H., pp. 27- 48. London: Metheun. 1987d. Colonialism, • Development and Degradation. In Land Degradation and Society, eds. Blaikie, P. and Brookfield, H., pp. 100- 121. London: Metheun. Blarel, B., Hazell, P., Place, F. and Quiggin, J. 1992. The Economics of Farm Fragmentation: Evidence from Ghana and Rwanda. World Bank Economic Review 6:233-254. Blustain, H. 1982. Social Issues in Technology Choice: Soil Conservation in Jamaica. Journal of Soil and Water Conservation 37:323-325. 1985. The Political • Context of Soil Conservation Programs in Jamaica. Human Organization 44:124-131. Bocco, G. 1991. Traditional Knowledge for Soil Conservation in Central Mexico. Journal of Soil and Water Conservation. 46:346-348. Bodley, J. ed. 1988. Tribal Peoples and Development Issues: A Global Overview. Mountain View, CA: Mayfield Publishing. Bojo, J. 1991. Economics and Land Degradation. Ambio 20:75-79. Bond, G.C. and Vincent, J. 1991. Living on the Edge: Changing Social Structures in the Context of AIDS. In Changing Uganda, ed. Hansen and Twaddle pp. 113-129. London: James Curry. Bongaarts, J. 1996. Global Trends in AIDS Mortality. Population and Development Review 22:21-45. . 279 Boserup, E. 1965. The Conditions of Agricultural Growth: The Economics of Agrarian Change Under Population Pressure. Chicago: Aldine. 1981. • Population and Technological Change: A Study of Long-Term Trends. Chicago: University of Chicago Press. • 1985. Economic and Demographic Interrelationships in Sub-Saharan Africa. Population and Development Review 11:383-397. Brasnett, N. 1936. Soil Erosion. Uganda Journal 4:156-161. Brett, E. 1992. Colonialism and Underdevelopment in East Africa. The Politics of Economic Change, 1919-1939. Brookfield, VT: Gregg Revivals. Broad, R. 1994. The Poor and the Environment: Friends or Foes? World 1 Development 22 : 8 1 -822 Brokensha, D., Warren, D. and Werner, O. eds. 1980. Indigenous Knowledge Systems and Development. Washington, DC: University Press of America. Brookfield, H. 1972. Intensification and Disintensification in Pacific Agriculture: A Theoretical Approach. Pacific Viewpoint 13:30-48. . 1984. Intensification Revisited. Pacific Viewpoint 25:15-44. Brookfield, H. and Padoch, C. 1995. Appreciating Agrodiversity: A Look at the Dynamism and Diversity of Indigenous Farming Practices. Environment 36:7-11, 37-45. Brouwers, J. 1993. Rural People's Response to Soil Fertility Decline. The Adja Case (Benin). Wageningen University Papers 93-4. Wageningen: Agricultural University Wageningen. Browder, J. 1995. Redemptive Communities: Indigenous Knowledge, Colonist Farming Systems, and Conservation of Tropical Forests. Agriculture and Human Values XII: 17-30. Brown, L. 1988. The Changing World Food Prospect: The Nineties and Beyond. WorldWatch Paper 60. New York: W.W. Norton. ed. 1989. State of the World: 1990. New York: W.W. Norton. -ed. 1994. State of the World: 1995. New York: W.W. Norton. Brown, L. and Kane, H. 1994. Full House: Reassessing the Earth's Population Carrying Capacity. New York: W.W. Norton & Co. Brown, L. and Wolf, E. 1984. Soil Erosion: Quiet Crisis in the World Economy. WorldWatch Paper 60. New York: W.W. Norton. • 1985. Reversing Africa's Decline. Worldwatch Paper 65. New York: W.W. Norton. . 1995. Facing Food Scarcity. World Watch November/December: 10- 20. Brush, S. and Stabinsky, D., eds. 1996. Valuing Local Knowledge: Indigenous People and Intellectual Property Rights. Covelo, CA: Island Press. Brush, S. and Turner, B.L. II. 1987. The Nature of Farming Systems and Views of Their Change. In Comparative Farming Systems, ed. B.L. Turner and S. Brush, pp. 11-48. New York: Guilford Press. Bryan, R. 1968. The Development, Use and Efficiency of Indices of Soil Erodibility. Geoderma 2:5-26. Bryant, R. 1992. Political Ecology: An Emerging Research Agenda in Third- World Studies. Political Geography 11:12-36. Bryceson, D. 1996. Deagrarianization and Rural Employment in Sub-Saharan Africa: A Sectoral Perspective. World Development 24:97-111. Buohl, S., Hole, F. and McCracken, R. 1989. Soil Genesis and Classification. 3rd Edition. Ames, IA: Iowa State University Press. Burwell, T. 1995. Bootlegging on a Desert Mountain: The Political Ecology of Agave (Agave spp.) Demographic Change in the Sonora River Valley, Sonora, Mexico. Human Ecology 23: 407-428. Butzer, K. 1989. Cultural Ecology. In Geography in America, eds. Gary L. Gaile and Cort J. Willmott, pp. 192-207. Columbus, OH: Merrill. 1990. • The Realm of Cultural-Human Ecology: Adaptation and Change in Historical Perspective. In The Earth as Transformed by Human Action, eds. B.L. Turner II, W. Clark, R. Kates, J. Richards, J. Mathews and W. Meyer, pp. 785-701. Cambridge: Cambridge University Press and Clark University. Byagagaire, J. and Lawrence, J. 1957. The Effect of Customs of Inheritance on the Sub-Division and Fragmentation of Land in South Kigezi. In Land Tenure in Uganda, Anonymous editor, pp. 17-22. Entebbe: Uganda Protectorate. Cahen, L. and Snelling, N. 1966. The Geochronology of Equatorial Africa. Amsterdam: North-Holland Publishing. Caldwell, J. and Caldwell, P. 1987. The Cultural Context of High Fertility in Sub-Saharan Africa. Population and Development Review 13:409-436. . 1990. High Fertility in Sub-Saharan Africa. Scientific American 5:118- ' 125. . 1993a. The Nature and Limits of the Sub-Saharan African AIDS Epidemic: Evidence from Geographic and Other Patterns. Population and Development Review 19:817-848. . 1993b. The South African Fertility Decline. Population and Development Review 19:225-262. Caldwell, J., Caldwell, P. and Quiggin, P. 1989. The Social Context of AIDS in Sub-Saharan Africa. Population and Development Review 15:185- 234. Caldwell, J., Orubuloye and Caldwell, P. 1992. Fertility Decline in Africa: A New Type of Transition? Population and Development Review 18:211- 242. Camp, S. 1992. Population, Pressure, Poverty and the Environment. Integration 32:24-27. Campbell, D. and Riddell, J. 1984. Social and Economic Change and the Intensity of Land Use in the Mandara Mountain Region of North Cameroon. Tijdschrift voor Economic en Sociale Geografie 75:335-343. CARE. 1994. The People ofBwindi Impenetrable National Park. Baseline Survey. Kabale, Uganda: CARE/Development Through Conservation. Carney, J. 1991. Indigenous Soil and Water Management in Senegambian Rice Farming Systems. Agricultural and Human Values VIII: 37-48. 1993. Converting • the Wetlands, Engendering the Environment: The Intersection of Gender with Agrarian Change in the Gambia. Economic Geography 69:329-348. Carter, S. 1993. Soil, Scale and Ideology: Some Thoughts on the Use of Small- Scale Soil Maps in an African Context. Paper presented at the Maps and Africa Colloquim, African Studies Group, University of Aberdeen, Aberdeen, Scotland, April 5-6. Carter, S. and Murwira, H. 1994. Spatial Variability in Soil Fertility Management and Crop Response in Mutuko Communal Area, Zimbabwe. Ambio 24:77-84. Carter, J. 1995. Alley Farming: Have Resource Poor Farmers Benefitted? Natural Resource Perspectives (ODI), No. 3, June. Casley, D. and Kuma, K. 1988. The Collection, Analysis and Use of Monitoring and Evaluation Data. Baltimore, MD: The Johns Hopkins University Press. Chambers, R. 1983. Rural Development: Putting the Last First. Harlow, Essex, U.K.: Longman Scientific and Technical. • 1994a. The Origins and Practice of Participatory Rural Appraisal. World Development 22:953-969. • 1994b. Participatory Rural Appraisal (PRA): Analysis of Experience. World Development 22:1253-1268. • 1994c. Participatory Rural Appraisal (PRA): Challenges, Potential and Paradigm. World Development 22:1437-1454. Chambers, R., Pacey, A. and Thrupp, L. 1989. Farmer First: Farmer Innovation and Agricultural Research. London: Intermediate Technology Publications. Chandler, P. 1994. Adaptive Ecology of Traditionally Derived Agroforestry in China. Human Ecology 22:415-443. Chazan, N. 1988. Patterns of State-Society Incorporation and Disengagement in Africa. In The Precarious Balance: State and Society in Africa, eds. D. Rothchild and N. Chazan, pp. 121-148. Boulder, CO: Westview Press. 283 Chazan, N. and Shaw, T. eds. 1988. Coping with Africa's Food Crisis. Boulder, CO: Lynne Reinner. Chenery, E.M. 1960. An Introduction to the Soils of the Uganda Protectorate. Memoirs of the Research Division, Series 1 (Soils), No. 1. Kampala: Department of Agriculture, Uganda Protectorate. Clarke, W. 1990. Learning from the Past: Traditional Knowledge and Sustainable Development. The Contemporary Pacific 2:233-253. Clarke, W. and Thaman, R. eds. 1993. Agroforestry in the Pacific Islands: Systems for Sustainability. Tokyo: United Nations University Press. Clay, D.C. and Lewis, L.A. 1990. Land Use, Soil Loss, and Sustainable Agriculture in Rwanda. Human Ecology 18:147-161. Clay, J. 1988. Indigenous Knowledge and Tropical Forests: Models of Land Management from Latin America. Cultural Survival Report 27. Cambridge, MA: Cultural Survival. Cleave, J. and Jones, D. 1970. Economic Background. In Agriculture in Uganda, ed. J. Jameson, pp. 109-139. Oxford: Oxford University Press. Cleaver, K. and Schrieber, G. 1992. The Population, Environment and Agriculture Nexus in Sub-Saharan Africa. Washington, DC: World Bank. Cochran W. 1977. Sampling Techniques. New York: Wiley. Cochrane, S. and Farid, S. 1989. Fertility in Sub-Saharan Africa. World Bank Discussion Paper 43. Washington, DC: World Bank. Colfer, C, and Newton, B. 1989. Ethnicity: An Important Consideration in Indonesian Agriculture. Agriculture and Human Values VL52-67. Collins, J. 1986. Smallholder Settlement of Tropical South America: The Social Causes of Ecological Destruction. Human Organization 45:1-10. Conelly, T. 1994. Population Pressure, Labor Availability, and Agricultural Disintensification: The Decline of Farming on Rusinga Island, Kenya. Human Ecology 22:145-170. Conklin, H. 1954. An Ethnoecological Approach to Shifting Cultivation. Transactions of the New York Academy of Sciences (Series 2) 17:133- 42. 1980. Ethnographic . Atlas of the lfugao: A Study of Environment, Culture, and Society in Northern Luzon. New Haven: Yale University Press. Conway, G. 1987. The Properties of Agroecosystems. Agricultural Systems 24:95-117. Coombe, A.D. and Simmons, W.C. 1933. The Volcanic Area ofBufumbira (Part I). Geological Survey of Uganda. Memoir No. 3. Entebbe: Government of Uganda. Critchley, W. 1991. Looking After Our Land. New Approaches to Soil and Water Conservation in Dryland Africa. Oxford: Oxfam Publications. Critchley, W., Reij, C. and Willcocks, T. 1994. Indigenous Soil and Water Conservation: A Review of the State of Knowledge and Prospects for Building on Traditions. Land Degradation and Rehabilitation 5:293- 314. Cunningham, A. 1992. People, Park and Plant Use: Research Recommendations for Multiple-Use Zones and Development Alternatives Around Bwindi-Impenetrable National Park, Uganda. Unpublished Report prepared for Development Through Conservation (DTC) Project, CARE, Kabale Uganda. Kampala, Uganda: CARE- International. Cunningham, A., Wild, R., Mutebi, J. and Tskeli, A. 1993. People and Wild Plant Use: Mgahinga Gorilla National Park. Unpublished Report prepared for Development Through Conservation (DTC) Project, CARE, Kabale Uganda. Kampala, Uganda: CARE-International. Current, D., Lutz, E. and Scherr, S. 1995. The Costs and Benefits of Agroforestry to Farmers. The World Bank Research Observer 10:151- 180. DAO. 1992. Soil and Water Conservation By-Laws, Kabale District. Mimeo. Kabale, Uganda: District Agricultural Office (DAO), Kabale District. 1993. Soil • and Water Conservation By-Laws, Kisoro District. Mimeo. Kisoro, Uganda: District Agricultural Office (DAO), Kisoro District. . 1994. Meteorological Records. Kabale, Uganda: Meteorological Bureau, District Agricultural Office (DAO), Kabale District. den Biggelaar, C. 1991. Farming Systems Development: Synthesing Indigenous and Scientific Knowledge Systems. Agriculture and Human Values VIII: 25-36. Denevan, W. 1983. Adaptation, Variation and Cultural Geography. The Professional Geographer 35:399-407. Denoon, D. ed. 1971. A History ofKigezi. Kampala: The National Trust. de Boef, W., Amanor, K., Wellard, K., with Bebbington, A. eds. 1993. Cultivating Knowledge: Genetic Diversity, Farmer Experimentation and Crop Research. London: Intermediate Technology Publications. de Schlippe, P. 1956. Shifting Cultivation in Africa. London: Routledge and Kegan Paul. de Wilde, J. 1967. Experiences with Agricultural Development in Tropical Africa. Vol.1: The Synthesis. Baltimore, MD: The Johns Hopkins University Press. Diamond, I. and Rutenberg, N. 1995. Recent Trends in Fertility in Botswana. Journal of International Development 7:145-161. Dommen, A. 1988. Innovation in African Agriculture. Boulder, CO: Westview Press. Doolittle, W. 1984. Agricultural Change as an Incremental Process. Annals of the American Association of Geographers 74:124-137. Doornkamp, J. 1968. The Geomorphology of Southwest Uganda. East African Geographical Review. 6:75-81. Dorsey, J., Elliot, B., Arnold, C. and Wanakwanyi, T. 1990. Soil Conservation in Uganda. Paper presented at The American Society of Agricultural Engineers, Chicago, IL, December 18-21. Dovers, S. and Handmer, J. 1992. Uncertainty, Sustainability and Change. Global Environmental Change, December, pp. 262-276. Dregne, H.E. 1982. Impact of land degradation on future world food production. Washington, DC: United States Department of Agriculture. . 1986. Soil and Water Conservation: A Global Perspective. Interciencia, 2:17-25. . 1990. Erosion and Soil Productivity in Africa. Journal of Soil and Water Conservation 4:431-436. Dudal, R. 1968. Definitions of Soil Units for the Soil Map of the World. World Soil Resources Report No. 33. Rome: Food and Agriculture Organization (FAO) of the United Nations. Dunbar, A.R. 1969. The Annual Crops of Uganda. Kampala: East African Literature Bureau. Dvorak, K. 1989. Indigenous Soil Classification in Semi-arid Tropical India. Economics Group Progress Report 84. Patancheru P.O., Andhra Pradesh, India: International Crops Research Institute for the Semi-Arid Tropics. East Africa High Commission. 1950. African Population of Uganda Protectorate. Geographical and Tribal Studies. East African Population Census, 1948. Nairobi: East African Statistical Department. . 1959. Uganda Census, 1959. African Population. Nairobi: East African Statistical Department. . 1959. Uganda Census, 1959. Non-African Population. Nairobi: East African Statistical Department Ede, K. 1981. An Analysis of Regional Inequality in Uganda. Tiidschrift voor Economische en Sociale Geographie 72:296-303. Edel, M. 1969. The Chiga of Western Uganda. London: International African Institute. Edje, O., Semoka, J. and Haule, K. 1988. Traditional Forms of Soil Fertility Maintenance. In Soil Fertility Research for Bean Cropping Systems in Africa, ed. Charles Wortmann, pp. 7-30. CIAT African Workshop Series 3. Addis Ababa, Ethiopia, September 5-9. Edwards, R. 1989. "Farmer' Knowledge: Utilization of Farmers' Soil and Land Classification". In Farmer First: Farmer Innovation and Agricultural Research, eds. R. Chambers, A. Pacey, and L. Thrupp, pp. 37-44. London: Intermediate Technology Publications. Ehrlich, P. 1968. The Population Bomb. New York: Ballantine Books. . 1970. Population, Resources, Environment: Issues in Human Ecology. San Francisco: W.H. Freeman. Ehrlich, P. and Ehrlich, A. 1990. The Population Explosion. New York: Simon and Schuster. Ehrlich, P., Ehrlich, A. and Daily, G. 1993. Food Security, Population and Environment. Population and Development Review. 19:1-32. EIU. 1995. Uganda: Country Profile, 1994-95. London: The Economist Intelligence Unit (EIU). Ellen, R. 1982. Environment, Subsistence and System: The Ecology of Small- Scale Social Formations. Cambridge: Cambridge University Press. • 1988. Persistence and Change in the Relationship Between Anthropology and Geography. Progress in Human Geography 12:229- 262. Emel, J. and Peet, R. 1989. Resource Management and Natural Hazards. In New Models in Geography: The Political Economy Perspective, eds. Richard Peet and Nigel Thrift, pp. 49-76. London: Unwin Hyman. EPL. 1995. Uganda. In Africa South of the Sahara: 1995, pp. 966-987. London: Europa Publications Limited (EPL). Erlich, C. 1965. The Uganda Economy, 1903-1945. In History of East Africa, Vol.11, eds. V. Harlow and E. Chilver, pp. 395-475. Oxford: Clarendon Press. Ewel, J. 1986. Designing Agricultural Ecosystems for the Tropics. Annual Review of Ecological Systems 17:245-271. Fairhead, J. 1990. Fields of Struggle: Towards a Social History of Farming Knowledge and Practice in a Bwisha Community, Kivu, Zaire. Ph.D. Dissertation, University of London. 1993. • Representing Knowledge: The "New Farmer" in Research Fashion. In Practising Development: Social Science Perspectives, ed. J. Pottier, pp. 187-204. London: Routledge. 288 Fairhead, J. and Leach, M. 1994. Contested Forests: Modern Conservation and Historical Land Use of Guinea's Ziama Reserve. African Affairs 93:481-512. . 1995. False Forest History, Complicit Social Analysis: Rethinking Some West African Environmental Narratives. World Development 23:1023-1035. FAO. 1977. Guidelines for Soil Profile Descriptions. 2nd Edition. Rome: Food and Agriculture Organization (FAO) of the United Nations. . 1978. Report on the Agro-ecological Zones Project. Vol.1: Methodology and Results for Africa. World Soil Resources Report No. 48. Rome: Food and Agriculture Organization (FAO) of the United Nations. . 1980. Land Resources For Populations of the Future. Rome: Food and Agriculture Organization (FAO) of the United Nations. 1983. • Guidelines: Land Evaluation for Rainfed Agriculture. FAO Soils Bulletin No. 52. Rome: Food and Agriculture Organization (FAO) of the United Nations. 1991. • Agro-Ecological Land Resources Assessment for Agricultural Development Planning. A Case Study of Kenya: Resources Data Base and Land Productivity. Main Report. Rome: Food and Agriculture Organization (FAO) of the United Nations. 1992a. The • State of Food and Agriculture: A Decade of Change. Rome: Food and Agriculture Organization (FAO) of the United Nations. • 1992b. Ninth Meeting of the East and Southern African Sub-Committee for Soil Correlation and Land Evaluation. Lilongwe, Malawi, 25 November - 2 December 1991. World Soil Resources Reports No.70. Rome: Food and Agriculture Organization (FAO) of the United Nations. 1993a. • World Soil Resources: An Explanatory Note on the FAO World Soil Resources Map at 1:25,000,000 Scale. World Soil Resources Report 66 Rev.l. Rome: Food and Agriculture Organization (FAO) of the United Nations. . and National . 1993b. Global Soils and Terrain Digital Databases: Procedures Manual. World Soil Resources Report No. 74. Rome: Food and Agriculture Organization (FAO) of the United Nations. • 1994a. FAO Production Yearbook, 1993. FAO Statistics Series No. 1 17. Rome: Food and Agriculture Organization of the United Nations. • 1994b. FAO Trade Yearbook, 1993. FAO Statistics Series No. 121. Rome: Food and Agriculture Organization of the United Nations. • 1994c. FAO Fertilizer Yearbook, 1993. Vol. 43. Rome: Food and Agriculture Organization of the United Nations. Fedele, F.G. 1984. Toward a Human Ecology of Mountains. Current Anthropology 25 : 688-69 1 Feirman, S. 1993. Defending the Promise of Subsistence: Population Growth and Agriculture in the West Usambara Mountains, 1920-1980. In Population Growth and Agricultural Change in Africa, eds. G. Hyden, R.W. Kates and B.L. Turner II, pp. 114-144. Gainesville, FL: University of Florida. Ferguson-Bisson, D. 1992. Rational Land Management in the Face of Demographic Pressure: Obstacles and Opportunities for Rural Men and Women. Ambio 21:90-94. Fiennes, R. 1939. Soil Erosion and Agricultural Planning. Uganda Journal 6:137-147. Fitzpatrick, E. 1983. Soils: Their Formation, Classification and Distribution. New York: Longman Scientific Publishing. Flury, M. 1988. Small-scale Farming and Changes of Land Use in the Highlands of Laikipia, Kenya. Mountain Research and Development 8:265-272. Foote, K., Hill, K. and Martin, L. eds. 1993. Demographic Change in Sub- Saharan Africa. Washington, DC: National Academy Press. Ford, R. 1990. The Dynamics of Human-Environment Interactions in the Tropical Montane Agrosystems of Rwanda: Implications for Economic Development and Environmental Stability. Mountain Research and Development 10:43-63. . 1992. Humans and the Sahelian Environment. National Geographic Research and Exploration, 8:460-475. . 1993. Marginal Coping in Extreme Land Pressures: Ruhengeri, Rwanda. In Population Growth and Agricultural Change in Africa, eds. G. Hyden, R.W. Kates and B.L. Turner II, pp. 145-186. Gainesville, FL: University of Florida. Fortmann, L. 1990. Women's Role In Small Farm Agriculture. In Agroecology and Small Farm Development, eds. M. Altieri and S. Hecht, pp. 35-44. Boca Raton, FL: CRC Press. Foster, H. 1971. Rapid Routine Soil and Plant Analysis Without Automatic Equipment: I. Routine Soil Analysis. East African Agricultural and Forestry Journal October: 160-165. Foster, B. 1989. Rwandese Refugees in Uganda. In Ecology and Politics, ed. A. Ornas and M. Salih, pp. 145-156. Motala, Sweden: Scandinavian Institute of African Studies. Fox, R.C. 1988. Environmental Problems and the Political Economy of Kenya: An Appraisal. Applied Geography 8:315-335. Franzel, S. 1992. Impact, Institutionalization and Methodology: Research with Farmers in Ethiopia. In Research with Farmers: Lessons from Ethiopia, eds. S. Franzel and H. Houten, pp. 243-264. Wallingford, Oxon, UK: Commonwealth Agricultural Bureau. Freeman, M. and Carbyn, L. eds. 1988. Traditional Knowledge and Renewable Resource Management in Northern Regions. Edmonton: University of Alberta. Fricke, T. 1989. Introduction: Human Ecology in the Himalaya. Human Ecology 17:131-145. Friis-Hansen, A. 1995. Seeds for Africa Peasants: Peasant's Needs and Agricultural Research - The Case of Zimbabwe. Publication No. 9. Copenhagen: Centre for Development Research. Fujisaka, S. 1987. Participation by Farmers, Researchers and Extension Workers in Soil Conservation. Gatekeeper Series SA16. London: International Institute for Environment and Development. 291 Furbee, L. 1989. A Folk Expert System: Soils Classification in the Colca valley, Peru. Anthropological Quarterly 62:83-103. Geertz, C. 1963. Agricultural Involution: The Processes of Ecological Change in Indonesia. Berkeley: University of California Press. Getahun, A. 1991. Agricultural Growth and Sustainability: Conditions for their Compatibility in the Tropical East Africa Highlands. In Agricultural Sustainability, Growth and Poverty Alleviation: Issues and Policies, eds. S. Vosti, T. Reardon and W. von Urff, pp. 451-468. Washington DC: International Food Policy Research Institute. Gibb, A. and Partners. 1967. Water Resource Survey of Uganda, 1954-55. 2nd Edition. Entebbe: Department of Lands and Survey. Gillespie, A. 1989. Potential Impact of AIDS on Farming Systems: Case Study from Rwanda. Land Use Policy 6:301-312. Glantz, M. ed. 1987. Drought and Hunger in Africa. Cambridge: Cambridge University Press. Gladwin, C. 1989a. Ethnographic Decision Tree Modeling. Newbury Park, CA: Sage. Gladwin, C. 1989b. Indigenous Knowledge Systems, The Cognitive Revolution, and Agricultural Decision Making. Agriculture and Human Values VI:32-41. Godoy, R. 1984. Ecological Degradation and Agricultural Intensification in the Andean Highlands. Human Ecology 12:359-383. Goland, C. 1993. Field Scattering as Agricultural Risk Management: A Case Study From Cuvy Cuyo, Department of Puno, Peru. Mountain Research and Development 13:317-338. Goldman, A. 1993a. Population Growth and Agricultural Change in Imo State, Southeastern Nigeria. In Population Growth and Agricultural Change in Africa, eds. B.L. Turner, G. Hyden, and R.W. Kates, pp. 250-301. Gainesville, FL: University of Florida. • 1993b. Agricultural Innovation in Three Areas of Kenya; Neo- Boserupian Theories and Regional Characterization. Economic Geography 69:44-71. 292 . 1995. Threats to Sustainability in African Agriculture. Human Ecology 23: 291-334. • 1996. Pest and Disease Hazards and Sustainability in African Agriculture. Experimental Agriculture 32:199-211. Goliber, T.J. 1985. Sub-Saharan Africa: Population Pressures on Development. Population Bulletin 40. . 1989. Africa's Expanding Population: Old Problems, New Policies. Population Bulletin 44. GOU. 1965a. Kabale. Map: East Africa (Uganda) 1:50,000. Series Y732, Sheet 93/4. Entebbe: Lands and Survey Department. . 1965b. Maziba. Map: East Africa (Uganda) 1:50,000. Series Y732, Sheet 94/3. Entebbe: Lands and Survey Department. . 1965c. Kisoro. Map: East Africa (Uganda) 1:50,000. Series Y732, Sheet 93/3. Entebbe: Lands and Survey Department. . 1965d. Rubanda. Map: East Africa (Uganda) 1:50,000. Series Y732, Sheet 93/2. Entebbe: Lands and Survey Department. . 1965e. Report on Uganda Census of Agriculture. Vol.1, 1965. Entebbe: Ministry of Agriculture and Cooperatives, Government of Uganda (GOU). . 1966. Report on Uganda Census of Agriculture. Vol.111, 1966. Entebbe: Ministry of Agriculture and Cooperatives, Government of Uganda (GOU). • 1967. Atlas of Uganda. Entebbe, Uganda: Department of Lands and Survey. . 1971. Report on the 1969 Population Census, Volume 1. Kampala: Ministry of Planning and Economic Development, Government of Uganda. • 1975. The Land Reform Decree. Entebbe: Government Printer. • 1982. Report on the 1980 Population Census, Volume 1. Kampala: Ministry of Planning and Economic Development, Government of Uganda. 293 1992a. The • 1991 Population and Housing Census. Kabale District, Uganda. (District Summary Series). Entebbe: Ministry of Finance and Economic Planning, Government of Uganda. • 1992b. The 1991 Population and Housing Census, Kisoro District, Uganda. (District Summary Series). Entebbe: Ministry of Finance and Economic Planning, Government of Uganda. . 1993a. Preliminary Estimates of Fertility and Mortality. The 1991 Population and Housing Census, Uganda. Entebbe: Ministry of Finance and Economic Planning, Government of Uganda. .1993b. Technical Report: National Biomass Study, Phase I. November 1989 - December 1991. Kampala: Forest Department, Government of Uganda. . 1994a. The 1991 Population and Housing Census, Uganda. National Summary. Entebbe: Ministry of Finance and Economic Planning, Government of Uganda. - . 1994b. State of the Environment Report for Uganda 1994. Ministry of Natural Resources. Entebbe: Government of Uganda. Greenland, D.J. and Lai, R. 1977. Soil Conservation and Management in the Humid Tropics. New York: John Wiley & Sons. Greenland, D. and Szabolcs, I. eds. 1994. Soil Resilience and Sustainable Land Use. Wallingford, Oxon, UK: Commonwealth Agricultural Bureau International. Grenzebach, K. 1981. Methods of Regional Land-Use Analysis in Southwest Uganda. Applied Geography 3:29-52. Grigg, D. 1979. Ester Boserup's theory of agrarian change: a critical review. Progress in Human Geography 3:65-83. Grisley, W. and Mwesigwa, D. 1994. Socio-economic Determinants of Seasonal Cropland Fallowing Decisions: Smallholders in Southwest Uganda. Journal of Environmental Management 42:81-90. Grosjean, M. and Messerli, B. 1988. African Mountains and Highlands: Potentials and Constraints. Mountain Research and Development 8:111- 122. Grossman, L. 1977. Man-Environment Relationships in Anthropology and Geography. Annals of the American Association of Geographers 67:126-144. . 1984. Peasants, Subsistence Ecology, and Development in the Highlands of Papua New Guinea. Princeton: Princeton University Press. Grove, A. and Sutton, J. 1989. Agricultural Terracing South of the Sahara. Azania XXIV: 113-122. Guillet, D. 1983. Toward a Cultural Ecology of Mountains: The Central Andes and the Himalayas Compared. Current Anthropology 24:561- 574. . 1987. Terracing and Irrigation in the Peruvian Andes. Current Anthropology 28:409-417. Gustafson, A. 1937. Conservation of Soil. New York: McGraw-Hill. Hakansson, T. 1989. Social and Political Aspects of Intensive Agriculture in East Africa: Some Models from Cultural Anthropology. Azania XXIV: 12-20. Hallsworth, E. 1987. Anatomy, Physiology and Psychology of Erosion. Chichester, UK: Wiley. Hamilton, A. 1969. The Vegetation of Southwest Kigezi. Uganda Journal 33:175-199. . 1982. Environmental History of East Africa. London: Academic Press. • 1984. Deforestation in Uganda. Nairobi: Oxford University Press. Hamilton, A., Taylor, D. and Vogel, J. 1986. Early Forest Clearance and Environmental Degradation in Southwest Uganda. Nature 320:164-167. • 1989. Neolithic Forest Clearance at Ahakaghezi, Western Uganda. In Quaternary and Environmental Research on East African Mountains, ed. W. Mahaney, pp. 435-463. Rotterdam: Balkema. Hanna, L. 1971. Climate and Crop Potential in Uganda. In Studies in East African Geography and Development, ed. S.H. Ominde, pp. 99- 112. Berkeley, CA: University of California Press. Hansen, A. 1994. The Illusion of Local Sustainability and Self-Sufficiency: Famine in a Border Area of Northwestern Zambia. Human Organization 53:11-20. Hansen, H.B. and Twaddle, M. 1988. Uganda Now: Between Decay and Development. London: James Curry. . 1990. Changing Uganda Now: The Dilemmas of Structural Adjustment & Revolutionary Change. London: James Curry. Hardesty, D. 1986. "Rethinking Cultural Adaptation". Professional Geographer 38:11-18. Hardin, G. 1968. The Tragedy of the Commons. Science 162:1243-1248. Hare, 1980. The Planetary Environment: Fragile or Sturdy? Geographical Journal 146:379-95. Harrison, P. 1987. The Greening of Africa. Hammondsworth, England: Penguin Books. Harrop, J. 1960. The Soils of the Western Province of Uganda. Memoirs of the Research Division, Series 1 (Soils), No.6. Kampala: Department of Agriculture, Uganda Protectorate. 24- . 1970. Geology. In Agriculture in Uganda, ed. J.D. Jameson, pp. 29. Oxford: Oxford University Press. . 1970. Soils. In Agriculture in Uganda, ed. J. Jameson, pp. 43-71. Oxford: Oxford University Press. Hecht, S. 1985. Environment, Development and Politics: Capital Accumulation and the Livestock Sector in Eastern Amazonia. World Development 13:663-684. Hecht, S. 1990. Indigenous Soil Management in the Latin American Tropics: Neglected Knowledge of Native Peoples. In Agroecology and Small Farm Development, eds. M. Altieri and S. Hecht, pp. 151-156. Boca Raton, FL: CRC Press. Hecht, S. and Posey, D. 1990. Indigenous Soil Management in the Latin American Tropics: Some Implications for the Amazon Basin. In Ethnobiology: Implications and Applications, Vol. 2, ed. D. Posey et al., pp. 73-86. Belem, Brazil: Museu Paraense Emilio Goeldi. 296 Higgins, G., Kassam, A., Naiken, L., Fischer, G. and Shah, M. 1982. Potential Population Supporting Capacities of Lands in the Developing World. Technical Report on FAO/UNFPA Project, INT 75/P13. Rome: Food and Agriculture Organization. Higgins, H., Kassam, A., Van Velthuizen, H. and Purnell, M. 1987. Methods Used by FAO to Estimate Environmental Resources, Potential Outputs of Crops, and Population Supporting Capacities in the Developing Nations. In Agricultural Environments, ed. A.B. Bunting, pp. 171-184. Wallingford, Oxon, UK: Commonwealth Agricultural Bureau International. Higgins, G., Kassam, A. and Shah, M. 1984. Land, Food and Population in the Developing World. Nature and Resources XX:2-10. Higgins, G. and Kassam, A. 1985. Land, Food and Population in Africa. In Advancing Agricultural Production in Africa, ed. D.L. Hawksworth, pp. 337-341. Farnham Royal, Slough, UK: Commonwealth Agricultural Bureau. Hill, A.R. 1987. Ecosystem Stability: Some Recent Perspectives. Progress in Physical Geography 11:315-333. Ho, T. 1990. Population Growth and Reproduction in Sub-Saharan Africa: Technical Analyses of Fertility and Its Consequences, eds. Ascadi, G., Johnson-Ascadi, G. and Bulatalo, R., pp.3 1-43. Washington, DC: World Bank. Holling, C.S. 1973. Resilience and Stability of Ecological Systems. Annual Review of Ecology and Systematics 4:1-23. and . 1986. The Resilience of Terrestrial Ecosystems: Local Surprise Global Change. In Sustainable Development of the Biosphere, eds. W.C. Clark and R.E. Munn, pp. 285-300. Cambridge: Cambridge University Press. Holling, C.S. and Bocking, S. 1987. Surprise and Opportunity: In Evolution, in Ecosystem, in Society. In Planet Under Stress, eds. C. Mungall and D.J. McClaren, Oxford: Oxford University Press. Holmes, A. and Harwood, H.F. 1937. The Volcanic Area ofBufumbira (Part II): The Petrology of the Volcanic Field ofBufumbira, Southwest Uganda. Geological Survey of Uganda, Memoir No. 3. Entebbe: Government of Uganda. 297 Holmgren, P., Masakha, E. and Sjoholm, H. 1994. Not All African Land is Being Degraded: A Recent Survey of Trees on Farms in Kenya Reveals Rapidly Increasing Forest Resources. Ambio 23:390-395. Hopkins, E. 1968. The Politics of Conquest: The Pacification of Kigezi District. Ph.D. Dissertation, Columbia University. Hoskins, M. 1984. Observations on Indigenous and Modern Agroforestry in West Africa. In Social, Economic and Institutional Aspects of Agroforestry, ed. J. Jackson, pp. 46-50. Tokyo: The United Nations University. Hudson, N. 1976. Soil Conservation. London: B.T. Batsford. . 1992. Success and Failure of Soil Conservation Programs. In Soil Conservation for Survival, eds. K. Tato and H. Hurni, pp. 129-142. Ankeni, IW: Soil and Water Conservation Society. Hughes, R. and Hughes, J. 1992. A Directory of African Wetlands. Gland, Switzerland: International Union for the Conservation of Nature (IUCN). Hunter, J. 1966. Ascertaining Population Carrying Capacity Under Traditional Systems of Agriculture in Developing Countries. Professional Geographer 18:151-4. Hyde, R. 1975. The Analysis of Crop Distributions in Uganda. Department of Geography. Occasional Paper No. 65. Kampala: Makerere University. Hyden, G. 1986. The Invisible Economy of Smallholder Agriculture in Africa. In Understanding Africa's Rural Households and Farming Systems, ed. J. Moock, pp. 11-35. Boulder: Westview Press. Hyden, G., Kates, R. and Turner, B.L. II, eds. 1993. Beyond Intensification. In Population Growth and Agricultural Change in Sub-Saharan Africa, eds. B.L. Turner II, G. Hyden, R. Kates pp. 401-439. Gainesville, FL: University of Florida Press. ICRAF. 1987. Agroforestry Potentials for the Land-Use Systems in the Bi- modal Highlands of East Africa: Uganda. AFRENA Report No. 4. Nairobi: International Center for Research on AgroForestry (ICRAF). 1993. Integrated . Natural Resource Management Research for the Highlands of East and Central Africa. Consultative Workshop Report. Nairobi: International Center for Research on AgroForestry (ICRAF). IFAD. 1992. Soil and Water Conservation in Sub-Saharan Africa: Towards Sustainable Production by the Rural Poor. Report prepared for the International Fund for Agricultural Development (IFAD), Rome. Amsterdam: Center for Development Cooperative Services, Free University. Igbozurike, M. 1970. Fragmentation in Tropical Africa: An Overrated Phenomenon. Professional Geographer 22:132-135. Inglis, J. ed. 1993. Traditional Ecological Knowledge: Concepts and Cases. Ottawa: Canadian Museum of Nature/International Development Research Center. Ives, J.D. and Messerli, B. 1989. The Himalaya Dilemma: Reconciling Development and Conservation. New York: Routledge. Ives, J.D. and Pitt, D.C. eds. 1988. Deforestation: Social Dynamics in Watersheds and Mountain Ecosystems. London: Routledge. Jack, E. 1914. On the Congo Frontier. London: Fisher-Unwin. Jacks, G. and Whyte, R. 1938. The Rape of the Earth. London: Faber and Faber. Jaeger, W. 1992. The Causes of Africa's Food Crisis. World Development 20:1631-1646. Jagannathan, N. Mori, H. and Hassan, H. 1990. Applications of Geographical Information Systems in Economic Analysis: A Case Study of Uganda. Environment Working Paper 27. Washington, DC: World Bank. Jamal, V. 1978. Taxation and Inequality in Uganda, 1900 - 1964. Journal of Economic History 18:418-438. • 1991. The Agrarian Context of the Ugandan Crisis. In Changing Uganda, ed. H. Hansen and M. Twaddle, pp. 78-97. London: James Curry. Jameson, J.D. ed. 1970. Agriculture in Uganda. Oxford: Oxford University Press. Jameson, J.D. and McCallum, D. 1970. Climate. In Agriculture in Uganda, ed. J. Jameson, pp. 12-23. Oxford: Oxford University Press. Jarosz, L. 1993. Defining and Explaining Tropical Deforestation: Shifting Cultivation and Population Growth and Population Growth in Colonial Madagascar (1896-1940). Economic Geography 69:366-379. Jiggins, J. 1989. An Examination of the Impact of Colonialism in Establishing Negative Values and Attitudes Towards Indigenous Agricultural Knowledge. In Indigenous Knowledge Systems for Agricultural and International Development, eds. D. Warren, D. Brokensha and L. Slikkerveer, pp. 68-78. Studies in Technology and Social Change, No. 1 1 . Ames, IW: Iowa State University. Jodha, N. 1990. Mountain Agriculture: The Search for Sustainability. Journal of Farming Systems Research-Extension 1:55-76. . 1992. Common Property Resources: A Missing Dimension of Development Strategies. Washington, DC: World Bank. Jodha, N., Bankota, S. and Partap, T. 1992. Sustainable Mountain Agriculture: Farmer's Strategies and Innovative Approaches. Vol. 1 & 2. New Delhi: Oxford and IBH Publishing. Johannes, R. ed. 1989. Traditional Ecological Knowledge: A Collection of Essays. Gland, Switzerland: International Union for Conservation of Nature and Natural Resources. Johnson, D. and Lewis, L. 1995. Land Degradation. Cambridge, MA: Blackwell. Johnson, M. ed. 1993. Lore: Capturing Traditional Knowledge Environmental Knowledge. Ottawa: International Development Research Center (IDRC) Books. Johnston, R., Gregory, D. and Smith, D. 1994. The Dictionary of Human Geography. Oxford: Basil Blackwell. Jolly, C. 1994. Four Theories of Population Change and the Environment. Population and Environment: A Journal of Interdisciplinary Studies 16:61-90. Jones, J. and Wallace, B. 1986. Social Science in Farming Systems Research. In Social Sciences and Farming Systems Research, eds. J. Jones and B. Wallace, pp. 1-22. Boulder, CO: Westview. Jones, W. and Egli, R. 1984. Farming Systems in Africa: The Great Lakes Highlands of Zaire, Rwanda and Burundi. Washington, DC: World Bank. Jorgensen, J. 1991. Uganda: A Modern History. New York: St. Martin's Press. Jurion, F. and Henry, J. 1969. Can Primitive Farming Can be Modernized? Brussels: INEAC/Belgian Cooperative and Development Office. Kabera, J. 1982. Rural Population Redistribution in Rural Uganda since 1900. In Redistribution of Population in Africa, eds. J. Clarke and L Kosinski, pp. 192-201. London: Heinemann. Kagambirwe, E. 1972. Causes and Consequences of Land Shortage in Kigezi. Department of Geography Occasional Paper No. 23. Kampala: Department of Geography, Makerere University. . 1984. Land Settlement in Uganda: A Comparative Study of Rural Migrations from Kigezi and Bugisu. Ph.D. Dissertation, Makerere University. Kaijuka, E., Kaija, E., Cross, A. and Loaiza, E. 1989. Uganda Demographic and Health Survey, 1988/1989. Ministry of Health, Government of Uganda. Columbia, MD: Institute for Resource Development/Macro Systems. Kakuru, A. 1993. The Potential Use of Contour Hedges for Soil Erosion Control in the Highlands ofKabale District, Southwest Uganda. Masters Thesis, Environment and Natural Resources. Kampala: Makerere University. Kamugisha, J. 1993. Management of Natural Resources and Environment in Uganda. Policy and Legislation Landmarks, 1890 - 1990. RSCU Report No. 11. Nairobi: SIDA and Regional Soil Conservation Unit (RSCU). Karani, P. 1982. Deforestation and Its Deleterious Effects on the Human Environment. Uganda Journal 40:1-13. Kakarikawe, E. and Yeld, R. 1966. Agricultural Development and Resettlement Schemes in Kigezi, Western Uganda. World Bank Report. Kateete, B. 1976. Land Tenure and Land Use in Hamurwa Parish, Ikumba, Rubanda. In Essays on Land Fragmentation in Kigezi District, eds. R. Tindituuza and B. Kateete, pp. 47-93. Department of Geography Occasional Paper No. 32. Kampala: Department of Geography, Makerere University. Kates, R., Hyden, G. and Turner II, B.L. 1993. Theory, Evidence and Design. In Population Growth and Agricultural Change in Africa, eds. G. Hyden, R.W. Kates and B.L. Turner II, pp. 1-40. Gainesville, FL: University of Florida. Kebedde, G. and Jacobs, M. 1988. Drought, Famine and the Political Economy of Environmental Degradation in Ethiopia. Geography 73:65- 70. Kelley, H. 1983. Keeping the Land Alive: Soil Erosion - Its Causes and Cures. FAO Soils Bulletin, No. 50. Rome: Food and Agriculture Organization (FAO) of the United Nations. Kisakye, J. 1987. A Diagnostic Survey of Kabale District. In Bean Research in Eastern Africa, CIAT African Workshop Series, No. 2, ed. R.A. Kirkby, pp. 53-62. Mukono, Uganda, June 22-25. Kisitu, V. 1991. Fertilizer Material Occurrences in Uganda. Fertilizer Research 30:187-189. Kloppenburg, Jr., J. 1991. Social Theory and the De/Reconstruction of Agricultural Science: Local Knowledge for an Alternative Agriculture. Rural Sociology 56:519-548. Knight, C. 1974. Ecology and Change: Rural Modernization in an African Community. New York: Academic Press. . 1980. Ethnoscience and the African farmer: Rationale and Strategy. In Indigenous Knowledge Systems and Development, eds. D. Brokensha, D. Warren, and O. Werner, pp. 205-231. Lanham, MD: University Press of America. 302 Krokfors, C. 1989. Population and Land Degradation: A Political Ecology Approach. In Ecology and Politics: Environmental Stress and Security in Africa, eds. H. Anders and M. Salih, pp. 197-209. Uppsala: Scandinavian Institute of African Studies. Kururagire, A. 1969. Land Fragmentation at Rugarama, Kigezi. Uganda Journal 33:59-64. Lai, R. 1985. Soil Erosion and Its Relation to Productivity in Tropical Soils. In Soil Erosion and Conservation, eds. S. El-Swaify, W. Middenhaurer and A. Lo, pp. 27-36. Ankeny, IA: Soil Conservation Society of America. . 1987. Managing the Soils of Sub-Saharan Africa. Science 236: 1069- ' 1076. . 1988a. Soil Degradation and the Future of Agriculture in Sub-Saharan Africa. Journal of Soil and Water Conservation 43:444-451. . 1988b. Monitoring Soil Erosion's Impact on Crop Productivity. In Soil Erosion Research Methods, ed. R. Lai, pp. 126-137. Ankeny, IA. . 1990. Low-Resource Agriculture Alternatives in Sub-Saharan Africa. Journal of Soil and Water Conservation 45:437-445. . 1994. Sustainable Land Use Systems and Soil Resilience. In Soil Resilience and Sustainable Land Use, eds. D. Greenland and I. Szabolcs, pp. 41-67. Wallingford, Oxon, UK: Commonwealth Agricultural Bureau International. Lai, R. and Sanchez, P. ed. 1992. Myths and Science of Soils of the Tropics. SSSA Special Publication Number 29. Madison, WI: Soil Science Society of America and America Society of Agronomy. Landon, J. 1991. Booker Tropical Soil Manual. New York: John Wiley & Sons, and Longman Scientific and Technical. Langlands, B. 1971. The Population Geography of Kigezi. Department of Geography, Occasional Paper No. 2. Kampala: Department of Geography, Makerere University. Soil . 1974a. Productivity Map of Uganda. In Soil Productivity and Land Availability Studies, ed. B. Langlands, pp. 1-26. Department of Geography, Occasional Paper No. 54. Kampala: Department of Geography, Makerere University. . 1974b. Cultivable Land: Population Ratios, 1959 and 1969. In Soil Productivity and Land Availability Studies, ed. B. Langlands, pp. 55- 69. Department of Geography, Occasional Paper No. 54. Kampala: Department of Geography, Makerere University. Langlands, B. and Mbakyenga, S. 1974. Soil Classification by Counties. In Soil Productivity and Land Availability Studies, ed. B. Langlands, pp. 27-53. Department of Geography, Occasional Paper No. 54. Kampala: Department of Geography, Makerere University. Lee, R., Arthur, W., Kelley, A., Rodgers, G. and Srinivsan, T., eds. 1988. Population, Food and Rural Development. Oxford: Clarendon Press. Lele, U. and Stone, S. 1989. Population Pressure, The Environment and Agricultural Intensification. MADIA Discussion Paper 4. Washington, DC: World Bank. Lesthaege, R. and Jolly, C. 1995. The Start of the Fertility Transition: Some Answers and Many Questions. Journal of International Development 7:25-45. Lewis, L. 1992. Terracing and Accelerated Soil Loss on Rwandan Steeplands: A Preliminary Investigation of the Implications of Human Activities Affecting Soil Movement. Land Degradation and Rehabilitation 3:241- 246. Lewis, H. 1993. Traditional Ecological Knowledge: Some Definitions. In Traditional Ecological Knowledge: Wisdom for Development, eds. N. Williams and G. Baines, pp. 8-12. Canberra: Centre for Resource and Environmental Studies, Australian National University. Lindblade, K., Tumhairwe, J., Carswell, G., Nkwiine, C, and Bwamiki, D. 1996. More People, More Fallow: The Myth of Overcultivation in Kabale District, Uganda. Report to the Rockefeller Foundation, Nairobi, Kenya. Lindsey, J. and Gumbs, F. 1982. Erodibility Indices Compared to Measured Values of Selected Trinidad Soils. Soil Science Society ofAmerica Journal 46:393-396. Lindskog, P. and Tengberg, A. 1994. Land Degradation, Natural Resources and Local Knowledge in the Sahel Zone of Burkina Faso. GeoJournal 33:365-375. Little, P. and Horowitz, M. 1987. Introduction: Social Science Perspectives on Land, Ecology and Development. In Lands at Risk in the Third World, eds. P. Little and M. Horowitz, pp. 1-15. Boulder, CO: Westview Press. Livingstone, D. 1984. Interactions of Food Production and Changing Vegetation in Africa. In From Hunters to Farmers, eds. J. Clark and S. Brandt, pp. 22-25. Berkeley: University of California Press. Lockwood, M. 1995. Demographic Policy and the African Demographic Transition: Issues and Questions. Journal of International Development 7:1-24. Locoh, T. and Hertich, V. eds. 1994. The Onset of Fertility Transition in Sub- Saharan Africa. Leige: Ordina Editions. Logan, B. 1991. Overpopulation and Poverty in Africa: Rethinking the Traditional Relationship. Tijdschrift voor Economic en Sociale Geographie 82:40-57. Lubanga, L. 1988. Effects of the Traditional Cropping System on Soil Fertility in South Kivu, Zaire. In Soil Fertility Research for Bean Cropping Systems in Africa, ed. Charles Wortmann, pp. 7-30. CIAT African Workshop Series 3. Addis Ababa, Ethiopia, September 5-9. Ludwig, H. 1968. Permanent Farming on Ukara: The Impact of Land Shortage on Husbandry Practices. In Smallholder Farming and Smallholder Development in Tanzania, ed. H. Ruthenberg, pp. 87-136. Munich: Weltforum Verlag. Lury, D. 1976. Dayspring Mishandled? The Uganda Economy, 1945 - 1960. In History of East Africa, Vol.111, eds. D. Low and A. Smith, pp. 212- 250. Oxford: Clarendon Press. Lutz, E., Pagiola, S. and Reiche, C. 1994. The Costs and Benefits of Soil Conservation: The Farmer's Viewpoint. The World Bank Research Observer 9:273-295. Mabogunje, A. 1995. The Environmental Challenges in Sub-Saharan Africa. Environment 37:4-9, 31-35. 305 Magunda, M.K. 1992. Influence of Some Physico-Chemical Properties on Soil Strength, Stability of Crusts and Soil Erosion. Ph.D. Dissertation, University of Minnesota. Malthus, T.R. 1960. On Population. First Essay on Population, 1798, and Second Essay on Population, 1803. New York: Modern Library and Random House. Maro, P. 1988. Agricultural Land Management Under Population Pressure: The Kilamanjaro Experience. Mountain Research and Development 8:273-282. Martin, W. 1944. Soil Structure. East African Agricultural Journal IX: 189. . 1945a. Notes on Soils and Soil Conservation for the Kigezi Committee. Unpublished Report of the District Agricultural Office. Kabale, Kigezi District, Uganda: District Agricultural Office (DAO). - . 1945b. Kigezi Committee Results of Soil Examination. Unpublished Report of the District Agricultural Office. Kabale, Kigezi District, Uganda: District Agricultural Office (DAO). . 1946. The Kigezi Overpopulation Problem. In Uganda Department of Agriculture: Annual Report, 1944-45, pp. 15-23. Entebbe: Uganda Protectorate. Martin, A. 1963. The Marketing of Minor Crops in Uganda. Department of Technical Co-operation. Overseas Research Publication No.l. London: Her Majesty's Stationery Office. Martin, G. and Vityakon, P. 1986. Soil Management in Traditional Agriculture. In Traditional Agriculture in Southeast Asia: A Human Ecology Perspective, ed. G. Martin, pp. 199-225. Boulder, CO: Westview Press. Martin, S. 1987. Boserup Revisited: Population and Technology in African Agriculture, 1900-1940. Journal of Imperial and Commonwealth Agriculture 16:109-23. Mascarenhas, O. and Veit, P. 1995. Indigenous Knowledge in Resource Management: Irrigation in Msanzi, Tanzania. Washington, DC: World Resources Institute. Masefield, G. 1962. Agricultural Change in Uganda, 1945 - 1960. Stanford: Food Research Institute. Mateke, P. 1970. The struggle for dominance in Bufumbira, 1830-1920. Uganda Journal, 34:35-57. Matheson, J. and Bovill, E.W. 1950. East African Agriculture. London: Oxford University Press. Matlon, P. and Spencer, D. 1986. Increasing Food Production in Africa. American Journal of Agricultural Economics 66:671-676. Mausbasch, M. and Wilding, L., eds. 1991. Spatial Variabilities of Soils and Landforms. SSSA Special Publication No. 28. Madison, WI: Soil Science Society of America. McCall, M. 1993. Indigenous Technical Knowledge in Farming Systems of Eastern Africa: A Bibliography. Working Paper No. 67. Enschede, The Netherlands: Universiteit Twente. . 1996. ITK in East African Farming Systems. Indigenous Knowledge and Development Monitor 4:20-22. McCorkle, C. 1989a. Toward A Knowledge of Local Knowledge and Its Importance for Agricultural RD&E. Agriculture and Human Values VL4-12. . 1989b. Veterinary Anthropology. Human Organization 48: 156-162. McMaster, D. 1962. A Subsistence Crop Geography of Uganda. Cornwall, UK: Geographical Publications. . 1966. Changes in the Extent and the Distribution of Cultivation in Uganda, 1952 - 1958. Uganda Journal 30:63-74. . 1968. Uganda: Initiatives in Agriculture, 1888-1966. In Land Use and Land Resources, ed. IBG, pp. 241-258. London: Institute of British Geographers (IBG). Meadows, D.H., Meadows, D.L., Randers, J. and Behrens III, W. 1974. The Limits to Growth. A Report for the Club of Rome's Project on the Predicament of Mankind. New York: Universe Books. Meelu, O., Singh, Y. and Singh, B. 1994. Green Manuring for Soil Productivity Improvement. World Soil Resources Report 76. Rome: Food and Agriculture Organization of the United Nations. Messerli, B., Hurni, H., Wolde-Semayat, B., Tedla, S., Ives, J. and Wolde- Mariam, M. 1988. African Mountains and Highlands: Introductions and Resolutions. Mountain Research and Development 8:93-100. Messerschmidt, D. 1984. More on a Cultural Ecology of Mountains. Current Anthropology 25:357-358. Middleton, H. 1930. Properties of Soils Which Influence Soil Erosion. USDA Technical Bulletin No. 178. Washington, DC: USDA. Miller, N. and Rockwell, R. 1988. AIDS in Africa. Lewiston, NY: Edwin Mellen. Millington, A. 1987a. Soil Conservation and Shifting Cultivators. ILEIA Newsletter 3:15-16. . 1987b. Soil Conservation and Shifting Cultivators. ILEIA Newsletter 3:18-19. . 1987c. Environmental Degradation, Soil Conservation and Agricultural Policies in Sierra Leone, 1895-1984. In Conservation in Africa, eds. D. Anderson and R. Grove, pp. 229-248. Cambridge: Cambridge University Press. . 1992. Soil Erosion and Conservation. In Environmental Issues in the 1990's, eds. A. Mannion and S. Bowlby, pp. 236-244. New York: John Wiley & Sons. Millington, A., Mutiso, S., Kirkby, J. and O'Keefe, P. 1989. African Soil Erosion-Nature Undone and the Limitation of Technology. Land Degradation and Rehabilitation 1:279-290. Milne, G. 1935. Some Suggested Units of Classification and Mapping, Particularly for East African Soils. Soil Research 4:183-198. . 1936. A Provisional Soil Map of East Africa (Kenya, Uganda, Tanganyika and Zanzibar), With Explanatory Memoir. London: Amani Memoirs. Miracle, M. 1967. Agriculture in the Congo Basin. Madison, WI: University of Wisconsin Press. Mitchell, A. 1985. Human Carrying Capacity in Existing Farming Systems. In Advancing Agricultural Production in Africa, ed. D.L. Hawksworth, pp. 361-367. Farnham Royal, Slough, UK: Commonwealth Agricultural Bureau International. Moldenhauer, W. and Hudson, N. eds. 1988. Conservation Farming on Steep Slopes. Ankeny, IA: Soil and Water Conservation Society. Moore, D. 1993. Contesting Terrain in Zimbabwe's Eastern Highlands: Political Ecology, Ethnography and Peasant Resource Struggles. Economic Geography 69:380-401. Moran, E. 1982. Human Adaptability: An Introduction To Ecological Anthropology. Boulder, CO: Westview. . 1990. The Ecosystem Approach in Anthropology. Boulder, CO: Westview. . 1991. Human Adaptive Strategies in Amazonian Blackwater Ecosystems. American Anthropologist 93:361-382. Morgan, W. 1969. Peasant Agriculture in Tropical Africa. In Environment and Land Use in Africa, eds. M. Thomas and G. Whittington, pp. 241-272. London: Metheun. Morgan, W. and Solarz, J. 1994. Agricultural Crisis in Sub-Saharan Africa: Development Constraints and Policy Problems. Geographical Journal 160:57-73. Morrison, M. 1968. Vegetation and Climate in the Uplands of Southwest Uganda During the Later Pleistocene Period, 1. Muchoya Swamp, Kigezi. Journal of Ecology 56:363-377. Morrison, M. and Hamilton, A. 1974. Vegetation and Climate in the Uplands of Southwest Uganda During the Late Pleistocene Period, 2. Forest Clearance and Other Vegetational Changes in the Rukiga Highlands During the Past 8,000 Years. Journal of Ecology 62:1-31. Mortimore, M. 1989. Adapting to Drought: Farmers, Famines and Desertification in West Africa. Cambridge: Cambridge University Press. . 1993. Population Growth and Land Degradation. GeoJournal 31:15-21. Mortimore, M. and Tiffen, M. 1994. Population Growth and a Sustainable Environment. Environment 36:11-20, 28-32. Moss, R., ed. 1968. Soils, Slopes and Surfaces in Africa. In The Soil Resources of Tropical Africa, ed. R. Moss, pp. 29-60. Cambridge: Cambridge University Press. Moss, R. and Rathbone, R. 1975. The Population Factor in African Studies. New York: Africana. Motoyoshi, O. 1973. "Modernization in a Rural Kiga Community". The Japanese Journal of Ethnology 38:12-21. Mudoola, D. 1993. Religion, Ethnicity and Politics in Uganda. Kampala: Fountain Publishers. Mugisha, R. 1992. Emergent Changes and Trends in Land Tenure and Land Use in Kabale and Kisoro Districts. CBR Working Paper No. 26. Kampala: Centre for Basic Research (CBR). Mukasa, S. and Thomas, D. 1970. Staple Food Crops. In Agriculture in Uganda, ed. J. Jameson, pp. 140-153. Oxford: Oxford University Press. Mulongoy, K. and Merckx, R. eds. 1993. Soil Organic Matter Dynamics and Sustainability of Tropical Agriculture. Chichester, West Sussex, UK: John Wiley. Muller-Boker, U. 1991. Knowledge and Evaluation of the Environment in Traditional Societies of Nepal. Mountain Research and Development 11:101-114. Murindwa, R. 1991. Nyabingi Movement: People's Anti-Colonial Struggles in Kigezi, 1910-1930. CBR Working Paper No. 18. Kampala: Centre for Basic Research (CBR). Murti, A. and Hinde, P. 1995. Recent Demographic Changes in Tanzania: Causes, Consequences and Future Prospects. Journal of International Development 7:117-134. Musere, J. 1990. African Sleeping Sickness: Political Ecology, Colonialism and Control in Uganda. Lewiston, NY: E. Mellen Press. 310 Mutibwa, P. 1990. Uganda Since Independence: A Story of Unfulfilled Hopes. Kampala: Fountain Publishers. Mwaka, V. 1978. Agricultural Marketing Co-operatives in Uganda. Department of Geography Occasional Paper No. 69. Kampala: Makerere University. National Research Council. 1996. Lost Crops of Africa. Volume 1: Grains. Washington, DC: National Academy Press. Napier, T., Camboni, S. and El-Swaify, S. eds. 1994. Adopting Conservation on the Farm: An International Perspective on the Socioeconomics of Soil and Water Conservation. Ankeny, IA: Soil and Water Conservation Society. Ndiaye, S. and Sofranko, A. 1994. Farmer's Perceptions of Resource Problems and Adoption of Conservation Practices in a Densely Populated Area. Agriculture, Ecosystems and Environment 48:35-47. Neimeijer, D. 1995. Indigenous Soil Classifications: Complications and Considerations. Indigenous Knowledge and Development Monitor 3:20- 21. . 1996. The Dynamics of African Agricultural History: Is it Time for a New Development Paradigm? Development and Change 27:87-110. Netting, R. 1968. Hill Farmers of Nigeria: Cultural Ecology of the Kyofar of the Jos Plateau. Seattle, WA: University of Washington Press. • 1986. Cultural Ecology. Prospects Heights, IL: Waveland Press. 1993. • Householders, Smallholders: Farm Families and the Ecology of Intensive Sustainable Agriculture. Stanford, CA: Stanford University Press. Netting, R., Stone, M. P. and Stone, G. 1989. Kofyar Cash-Cropping: Choice and Change in Indigenous Agricultural Development. Human Ecology 17:299-319. Ngabirano, F. 1993. Land Degradation on Kabale Slopes. The New Vision (Newspaper). Wednesday, May 26; pp. 12-13. Ngologoza, P. 1969. Kigezi and Its People. Nairobi: East African Literature Bureau. 311 Nicholaides III, J. and Moran, E. 1995. Soil Indices for Comparative Analysis of Agrarian Systems. In The Comparative Analysis of Human Societies: Toward Common Standards for Data Collection and Reporting, ed. E. Moran, pp. 39-54. Boulder: Lynne Reinner. Nsibambi, A. 1988. Solving Uganda's Food Problem. In Uganda Now, eds. H. Hansen and M. Twaddle, pp. 135-157. London: James Curry. Nyamaguru, V. 1992. Some Ecological Aspects of the Kigeyo Swamp: A Highland Wetland Ecosystem In Rubanda County, Kabale District, In Southwestern Uganda. MS Thesis, Makerere University, Kampala, Uganda. Obol-Ochola, J. 1971. Customary Law and the Economic Development of Uganda. LL.M. Dissertation, University of Dar es Salaam. Okafor, F. 1987. Population Pressure and Land Resource Depletion in Southeastern Nigeria. Applied Geography 7:243-256. Okereke, O. 1974. The Economic Impact of Uganda Co-operatives. Nairobi: East Africa Literature Bureau. Okorio, J. et al. 1988. Agroforestry Research Project Proposal for the Kigezi Annual Montane Food Crop System in the Highlands of Uganda, AFRENA Report No. 11. Nairobi: International Centre for Research in AgroForestry. Oldeman, L. 1994. The Global Extent of Soil Degradation. In Soil Resilience and Sustainable Land Use, eds. D. Greenland and I. Szabolcs, pp. 99- 118. Wallingford, Oxon, UK: Commonwealth Agricultural Bureau International. Oilier, C, Lawrance, C, Beckett, P., and Webster, R. 1969. Land Use Systems of Uganda. (MEXE Report No. 59). Christchurch, Hampshire, UK: Military Engineering Experimental Establishment (MEXE). Onuhua, N. and Timeous, I. 1995. Has a Fertility Decline Begun in West Africa?. Journal of International Development 7:93-115. Opio-Odongo, J. 1992. Designs on the Land: Agricultural Research in Uganda, 1890 - 1990. Nairobi: African Centre for Technology Studies Press. 312 Orlove, B. 1980. Ecological Anthropology. Annual Review of Anthropology 9:235-273. Ostium, E. 1990. Governing the Commons. Cambridge, MA: Cambridge University Press. Osunade, M. 1988. Soil suitability classification by small farmers. Professional Geographer 40:194-201. • 1992a. Land Management in Nigeria. Journal of Farming Systems Research-Extension 1:131-144. . 1992b. Some Aspects of Land Management by Small Farmers of Southwestern Nigeria. Journal of Farming Systems Research-Extension 3:131-144. . 1994. Indigenous Grass Ecology and Socio-Economic Values in Swaziland. Journal of Environmental Management 41:283-292. Pahlman, C. 1991. Soil Erosion? That's Not How We See the Problem! ILEIA Newsletter l&2:24-26. Park, D. ed. 1993. Risk and Tenure in Arid Lands: The Political Ecology of Development in the Senegal River Basin. Tucson, AZ: University of Arizona Press. Parsons, D. 1960. The Montane Systems. The Systems of Agriculture Practised in Uganda, No. 4. Memoirs of the Research Division, Series 3, Department of Agriculture. Entebbe: Department of Agriculture, Uganda Protectorate. • 1970. Agricultural Systems. In Agriculture in Uganda, ed. J.D. Jameson, pp. 127-138. Oxford: Oxford University Press. Pawluk, R., Sandor, J. and Tabor, J. 1992. The Role of Indigenous Soil Knowledge in Agricultural Development. Journal of Soil and Water Conservation 47:298-302. Peden, D. and Kakuru, A. 1993. Hedgerows for Control of Soil Erosion in Kabale, Southwest Uganda. Paper presented at the Workshop on Land and Water Management: Towards Sustainable Land Use, Nairobi, Kenya, February 15-19. 313 Peden, D., Aluma, J., Okorio, J., Byenka, S. Wajja-Musukwe, N. and Muwanga, J. 1991. AFRENA Project: Uganda Progress Report for the Period Ending March, 1991. AFRENA Report Series No. 43. Nairobi: International Center for Research on AgroForestry (ICRAF). Peluso, N. 1992a. The Political Ecology of Extraction and Extractive Reserves in East Kalimantan, Indonesia. Development and Change 23: 49-74. Peluso, N. 1992b. Rich Forests, Poor People. Berkeley: University of California Press. Perrot-Maitre, D. and Weaver, T. 1992. Indigenous Knowledge and Fertilizer Strategies in Leyte, Philippines: Implications for Research and Demonstration Trials. Journal of Farming Systems Research-Extension 3: 21-32. Phillips, J. 1959. Agriculture and Ecology in Africa: A Study of Actual and Potential Development South of the Sahara. London: Faber and Faber. Pingali, P., Bigot, Y. and Binswanger, H. 1987. Agricultural Mechanization and the Evolution of Farming Systems in Sub-Saharan Africa. Baltimore, MD: Johns Hopkins University Press. Pingali, P. and Binswanger, H. 1988. Population Density and Farming Systems: The Changing Locus of Innovations and Technical Change. In Population, Food and Rural Development, eds. R. Lee et al., pp. 51- 76. Oxford: Oxford University Press. Pingali, P., Binswanger, H. and Bigot, Y. 1987. Agricultural Mechanization and the Evolution of Farming Systems in Sub-Saharan Africa. Baltimore, MD: Johns Hopkins University Press. Pitalbo, J. 1981. The North Mkata Plain, Tanzania: A Study of Land Capability and Land Tenure. Toronto: University of Toronto Press. Place, F. and Hazell, P. 1993. Productivity Effects of Indigenous Land Tenure Systems in Sub-Saharan Africa. America Journal of Agricutural Economics February, 10-19. Poponoe, H. 1986. African Soils: Opportunities and Constraints. In Food in Sub-Saharan Africa, ed. A. Hansen and D. McMillan, pp. 169-176. Boulder, CO: Lynne Reinner. . Porter, P. 1965. Environmental Potentials and Economic Opportunities: A Background for Cultural Adaptation. American Anthropologist 67:409- 420. . 1970. The Concept of Environmental Potential as Exemplified by Tropical Africa Research. In Geography and a Crowding World, eds. W. Zelinsky, L. Kosinski and R. Prothero, pp. 187-217. New York: Oxford University Press. . 1978. Geography as Human Ecology. A Decade of Progress in a Quarter Century. American Behavioral Scientist 22:15-39. . 1984. Food and Development in the Semi-Arid Zone of East Africa. Syracuse, NY: Maxwell School of Citizenship and Public Affairs, Syracuse University. . 1991. Cultural Ecology. In Modern Geography: An Encyclopedic Survey, ed. Gary S. Dunbar, pp. 38-39. NY: Garland. Posey, D. 1983. Indigenous Ecological Knowledge and Development. In The Dilemma ofAmazonian Development, ed. E. Moran, pp. 127-142. Boulder, CO: Westview Press. . 1993. Indigenous Knowledge in the Conservation and Use of World Forests. In World Forests for the Future: Their Use and Conservation, eds. K. Ramakrishna and G. Woodwell, pp. 59-77. New Haven: Yale University Press. Posey, D. and Balee, W. eds. 1989. Resource Management in Amazonia: Indigenous and Folk Strategies. Bronx, New York: The New York Botanical Garden. Powesland, P. 1954. History of Migration in Uganda. In Economic Development and Tribal Change: A study of immigrant labor in Buganda, ed. A. Richards, pp. 17-51. Cambridge: W. Heffer and Sons. Pretty, J. 1995. Participatory Learning for Sustainable Agriculture. World Development 23:1 247- 1 263 Pretty, J., Thompson, J. and Kiara, J. 1995. Agricultural Regeneration in Kenya: The Catchment Approach to Soil and water Conservation. Ambio 24:7-15. Purseglove, J. 1945. Report on the Kigezi Overpopulated Areas. Unpublished Report of the District Agricultural Office. Kabale, Kigezi District, Uganda: District Agricultural Office (DAO). . 1946. Land Use in the Overpopulated Areas of Kabale, Kigezi District, Uganda. East African Agriculture and Forestry Journal, 12:3-10. . 1947. Unpublished Memorandum of the District Agricultural Office. Kabale, Kigezi District, Uganda: District Agricultural Office (DAO). . 1950. Kigezi Resettlement. Uganda Journal 14:139-152. . 1951. Re-Settlement in Kigezi, Uganda. Journal of African Administration 3:13-21. Raikes, P. 1988. Modernising Hunger: Famine, Food Surplus and Farm Policy in the EEC and Africa. London: James Currey. Randolph, S.L. and Sanders, R. 1992. Female Farmers in the Rwandan Farming System: A Study of the Ruhengeri Prefecture. Agriculture and Human Values IX:59-66. Redclift, M. 1987. Sustainable Development: Exploring the Contradictions. London: Metheun. Reij, C. 1991. Indigenous Soil and Water Conservation in Africa. Gatekeepers Series 27. London: International Institute for Environment and Development. Repetto, R. 1987. Population, Resources, Environment: An Uncertain Future. Population Bulletin 42. Rhoades, J. 1982. Soil pH and Electroconductivity. In Methods of Soil Analysis, Part 2, eds. A. Page, R. Miller and R. Keeney, pp. 21. Madison, WI: American Society of Agronomy. Richards, P. 1983. Ecological Change and the Politics of Land Use. African Studies Review 25: 1-72. . 1985. Indigenous Agricultural Revolution. London: Hutchinson. • 1986. Coping With Hunger: Hazard and Experiment in an African Rice-Farming System. London: Allen & Unwin. in . 1989. Farmers Also Experiment: A Neglected Intellectual Resource African Science. Discovery and Innovation 1:19-25. Richards, J. and Tucker, R. eds. 1988. World Deforestation in the Twentieth Century. Durham, NC: Duke University Press. Riddell, J. and Campbell, D. 1986. Agricultural Intensification and Rural Development: The Mandara Mountains of North Cameroon. African Studies Review 29:89-106. Riley, B. and Brokensha, D. 1988. The Mbeere in Kenya, Vol.2. Botanical Identities and Uses. Binghamton, NY: Institute for Development Anthropology. Roberts, P. 1993. Farming by Soil. International Workshop on Sustainable Land Management for the 21st Century. University of Letherbridge, Letherbridge, Alberta, Canada. June 20-26. Robinson, W. and Harbison, S. 1995. The Fertility Decline in Kenya. Journal of International Development 7:81-91. Rocheleau, D. 1988. Gender, Resource Management and the Rural Landscape: Implications for Agroforestry and Farming Systems Research. In Gender Issues in Farming Systems Research and Extension, ed. S.V. Poats, M. Schmink and A. Spring, pp. 149-169. Boulder, CO: Westview. . 1991. Gender, Ecology and the Science of Survival: Stories and Lessons from Kenya. Agriculture and Human Values VIII: 156-165. Rocheleau, D., Steinberg, P. and Benjamin, P. 1995. Environment, Development, Crisis and Crusade: Ukambani, Kenya, 1890-1990. World Development 23:1037-1051. Roscoe, J. 1922. The Soul of Central Africa. London: Cassell and Company. Ross, S. 1993. Organic Matter in Tropical Soils: Current Conditions, Concerns and Prospects for Conservation. Progress in Physical Geography 17:265-305. Roth, M. 1993. Tenure Security, Credit Use, and Farm Investment in the Rujumbura pilot Land Registration Scheme, Rukungiri District, Uganda. Madison, WI: Land Tenure Center, University of Wisconsin- Madison. 317 Rugyema, K. 1974. A History of the Bakiga in Southwest Uganda and Northern Rwanda. Ph.D. Dissertation, University of Michigan. Ruthenberg, H. 1980. Farming Systems in the Tropics. Oxford: Clarendon Press. Rutunga, M. 1992. Conditions of Labour on Commercial Dairy Farms in Kabale District. Working Paper No. 1 . Kampala: Centre for Basic Research. Ryder, R. 1994. Farmer Perception of Soils in the Mountains of the Dominican Republic. Mountain Research and Development 14:261-266. SAA. 1994. Intellectual Property Rights for Indigenous Peoples: A Sourcebook. Oklahoma City, OK: Society for Applied Anthropology (SAA). Sanchez, P., Cuoto, W., and Buohl, S. 1982. The Fertility Capability Soil Classification System: Interpretation, Applicability and Modification. Geoderma 27:283-309. Sanchez, P. 1987. Soil productivity and sustainability in agroforestry systems. In Agroforestry: A decade of development, ed. H.A. Steppler and P.K. Nair, pp. 205-223. Nairobi: ICRAF. Sauer, C. 1925. The Morphology of Landscape. University of California Publications in Geography 2:11-54. . 1952. Agricultural Origins and Dispersals. New York: American Geographical Society. . 1956. The Agency of Man on Earth. In Man's Role in Changing the Face of the Earth, ed. W. Thomas, Jr., pp. 49-69. Chicago: University of Chicago Press. Scherer, F. 1969. The Development of Smallholder Vegetable Production in Kigezi, Uganda. Munich: Weltforum-Verlag. Schmink, M. and Wood, C.H. 1987. The "Political Ecology" of Amazonia. In Lands at Risk in the Third World, eds. P. Little and M.M. Horowitz, pp. 38-57. Boulder, CO: Westview. 318 Schoederer, R. 1993. Shady Practice: Gender and the Political Ecology of Resource Stabilization in Gambian Gardens/Orchards. Economic Geography 69:349-365. Schoenbrun, D.L. 1990. Early History in Eastern Africa's Great Lakes Region: Linguistic, Ecological and Archaeological Approaches, ca. 500 B.C. to ca. A.D. 1000. Ph.D. Dissertation, University of California- Los Angeles. . 1993. We Are What We Eat: Ancient Agriculture Between the Great Lakes. Journal of African History 34:1-31. . 1994. The Contours of Vegetation Change and Human Agency in Eastern Africa's Great Lakes Region: ca 2000 BC to ca 1000 AD. History in Africa 21:302-317. Schoepf, B. and Schoepf, C. 1988. Land, Gender, and Food Security in Eastern Kivu Zaire. In Agriculture, Woman and Land: The African Experience, ed. J. Davison, pp. 106-130. Boulder, CO: Westview Press. Schultz, T. 1964. Transforming Traditional Agriculture. New Haven: Yale University Press. Scoones, I. and Thompson, J. eds. 1994. Beyond Farmer First: Rural Peoples Knowledge, Agricultural Research and Extension Practice. London: Intermediate Technology Publications. Scott, C. and Walter, M. 1993. Local Knowledge and Conventional Soil Science Approaches to Erosional Processes in the Shivalik Himalaya. Mountain Research and Development 13:61-72. Sedjo, R.A. and Clawson, M. 1984. Global Forests. In The Resourceful Earth: A Response to Global 2000, ed. J.L. Simon and H. Kahn, pp. 128-171. Oxford: Basil Blackwell. Semmel, A. 1994. Geomorphology as Part of Development Aid: Examples from Central Africa and Cameroon. Applied Geography and Development. 23:7-19. Shannon, G. and Pyle, G. 1989. The Origin and Diffusion of AIDS: A View from Medical Geography. Annals of the Association ofAmerican Geographers 79:1-24. Shapiro, D. 1989. A Descriptive Overview of Traditional Farms and Farm Households in Zaire. The Review of Black Political Economy 18:86-96. Size, . 1990. Farm Household Size and Composition, and Women's Contribution to Agricultural Production: Evidence from Zaire. Journal of Development Studies 27:1-21. . 1995. Population Growth, Changing Agricultural Practices, and Environmental Degradation in Zaire. Population and Environment: A Journal of Interdisciplinary Studies. 16:221-236. Shaxson, T. et al. 1989. Land Husbandry: A Framework for Soil and Water Conservation. Ankeny, IW: Soil and Water Conservation Society. Shiva, V. 1988. Staying Alive: Women Ecology and Development. London: Zed Books. . 1991. The Green Revolution in the Punjab. The Ecologist 21:57-60. Showers, K. 1989. Soil Erosion in the Kingdom of Lesotho: Origins and Colonial Response, 1830's-1950's. Journal of Southern African Studies 15:263-286. Showers, K. and Malahleha, G. 1992. Oral Evidence in Historical Environmental Impact Assessment: Soil Conservation in Lesotho in the 1930's and 1940's. Journal of Southern African Studies 18:276-296. Shoji, S., Nanzyo, M. and Dahlgreen, R. 1993. Volcanic Ash Soils: Genesis, Properties and Utilization. New York: Elsevier Scientific Publications. Sikana, P. 1993a. Mismatched Models: How Farmers and Scientists See Soils. ILEIA Newsletter 9(1): 15-16. . 1993b. Indigenous Knowledge Characterisation and Farmer Participation in Northern Zambia: Implications for Research and Extension Delivery. In Rural People's Knowledge, Agricultural Research and Extension Practice: Africa Papers, ed. IIED, pp. 62-81. Research Series Vol.1, No.2. London: International Institute for Environment and Development (IIED). Simmons, O. 1988. Perspectives on Population and Development in the Third World. New York: Plenum. Simon, J.L. 1981. The Ultimate Resource. Princeton, NJ: Princeton University Press. Simon, J.L. and Kahn, H. 1984. The Resourceful Earth: A Response to Global 2,000. New York: Basil Blackwell. . 1989. Population Matters: People, Resources, Environment and Immigration. New Brunswick, NJ: Transaction Publishers. Smaling, E. 1991. Africa's Soils are Being Mined. ILEIA Newsletter 1/2:9-11. Smaling, E., Stoorvogel, J. and Windmeijer, P. 1993. Calculating Soil Nutrient Balances in Africa at Different Scales. II: District Scale. Fertilizer Research 35:237-250. Smith, S. and Reeves, E. 1989. Introduction. In Human Systems Ecology, eds. S. Smith and E. Reeves, pp. 1-18. Boulder, CO: Westview. Smith, T.E. and Blacker, J. 1963. Population Characteristics of the Commonwealth Countries of Tropical Africa. London: Athlone Press. Smyth, A. and Dumanski, J. 1993b. FESLM: An International Framework for Evaluating Sustainable Land Management. World Soil Resources Reports, No. 73. Rome: Food and Agriculture Organization (FAO) of the United Nations. Soil Management Support Services. 1987. Keys to Soil Taxonomy. 3rd Edition. Ithaca, NY: Cornell University. Soil Survey Staff. 1975. Soil Taxonomy: A Basic System of Soil Classification for Making and Interpreting Soil Surveys. Washington, DC: USDA Soil Conservation Service. . 1993. Soil Survey Manual. USDA Handbook No. 18. Washington, DC: United States Department Agriculture (USDA), Soil Conservation Service. . 1994. Keys to Soil Taxonomy. Washington, DC: United States Department Agriculture (USDA), Soil Conservation Service. Ssendiwanyo, E. 1992. Soil Survey ofKalengyere Research Station, Kabale District, Uganda. Unpublished manuscript. Kampala: Kawanda Agricultural Research Station, Soil Science Section. Stahl, M. 1993. Land Degradation in East Africa. Ambio 22:505-508. Stamp, D. 1958. Measurement of Land Resources. Geographical Review 48: 1-15. Steila, D. and Pond, T. 1989. The Geography of Soils: Formation, Distribution and Management. NY, NY: Rowman & Littlefield. Stephens, D. 1970. Soil Fertility. In Agriculture in Uganda, ed. J.D. Jameson, pp. 72-89. Oxford: Oxford University Press. Stephens, P., Bos, E., Vu, M. and Bulatao, R. 1991. Africa Region Population Projections. Washington, DC: World Bank. Steward, J. 1955. Theory of Culture Change. Urbana, IL: University of Illinois Press. Stheeman, H. 1932. The Geology of Southwestern Uganda. The Hague: Martinus Nijhoff. Stockdale, F. 1937. Soil Erosion in the Colonial Empire. Journal of Experimental Agriculture 20:3-19. Stocking, M. 1985. Soil Conservation Policy in Colonial Africa. Agricultural History 59:148-161. . 1987. Farming and Environmental Degradation in Zambia: the Human Dimension. Applied Geography 3:63-77. . 1988. Socio-Economics of Soil Conservation in Developing Countries. Journal of Soil and Water Conservation 5:381-85. . 1989. Measuring Land Degradation. In Land Degradation and Society, eds. P. Blaikie and H. Brookfield, pp. 49-63. London: Metheun. Stone, P. 1992. The State of the World's Mountains: A Global Report. London: Zed. Stonich, S. 1993. "I Am Destroying the Land!': The Political Ecology of Poverty and Environmental Destruction in Honduras. Boulder, CO: Westview Press. 322 Stoorvogel, J. and Smaling, E. 1990. Assessment of the Soil Nutrient Depletion in Sub-Saharan Africa. Report 28. Wageningen: Winand Staring Centre. Stoorvogel, J., Smaling, E., and Windmeijer, P. 1993. Calculating Soil Nutrient Balances in Africa at Different Scales. I: Supra-National Scale. Fertilizer Research 35:227-235. Struhsaker, T. 1987. Forestry Issues and Conservation in Uganda. Biological Conservation 39:209-234. Sutton, J. 1984. Irrigation and Soil-Conservation in African Agricultural History. Journal of African History 25:25-41. . 1989. Towards a History of Cultivating the Fields. Azania XXIV:99- 112. Tabor, J. A., Kilambya, D.W. and Kimbe, J.M. 1990. Reconnaissance Survey of the Ethno-Pedology of Kenya's Eastern Province. MIAC/KARI Report. Nairobi: United States Agency for International Development. Talbot, L.M. 1986. Demographic Factors in Resource Depletion and Environmental Degradation in East Africa. Population and Development Review 12:441-451. Taylor, D. 1990. Late Quaternary Pollen Records from Two Uganda Mires: Evidence for Environmental Change in the Rukiga Highlands of Southwest Uganda. Paleogeography, Paleoclimatology, Paleoecology 80:283-300. Temple, P. 1971. Lakes of Uganda. In Studies in East African Geography and Development, ed. S. Ominde, pp. 86-98. Los Angeles: University of California Press. Thapa, G. and Weber, K. 1991. Soil Erosion in Developing Countries: A Politico-Economic Explanation. Environmental Management 15:461- 473. Thomas, D. ed. 1970. Other Crops. In Agriculture in Uganda, ed. J. Jameson, pp. 207-216. Oxford: Oxford University Press. Thomas, H. and Scott, R. 1935. Uganda. London: Oxford University Press. 323 Thomas, M. and Whittington, G. 1969. Environment and Land Use in Africa. London: Metheun. Thompson, K. and Hamilton, A. 1983. Peatlands and Swamps of the African Continent. In Mires: Swamp, Bog, Fen and Moor. Ecosystems of the World, 4B, ed. A. Gore, pp. 331-374. New York: Elsevier Scientific Publishing. Thompson, J. and Scoones, I. 1994. Challenging the Populist Perspective: Rural People's Knowledge, Agricultural Research, and Extension Practice. Agriculture and Human Values 11:58-76. Throup, D. 1988. Economic and Social Origins ofMau Mau, 1945-1953. London: James Curry. Thrupp, L.A. 1989. Legitimizing Local Knowledge: From Displacement to Empowerment for Third World People. Agriculture and Human Values VI: 13-24. Thurston, H. 1992. Sustainable Practices for Plant and Disease Management in Traditional Farming Systems. Boulder, CO: Westview Press. Tiffen, M. 1995. Population Density, Economic Growth and Societies in Transition: Boserup Reconsidered in a Kenyan Case-Study. Development and Change 26:31-66. Tiffen, M., Mortimore, M. and Gichuki, F. 1994. More People, Less Erosion. New York: John Wiley & Sons. Tiffen, M. and Mortimore, M. 1994. Malthus Controverted: The Role of Capital and Technology in Growth and Environment Recovery in Kenya. World Development 22:997-1010. Timberlake, L. 1986. Africa in Crisis. Philadelphia: New Society Publishers. Tindituuza, R. 1976. Land Tenure and Land Use in Nyarurambi Parish, Rwamucucu, Rukiga. In Essays on Land Fragmentation in Kigezi District, eds. R. Tindituuza and B. Kateete, pp. 47-93. Department of Geography Occasional Paper No. 32. Kampala: Department of Geography, Makerere University. Tothill, J. ed. 1938. A Report on Nineteen Surveys Done in Small Agricultural Areas in Uganda With a View to Ascertaining the Position with Regard to Soil Deterioration. Department of Agriculture, Uganda Protectorate. Entebbe: Government Printer. Trapnell, C. 1953. The Soils, Vegetation and Agriculture of North-Eastern Rhodesia. Report of the Ecological Survey. Lusaka: Government Printer. (Originally published 1943). Trapnell, C. and Clothier, J. 1957. The Soils, Vegetation and Agricultural Systems of North-Western Rhodesia. Report of the Ecological Survey. Lusaka: Government Printer. (Originally published 1937). Trenbath, B., Conway, G. and Craig, I. 1990. Threats to Sustainability in Intensified Agricultural Systems: Analysis and Implications for Management. In Agroecology: Researching the Ecological Basis for Sustainable Agriculture, ed. S.R. Gleissman, pp. 337-366. New York: Springer-Verlag. Tucker, R. and Richards, J. eds. 1988. Global Deforestation and the Nineteenth Century World Economy. Durham, NC: Duke University Press. Tukahirwa, E. ed. 1992. Environmental and Natural Resource Management Policy and Law: Issues and Options. Vols.I. and 11. Washington, DC: World Resources Institute and Institute of Environment and Natural Resources, Makerere University (MUIENR). . 1994. Soil Erosion Research. The Uganda Experience. A Preliminary Report. Kampala: Makerere University Institute of Environment and Natural Resources (MUIENR). Tukahirwa, E. and Veit, P. 1992. Public Policy and Legislation in Environmental Management: Terracing in Nyarurembo, Uganda. Washington, DC: World Resources Institute. Tukahirwa, J. 1988. Soil Resources in the Highlands of Uganda: Prospects and Sensitivities. Mountain Research and Development 8:165-172. . 1992. Constraints on Soil Conservation in Uganda. In Soil Conservation for Survival, eds. K. Tato and H. Hurni, pp. 172-178. . 1995. Measurement, Prediction and Social Ecology of Soil Erosion in Kabale District, Southwest Uganda. Ph.D. Dissertation, Makerere University, Kampala, Uganda. Turner, B.L.,II, 1989. The Specialist-Synthesis Approach to the Revival of Geography: The Case of Cultural Ecology. Annals of the Association of American Geographers 79:88-100. Turner, B.L.,II, Hanham, R. and Portararo, A. 1977. Population Pressure and Agricultural Intensity. Annals of the Association of American Geographers 67:384-396. Turner, B.L.,11, Hyden, G., Kates, R. eds. 1993. Population Growth and Agricultural Change in Africa. Gainesville, FL: University of Florida. Turner, M. 1993. Overstocking the Range: A Critical Analysis of the Environmental Science of Sahelian Pastoralism. Economic Geography 69: 402-421. Turshen, M. 1984. The Political Ecology of Disease in Tanzania. New Brunswick, NJ: Rutgers University Press. Turyagenda, J. 1964. Overpopulation and its Effects in the Gombolola of Buhara, Kigezi. Uganda Journal 28:127-134. Uganda Protectorate. 1957. Kigezi District. Map: Western Province, Series Y532. 3rd Edition. Entebbe: Lands and Survey Department. UNEP. 1988. Uganda Case Study: A Sampler Atlas of Environmental Resources Data Set with GRID. Nairobi: United Nations Environment Programme (UNEP). . 1989. Environmental Data Report. Cambridge, MA: Basil Blackwell. . 1992. World Atlas of Desertification. London: Edward Arnold. UN. 1992. 1990 Demographic Yearbook. New York: United Nations. USCR. 1986. Human Rights in Uganda: A Season of Hope for its Refugees and Displaced Persons. Washington, DC: United States Committee for Refugees (USCR). , 1990. World Refugee Survey, 1991. Washington, DC: United States Committee for Refugees. 326 . 1991a. World Refugee Survey, 1992. Washington, DC: United States Committee for Refugees. . 1991b. Exile from Rwanda: Background to Invasion. Issue Paper. Washington, DC: United States Committee for Refugees. . 1992. World Refugee Survey, 1993. Washington, DC: United States Committee for Refugees. . 1993. World Refugee Survey, 1994. Washington, DC: United States Committee for Refugees. . 1994. World Refugee Survey, 1995. Washington, DC: United States Committee for Refugees. Valleroy, L., Harris, J. and Way, P. 1993. The Consequences of HIV/AIDS in Eastern Africa on Mothers, Children and Orphans. Population and Environment: A Journal of Interdisciplinary Studies 14:301-306. Veit, P., Mascarenhas, A. and Ampadu-Agyei, O. 1995. Lessons from the Ground Up: African Development That Works. Washington, DC: World Resources Institute. Verma, V., HussManns, R. and Mehran, F. 1990. Surveys of Economically Active Populations, Employment, Unemployment and Underemployment: An ILO Manual on Concepts and Methods. The Hague: Institute of Social Studies, The Netherlands. Vermeer, D. 1983. Food Sufficiency and Farming in the Future of West Africa: Resurgence of Traditional Agriculture. Journal of African Studies 10: 74-83. Vierich, H. and Stoop, W. 1990. Changes in West African Savanna Agriculture in Response to Growing Population and Continuing Low Rainfall. Agriculture, Ecosystems and Environment 31:115-132. Vogel, H. 1988. Deterioration of a Mountain Agro-Ecosystem in the Third World Due to the Emigration of Rural Labor. Mountain Research and Development 8:321-329. Walker, F.R. and Wortmann, C. 1993. Why Involve Farmers in Research? Paper presented at the International Crop Science Conference for Eastern and Southern Africa, Kampala, Uganda, June 14-18. Ward, W. and Deren, B. 1991. The Economics of Project Analysis. Washington, DC: World Bank. Warren, D. 1991. Using Indigenous Knowledge in Agricultural Development. World Bank Discussion Paper 127. Washington, DC: World Bank. Warren, D., Brokensha, D. and Slikkerveer, L., eds. 1991. Indigenous Knowledge Systems: The Cultural Dimension of Development. London: Kegan Paul International. Watts, M. 1983. Silent Violence: Food, Famine and Peasantry in Northern Nigeria. Berkeley, CA: University of California Press. . 1987. Powers of Production: Geographers Among the Peasants. Environment and Planning D: Society and Space 5:215-230. . 1989. The Agrarian Question in Africa: Debating the Crisis. Progress in Human Geography 13:1-41. Way, P. and Stanecki, K. 1993. How Bad Will It Be? Modelling the AIDS epidemic in Eastern Africa. Population and Environment: A Journal of Interdisciplinary Study 14:265-278. Wayland, E. and Brasnett, N. 1938. Soil Erosion and Water Supplies in Uganda. Geological Survey of Uganda, Memoir No.4. Entebbe: Government of Uganda. Wayland, E., ed. 1956a. The Geology of the Volcanic Area of Bufumbira, South-West Uganda. Memoir No. Ill, Part I. Entebbe: Geological Survey Department. . 1956b. The Petrology of the Volcanic Field of Bufumbira, South-West Uganda and Other Parts of the Birunga Field. Memoir No. Ill, Part II. Entebbe: Geological Survey Department. Webb, M., Conroy, C, Baguma, D. and Kabanyoro, R. 1994. Visit Report: Survey of the Socio-Economics of Weed Control on Smallholder Farms in the Districts of Uganda. Unpublished Report of Field Visit, 24 April -30 May, 1993. R2014 (S). London: Natural Resources Institute, Overseas Development Administration. Webster, J. ed. 1979. Chronology, Migration and Drought in Interlacustrine Africa. New York: Dalhousie University Press and Africana Publishing Company. Weinstock, D. 1984. Getting the Right Feel for the Soil: Traditional Methods of Crop Management. The Ecologist 14:146-149. Weismann, U. 1992. Socio-Economic Viewpoints on Highland-Lowland Systems: A Case Study on the Northwest Side of Mount Kenya. Mountain Research and Development 12:375-381. Weischet, W. and Caviedes, C. 1994. The Persisting Ecological Disadvantages of the Tropics. London: Longman Scientific & Technical. Were, J., Muhwezi, A. and Rutaremwa, G. 1992. Population Pressure, Land Use Changes and Consequences on the Environment in Kabale District, Uganda. Department of Geography Manuscript. Kampala: Department of Geography, Makerere University. White, T.A. and Jickling, J. 1995. Peasants, Experts, and Land Use in Haiti: Lessons from Indigenous and Project Technology. Journal of Soil and Water Conservation 50:7-14. Whitlow, R. 1988. Soil Conservation History in Zimbabwe. Journal of Soil and Water Conservation 43:299-303. Wijk, J. 1993. Intellectual Property Rights for Agricultural Biotechnology: Options and Implications for Developing Countries. The Hague: International Service for National Agricultural Research. Wilding, L., Smeck, N. and Hall, G., eds. 1983a. Pedogenesis and Soil Taxonomy: I. Concepts and Interactions. Developments in Soil Science, 11 A. New York: Elsevier Scientific Publishing. . 1983b. Pedogenesis and Soil Taxonomy: II. The Soil Orders. Developments in Soil Science, 11 A. New York: Elsevier Scientific Publishing. Wilkens, G. 1987. Good Farmers: Traditional Agricultural Resource Management in Mexico and Central America. Berkeley, CA: University of California Press. . 1989. Transferring Traditional Technology: A Bottom-Up Approach for Fragile Lands. In Fragile Lands of Latin America: Strategies for Sustainable Development, J. Browder, ed., pp. 44-57. Boulder, CO: Westview Press. Wischmeier, W. 1975. Estimating the Soil Loss Equation's Cover and Management Factor for Undisturbed Areas. In Present and Prospective Technology for Predicting Sediment Yield and Sources, USDA ARS-S- 40, pp. 118-125. Washington, DC: USDA. Wisner, B. 1989. Power and Need in Africa. Trenton, NJ: Africa World Press. Wood, A. 1992. Zambia's Soil Conservation Heritage: A Review of Policies and Attitudes Towards Soil Conservation from Colonial Times to the Present. In Soil Conservation for Survival, eds. K.Tato and H. Hurni, pp. 156-171. Ankeni, HV: Soil and Water Conservation Society. Woomer, P. and Swift, M. eds. 1994. The Biological Management of Tropical Fertility. Chichester, West Sussex, UK: John Wiley. World Bank. 1986. Population Growth and Policies in Sub-Saharan Africa. Washington, DC: World Bank. . 1987. Land Tenure and Soil Conservation. Final Report of Working Group 8, Agricultural Services Task Force. Kampala: Government of Uganda and the World Bank. . 1992a. World and Regional Supply and Demand Balances for Nitrogen, ' Phosphate, and Potash, 1990/1-1996/7. World Bank Technical Paper No. 176. Washington, DC: World Bank. . 1992b. Report on the Baseline Survey for the Farming Year 1989 (Mimeo). Mbarara, Uganda: Southwest Region Agricultural Rehabilitation Project, Monitoring and Evaluation Unit, World Bank. . 1993. Uganda: Agriculture. Washington, DC: World Bank. . 1994. Traditional Knowledge and Sustainable Development. Washington, DC: World Bank. . 1995. African Development Indicators. Washington, DC: World Bank. WRI. 1989. World Resources, 1989-90. Oxford: Oxford University Press. . 1994. World Resources, 1994-95. Oxford: Oxford University Press. Wrigley, C.C. 1959. Crops and Wealth in Uganda: a Short Agrarian History. Kampala: East African Institute of Social Research. 330 Yapa, L. 1993. What Are Improved Seeds? An Epistemology of the Green Revolution. Economic Geography 69:254-273. Yeld, R. 1965. Resettlement and its Effects on Kiga Patterns of Life. Sociology Working Paper No.l.. Kampala: MISR. Hunger in Kigezi, Southwest Uganda. Nkanga 3:24-37. . 1969a. Land Kampala: Makerere Institute for Social Research (MISR). in Social Change: A Study Among the Kiga . 1969b. The Family of ' Kigezi District, Southwest Uganda. Ph.D. Dissertation, University of East Africa. Yost, D. and Eswaran, H. 1990. Major Land Resource Areas in Uganda. Kampala: United States Agency for International Development (USAID). Young, A. 1989. Agroforestry for Soil Conservation. Wallingford, UK: Commonwealth Agricultural Bureau International. Young, C. 1988. The African Colonial State and Its Political Legacy. In The Precarious Balance: State and Society in Africa, eds. D. Rothchild and N. Chazan, pp. 25-66. Boulder, CO: Westview. Zake, J. 1991. Report of the Soil Fertility Survey of the South Western Region, Kabale and Rukungiri Districts. (Mimeo). Mbarara, Uganda: Southwest Region Agricultural Rehabilitation Project, Monitoring and Evaluation Unit, World Bank. Zelinsky, W., Kosinski, L. and Prothero, R. eds. 1970. Geography and a Crowding World. New York: Oxford University Press. Zerihun, T. 1994. Integration of Population Dynamics and Environmental Change in Agricultural Development Policies: The Soil Erosion and Conservation Issue in Uganda. In Inducing Food Insecurity: Perspectives on Food Policies in Eastern and Southern Africa, ed. M.Salih, pp. 164-179. Stockholm: Gotab. Zimmerer, K. 1991. Wetland Production and Smallholder Persistence: Agricultural Change in a Highland Peruvian Region. Annals of the American Association of Geographers 81:443-463. (Dis)courses in Cochabamba, Bolivia: . 1993. Soil Erosion and Social Perceiving the Nature of Environmental Degradation. Economic Geography 69:312-327. 1994. Local Soil Knowledge: Answering Basic Questions in Highland . Bolivia. Journal of Soil and Water Conservation 49:29-34. Zurick, David. Historical Links Between Settlement, Ecology, and Politics in the Mountains of West Nepal. Human Ecology 17 (1989): 229-23 8. BIOGRAPHICAL SKETCH Cary Farley was bom in Norway, Maine in 1961. He is the first of three sons of Carlton and Valkyrie Farley. Cary grew up in a number of different states in New England, and eventually graduated from high school at Pinkerton Academy in New Hampshire in 1979. He enrolled at Plymouth State College (New Hampshire) where he studied geography, history and Canadian studies, and graduated with a BA in 1984. Cary financed much of his undergraduate education by working on a dairy farm in New Hampshire; the summer following his graduation, he joined the Peace Corps and worked as an agricultural extension agent in Zaire. Upon completion of his Peace Corps service, Cary entered an MS program at the Department of Geography and Geology at the University of Massachusetts (Amherst). He conducted field research on tropical deforestation and forest resource use in Zaire, and completed his MS degree in 1990. Cary then enrolled in the PhD program in the Department of Geography at the University of Florida (Gainesville), where he studied tropical agriculture, resource management and smallholder production systems in the highlands of Uganda. Cary plans to continue working with smallholder agricultural production systems in tropical highland environments; he will begin a related research position as a post-doctorate fellow with CIAT in Eastern Africa in 1996. 332 I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Edward J. Malecki, Chairman Professor of Geography I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Abraham C. Goldman, Cochairman Assistant Professor of Geography I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Art Hansen Associate Professor of Anthropology I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Peter Nkedi-Kizza (/(/ Associate Professor of Soil and Water Science This dissertation was submitted to the Graduate Faculty of the Department of Geography in the College of Liberal Arts and Sciences and to the Graduate School and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy. August 1996 Dean, Graduate School 00 oo 00 Ov Ov Ov o O O fS ts ts 270n On On tn NO On On NO en o P- ON On fS a* VO "i es