Professional Documents
Culture Documents
Ruthenberg Farming Systems in The Tropics PDF
Ruthenberg Farming Systems in The Tropics PDF
HANS RUTHENBERG
CLARENDON
PRESS 1971
OXFORD
Preface
THIS book arose from an introductory course of lectures on the characteristics
of tropical farming systems for undergraduates at Stuttgart-Hohenheim and Gbttingen Universities and for graduates in general agriculture who were preparing for applied work in the tropics. Beyond its use in lecturing, the book is intended to provide the agricultural development worker with a framework into which he can fit the various types of farming that he will encounter, and at the same time to introduce him to the management problems that arise in particular farming systems. The subject is considered in the agro-economic setting. Emphasis is !aid on the interactions between technical and economic aspects of farming. Information about soils, climate, and socioinstitutional factors is deliberately kept short. Recently, much literature has become available about tropical agriculture, dealing with such crucial topics as soil fertility, irrigation, and practices in plant and animal husbandry. Land-tenure problems, farmers' behaviour, and agricultural-policy measures are increasingly dealt with in a growing number of publications. ! have, therefore, preferred to restrict this book to the farm-management aspects. The primary economic characteristics of the more important tropical farming systems and their broad lines of development are described and analysed. It is hoped that this information, as well as promoting understanding, will be helpful in the planning of meaningful development policies for agricultural production in tropical countries. The subject is by no means dealt with exhaustively, and many more farming systems than those depi-ted deserve consideration. I have tried in the illustrations to cover the major tropical crops, but no emphasis has been laid on a geographical balance. As far as possible, the case studies have been taken from areas that I have visited and where I was in a position to confirm the information cited in the relevant literature. This is the reason why most of the case studies are from Africa and Asia and very few are from Latin America. In order to describe the economic situation of each type of farming, I have used quantitative information about inputs and outputs, and much of it is expressed in monetary terms. These figures have, of ceurse, to be considered with great caution. They are difficult to compare because of wide differences in price levels and price relationships. Monetary information of this nature is necessary in order to help provide a clear picture of the circumstances of
vA
PREFACE
farming under different conditions of production. It should not be overlooked, however, that the relevance of the quantitative information depends on the year, on the country, on the farming system, and on the method of estimation. In the complex task of writing this book 1 have been greatly assisted by several co-workers. The main debt is, of course, to the numerous smallholders, farmers, and estate managers who so willingly discussed their work and problems. They in fact provided the information t'.at is presented in this book. Most valuable advice and criticism were given by my friend John D. MacArthur, who read the manuscript a number of times. To him I am indebted for a great number of suggestions and ideas, and for Chapters 2 and 10, which he wrote. Mr. and Mrs. P. C. Brown deserve my thanks for having translated the original German version of the manuscript. I owe a special word of thanks to the Volkswagen Foundation and the Fritz Thyssen Foundation. The book would not have been possible without the research funds that I received from them. The inclusion of the plates was made possible by a grant from Mr. Heinric'. Herrmann, Stuttgart. Stuttgart-Hohenheim
HANS RUTHENBERG
Contents
LIST OF PLATES KEY TO TERMS Xiii XV
1. INTRODUCTION 1.1. Definition of the geographical area 1.2. Definition of farming systems
1.2.1. Collecting 1.2.2. Cultivation 1.2.3. Grassland utilization
I 1 2
2 2 5
5 6
2.1.1. Climatic influences 2.1.2. Soil conditions 2.1.3. Biological environment 2.2. Some critical aspects of tropical farming 2.2.1. Problems of soil fertility 2.2.2. Coping with risk and uncertainty 2.2.3. Low labour productivity 2.2.4. The problem of seasonality 2.3. The dynamics of tropical farming 3. SHIFTING CULTIVATION SYSTEMS 3.1. Definition 3.2. Types of shifting cultivation and their geographical distribution 3.2.1. Vegetation systems 3.2.2. Migration systems 3.2.3. Rotation systems 3.2.4. Clearance systems 3.2.5. Cropping systems 3.2.6. Tool syst.ms
8 8
9 10 11 11 11 12 13 14 15
16 16 10i 17 17 20 22 24 24
CONTENTS 3.3. General characteristics of shifting systems 3.3.1. The spatial organization of cropping 3.3.2. Cropping principles 3.3.3. Characteristics of the fertilizer economy 3.3.4. Characteristics of animal husbandry 3.3.5. Characteristics of the labour economy 3.4. Three case studies 3.4.1. Extensive shifting cultivation with monocropping of the Iban, Sarawak 3.4.2. Shifting cultivation with mixed cropping in Congo-Kinshasa 3.4.3. Shifting cultivation with rotation among the Bemba, Rhodesia 3.5. Weaknesses of the system 3.5.1. Difficulties in the introduction of new methods 3.5.2. The tendency towards 'degraded' shifting systems 3.6. Development paths of shifting systems 3.6.1. Improved shifting systems 3.6.2. The change from shifting systems to more intensive types of farming
24 24 27 31 36 36 40 40 41 43 44 44 45 48 48 52
4. SEMI-PERMANENT CULTIVATION SYSTEMS 4.1. Definition 4.2. Types of semi-permanent cultivation and their geographical distribution 4.2.1. Semi-permanent cultivation or. the fertile soils of the humid savannas 4.2.2. Semi-permanent cultivation in connection with perennial crops 4.2.3. Semi-permanent cultivation in connection with irrigation farming 4.2.4. Unregulated ley systems in the drier savannas 4.2.i. Unregulated Icy systems in high-altitude areas 4.3. General characteristics of semi-permanent systems 4.2.1. Spatial organization of cropping 4.3.2. Cropping principles 4.3.3. Characteristics of the fertilizer economy 4.3.4. Characteristics of animal husbandry 4.3.5. Characteristics of the labour economy 4.4. Two case studies 4.4.1. Semi-permanent farming in thc subhumid savanna: rice holdings in the Kilombcro Valley, Tanzania 4.4.2. Unregulated Icy farming in the drier savanna: cotton-maizemillet holdings in Sukumaland, Tanzania 4.5. Weaknesses of the system 4.5.1. The tendency towards soil mining 4.5.2. Degeneratioa of the livestock economy 4.5.3. Bottlenecks in the labour economy
55 55 58 58 58 59 59 59 60 60 62 65 67 69 73 73 75 76 77 77 78
CONT.ENTS 4.6. Development paths of semi-permanent systems and unregulated Icy systems 4.6.1. Improved systems 4.6.2. Change towards more intensive types of farming 5. SYSTEMS WITH REGULATED LEY FARMING 5.1. Definition 5.2. Types of Icy farming and their geographical distribution 5.2.1. Regulated Icy systems in traditional smallholder areas 5.2.2. Regulated Icy systems in settlement schemes 5.2.3. Regulated Icy systems in large farms 5.3. General characteristics of regulated ley systems 5.3.1. Principles of crop and ley management 5.3.2. Characteristics of the fertilizer economy 5.3.3. Charactcristics of animal husbandry 5.3.4. Characteristics of the labour economy 5.4. Three case studies 5.4.1. Tobacco farms in Rhodesia 5.4.2. Large farms in the Kenya highlands 5.4.3. Smallholdings in the Kenya highlands 5.5. Weaknesses of the system 5.5.1. The low fertility effect of tropical grass leys 5.5.2. Unfavourable capital-output ratio 5.5.3. Dependence on improved animal husbandry 5.6. Development paths of regulated Icy systems 5.6.1. Improved Icy systems 5.6.2. Change from Icy systems to other types of farming 6. SYSTEMS WITH PERMANENT CULTIVATION ON RAIN-FED LAND 6.1. Definition 6.2. Types of permanent rain-fed cultivation and their geographical distribution 6.2.1. Permanent cultivation in tropical highlands 6.2.2. Permanent cultivation in the African savannas 6.2.3. Permanent cultivation in monsoon Asia 6.2.4. Permanent cultivation in the hot, humid tropics 6.3. General characteristics of land use with permanent rain-fed cultivation 6.3.1. Spatial organization of cropping 6.3.2. Cropping principles 6.3.3. Characteristics of the fertilizer economy 6.3.4. Characteristics of animal husbandry 6.3.5. Characteristics of the labour economy 6.4. Four case studies 6.4.1. Maize-millet holdings in Kakamega District, Kenya 6.4.2. Millet holdings in Ukara, Lake Victoria 6.4.3. Groundnut-millet holdings in Senegal 6.4.4. Millet-wheat holdings in the Bombay Deccan
ix 79 79 80 83 83 83 84 85 85 86 86 87 88 89 89 89 92 94 95 95 96 96 97 97 97 99 99 99 100 101 101 101 104 104 105 108 110 112 117 117 118 119 121
CONTENTS 6.5. Weaknesses of the system 6.6. Development paths of permanent cultivation systems 6.6.1. Improved permanent cultivation systems 6.6.2. Change from permanent rain-fed cultivation to other types of farming
7. SYSTEMS WITH ARABLE IRRIGATION FARMING 7.1. Definition 7.2. Types of irrigation farming and their geographical distribution 7.2.1. Water supply systems 7.2.2. Cropping systems 7.2.3. Exploitation systems 7.3. General characteristics of irrigation farming 7.3.1. A comparison with rain-fed farming 7.3.2. Spatial organization of cropping 7.3.3. Cropping principles 7.3.4. Characteristics of the fertilizer economy 7.3.5. Characteristics of animal husbandry 7.3.6. Characteristics of the labour economy 7.4. Eight case studies 7.4.1. Water availability and irrigation farming: from India, West Pakistan, and Singapore five case studies 7.4.2. Wet-rice systems: three case studies from East Pakistan, Thailand, and Taiwan 7.4.3. Farming systems in large irrigation schemes 7.5. Weaknesses of the system 7.5.1. Insufficient control of water supplies 7.5.2. Wasteful irrigation and husbandry practices 7.5.3. Dependence on inadequate institutional arrangements 7.6. Development paths of irrigation farming 7.6.1. Innovations and improved husbandry 7.6.2. Irrigation farming and mechanization 7.6.3. Introduction of intensive types of animal husbandry 8. SYSTEMS WITH PERENNIAL CROPS 8.1. Definition and genesis 8.1.1. Definition 8.1.2. Genesis 8.2. Types of farming with perennial crops and their geographical distribution 8.2.1. Cropping systems 8.2.2. Exploitation systems 8.3. General characteristics of farming systems with perennial crops 8.3.1. A comparison with arable farming 8.3.2. Some farm-management characteristics of perennial crops
132 132 132 133 138 138 139 139 142 147 152 154 155 160 160 166 176 181 182 183 183 185 185 186 188 189 189 189 189 191 192 193 197 197 201
CONTENTS 8.4. Estates and smallholdings with perennial field crops 8.4.1. Sugar-cane holdings 8.4.2. Sisal holdings 8.4.3. Banana holdings 8.5. Estates and smallholdings with shrub crops 8.5.1. Coffee holdings 8.5.2. Tea holdings 8.6. Estates and smallholdings with tree crops 8.6.1. Cocoa holdings 8.6.2. Rubber holdings 8.6.3. Oil-palm holdings 8.6.4. Coconut-palm holdings 8.7. Weaknesses of farming systems with perennial crops 8.7.1. Weaknesses of smallholder farming with perennial crops 8.7.2. Weaknesses of the estate economy 8.8. Development paths of farming systems with perennial crops 8.8.1. Increased application of technical innovations 8.8.2. From mixed cropping to pure stands and interculture 8.8.3. From holdings with one cash crop towards diversified commercial production 8.8.4. Increasing importance of perennial food and fruit crops 8.8.5. Integration of estate and smallholder production 9. GRAZING SYSTEMS 9.1. Definition and geographical distribution of grazing systems 9.2. Total nomadism 9.2.1. General characteristics of total nomadism 9.2.2. A case study: camel nomads in the Sahara 9.2.3. Weaknesses of total nomadism 9.2.4. Development paths of total nomadism 9.3. Semi-nomadism 9.3.1. General characteristics of semi-nomadism 9.3.2. A case study: semi-nomadism of the Bahima, Uganda 9.3.3. Weaknesses of semi-nomadism 9.3.4. Development paths of semi-nomadism 9.4. Ranching 9.4.1. General characteristics of ranching 9.4.2. Two case studies 9.4.3. Weaknesses of ranching 9.4.4. Development paths of ranching
xi 206 206 210 213 218 219 228 232 232 234 240 242 244 244 247 248 248 249 249 251 251 252 252 254 254 256 258 259 260 260 262 264 267 271 271 274 280 281
287
INDEX
303
List of Plates
(Plater appear between pp 144-145) Dahomey 1. Fallow vegetation, burnt by shifting cultivators, Cercle d'Allada, cultivator's shifting a in 2. Mixed cropping of groundnuts, sunflower, and maize field, Urambo, Tanzania being transformed into 3. Valley bottoms in the savanna zone of the Ivory Coast fields with irrigated rice farming in Nyeri District, 4. Dairy cattle on smallholdings with regulated ley Kenya. cultivation in the African 5. Tobacco is a cash crop well-suited to shifting Tanzania savannas. A small entrepreneur at Urambo, being transformed into 6. Valley bottoms in the rainforest zone of the Ivory Coast rice fields with irrigated country, Senegal 7. Cattle feeding on leaves of Fa.dherbiaalbida, in Screre Island, Tanzania Ulcara field, the to stable 8. Transporting manure from the Tanzania. The Island, Ukara on manure 9. Applying stable manure and green manure is placed at the base of the ridge 10. Smallholder Icy farming in Kisii District, Kenya used by shifting II. Swiss cows grazing on leys established on land traditionally was taken on picture The cultivators. Bush with tsetse-flies in the background. Dahomey Cotonou, to close a Swiss demonstration farm, 12. Two-wheel tract )r, ploughing a rice field in Taiwan performed by seasonal 13. Land preparation for rice in Ceylon. The work is being paid labour maturing rice, Taichung 14. Soil preparation for intercropping of sweet potatoes in District, Taiwan Note the seed cane in 15. Relay-planting of sugar-cane in growing rice in Taiwan. rice of rows the the mud between 16. Rice terraces and rain-fed farming in the Highland of Madagascar 17. Farmer in his rice field, Taichung District, Taiwan right shows improved 18. Two rice fields close to Madras, India. The field to the the left shows IR 8, to field The traditional varieties, which tend to lodge. doses of nitrogen high absorb productively can which a non-lodging variety buffaloes, in Ceylon. 19. Preparation of a rice field for planting by trampling with In Madagascar cultivation. plough or hoe after The work is performed the same method is applied and called 'pi6tinage' soil 20. View into the banana-coffee-bean plot of a Bahaya holding, Tanzania. The grass is covered with mulch
xiv LIST OF PLATES 21. Multiple cropping system of paddy fields in Taiwan, China 22. A field in the Gezira Scheme, in the Sudan, prepared for planting by the scheme management 23. A tenant in the Managil Extension of the Gezira Scheme, in the Sudan, irrigating his cotton 24. Well-terraced banana-coffee plot on Mt. Kiiimanjaro, Tanzania 25. Coffee plantation, Nyeri District, Kenya 26. Badly eroded tea plantation, Ceylon, medium-altitude level 27. Young tea plantation established with new clones, planted at high density, surrounded by tea which is about 50 years old, Ceylon, high-altitude level 28. Rubber plantation, Ivory Coast 29. Smallholder plot ready for the planting of oil-palms. The woman is planting a cover crop (Centrosemna pubescens) 30. Young oil-palm plantation, Ivory Coast 31. Root system of two piennial grasses in L dry savanna climate, Kongwa, Tanzania 32. Overgrazed and eroded land in Baringo District, Kenya
Note (2) In the tables and figures, quotients of standard scientific units have been abbreviated using the index notation: e.g. kilogrammes per hectare becomes kg ha-'.
1. Introduction
1.1. Definition of the geographical area
Geographically, this book deals with tropical farming systems only, although much of the information is valid also for subtropical farming. The tropical regions are generally defined as those lying between 2340 to the north and 23 to the south of the equator. This, however, is too rigid a definition for the purpose of dealing, with farming systems. I therefore prefer the definition of Waibel (1933), who used the criteria of economic geography, and defined as tropical those regions between the dotted lines in Fig. 1.1.1 This definition is generally applied in the following chapters.
.:.................'"... .:, .... i: . ... . .... ... > , .. .i~lili:: .... ... ....
FiG. 1.1. The tropics, defined according to the principles of economic geography (from Manshard 1968, p. 18). The main ecological regions within the tropics have been classified by Manshard (1968, pp. 36, 64), who distinguishes between (I) the rainforest, which has 0-21 arid months, (2) the humid savannas, having 24-6 arid months, (3) the dry savannas, with 6-81 arid months, (4) the thorn savannas, which have 84-10 arid months, (5) the semi-desert and full desert, where there are more than 10 arid months, and (6) the tropical high-altitude areas. identical results (see Manshard 1961, p. 13).
a
I Von Wissman's definition of the tropics, based on climatic criteria, gives almost
INTRODUCTION
INTRODUCTION
for severai years and is used for grazing. In the savannas there are extensive areas of wild or unregulatedley farming. In these areas, we find that, after a period of cultivation of several years, the field is covered with grass and shrubs for several more years and serves as rough pasture. Regulated ley farming, with established swards during the non-cropping period, is rare in the tropics, although it is found in some highlands (e.g. in Kenya) and in Latin America. Field systems occur where one arable crop follows another, and where established fields are clearly separated from each other. The grassland associated with field systems is usually treated as permanent grassland, whether it is rough or well cared for, and it is separatcd from the arable land. Systems with perennialcrops, i.e. with crops that cover the land for many years, are in a separate category. Within this category we distinguish between perennial field crops like sugar-cane and sisal, bush crops like tea and coffee, and tree crops proper like oil-palm and rubber. All kinds of rotation are found. In some cases tree crops are alternated with fallow, in others with arable farming, grazing, or other perennial crops. (b) Classification according to the intensity of the rotation (after Terra 1958; Nye and Greenland 1961; and Faucher 1949). The fallow and ley systems display considerable variations and degrees of intensity. This is especially true for field-bush rotations and for unregulated ley systems in the African savannas. A relatively simple and appropriate criterion for classification is the relationship between crop cultivation and fallowing within the total length of one cycle of land utilization. Following the suggestion of Joosten (1962), we define the symbol R as the number of years of cultivation multiplied by 100 and divided by the length of the cycle of land utilization. The length of the cycle is the sum of the number of years of arable farming plus the number of fallow years. The characteristic R indicates the proportion of the area under cultivation in relation to the total area available for arable farming. If, for instance, 40 per cent of the available arable land in one holding is cultivated, then R is 40. As long as fallow farming has an extensive character, in which many fallow years follow a short period of cultivation, R remains very small. If, for example, eighteen fallow years succeed two years of cultivation, as is frequently the case in the rainforest, R amounts to 10. This extensive type of fallow farming is generally designated shifting cultivation, because the shifting of fields within a broad area of wild vegetation usually results in the gradual relocation of the farming population. On the other hand, it should not be forgotten that there are a number of regions where stationary populations practise shifting cultivation. The larger R becomes, the higher is the percentage of the area cultivated annually in relation to the total area available for arable farming, and the more stationary the character of the farming becomes. When cultivation is
INTRODUCTION
extended so far at the expense of fallowing that the characteristic R reaches or exceeds the value 30, then we can hardly speak of a shifting of the fields any more. A level of intensity of land utilization has been achieved that Terra, Nye, and Greenland designate semi-permanent cultivation, and that Faucher designates stationary cultivation with fallowing. A characteristic R value of 50 is obtained, for example, if seven fallow years succeed seven years of cultivation. When the R value exceeds 70, and the soil is cultivated nearly every year or even more often, then permanentfarming is being practised. Permanent farming may again be conveniently classified according to the degree of multiple cropping. An R value of 150 would indicate that 50 per cent of the area is carrying two crops a year, and a value of 300 would indicate that three crops a year are being grown. (c) Classfication according to the water supply. One of the first steps in classifying land utilization is to question whether farming is practised with or without irrigation. In irrigation farming, to ensure that the moisture level of the soil is higher than would occur naturally, a water suppiy is directed into the fields, or the rain-water in the fields is purposefully dammed up. Farming without irrigation is widely referred to as dryfarming. In the interest of precise definition, however, it is advisable to speak of dry farming only where in semi-arid climates grain and leguminous crops are alternated with a fallow of one or two years. A distinction between irrigated farming and rainfedfarming is therefore preferred. (d) Classificationaccording to the croppingpatternand animalactivities. A most important aspect of the definition of farming systems is usually the classification according to the main crops and the livestock activities of the holdings. Each activity has different requirements as to climate, soils, markets, and inputs. Therefore those farms can be grouped together whose gross returns (sales plus household consumption plus changes in stock) are similarly constituted, to give, for example, coffee-banana holdings or rice-jute holdings. (e) Classificationaccordingto the implements usedfor cultivation. In addition, holdings are occasionally classified according to the main implements used. In various parts of the world, the land is cultivated by methods that require either no implements, or a few very simple tools. Millet is sown without fire-farming or soil preparation by a few nomads in the Sahara. Shifting cultivators frequently sow in ashes without touching the soil either beforehand or afterwards. Rice growers in Madagascar, Ceylon, and Thailand make use of the treading of animals: a large number of cattle are driven across the moist field to trample down the soil until it becomes a mire ready for planting. In some parts of the world, planting-sticks aod digging-sticks have not yet been replaced by hoes, spades, or ploughs. however, with the exception of these pre-technical methods, the main divisions are (1) hoe
INTRODUCTION
animal traction; and farming or spade farming; (2) farming with ploughs and (3) farming with ploughs and tractors. There are (f) Classification according to the degree of commercialization. to relation in a number of distinct types according to the percentage of sales are products the gross return. Subsistence farms are those where agricultural household; the of needs the only covering of raised with the main purpose in relation sales of share The products. these of selling is limited to surpluses are commercializedfarms Partly cent. per 25 to the gross returns is less than in addition those where the systematic cultivation of cash crops is practised sales still of percentage the to production for household needs. However, farms Semi.commercialized remains below 50 per cent of the gross returns. cash much so household are similar, but in addition to production for the of cent per 50-75 to cropping is carried out that the share of sales amounts in type, remaining the gross returns. Highly commercialized farms are the household. the by which less than 25 per cent of the gross output is consumed 1.2.3. Grassland utilization necessiThe low yields of grassland areas in the arid and semi-arid tropics of ranching tate either nomadism or semi-nomadism, or the development of livetypes different the classify to systems. It seems advisable, therefore, animals the both of stationariness of stock farming according to the degree main types can and those who tend them. According to Dittmar (1954), five be distinguished: owners do not (1) Total nomadism covers systems in which the animal regular cultivahave a permanent place of residence. They do not practise herds. the tion and their families move with owners have a (2) Semi-nomadism is a related system where the animal cultivation is supplementary permanent place of residence near which herds to their with travel time they - practised. However, for long periods of distant grazing areas. permanent place (3) Transhumance is tht situation in which farmers with a periods of time of residence send their herds, tended by herdsmen, for long to distant grazing areas. in (4) Partial nomadism is characterized by farmers who live continuously which remain permanent settlements, and who have herds at their disposal in the vicinity. remain on the (5) Stationary animal husbandr'v occurs where the animals year. entire holding or in the village throughout the 1.2.4. Procedure in this book
in The definition, classifications, and criteria outlined above are employed of their the following chapters. Collecting systems are omitted because
INTRODUCTION
economic insignificance. The book is restricted to cultivation systems and grazing systems, and its main chapzers are organized according to the type and intensity of land use. As major cultivation systems, a distinction is thus made between shifting cultivation systems, semi-permanent cultivation systems, systems with regulated ley farming, systems ,"ith permanent cultivation on rain-fed land, systems with arable irrigation farming, and systems with perennial crops. Under the heading of grazing systems, only nomadic grassland use and ranching are considered. The other forms of grassland use are described as part of the relevant cultivation system. 1.3. Definition of farms Farming within each system is carried out in holdings, which are more or less distinct managerial units. Following Woermann (1959), holdings are here defined as organized, economic units in which plant and animal products are produced for the purposes of gain. The economic aims that underlie such gainful employment vary from case to case. In larger holdings, market production and profits are the main objectives, whereas for the smallholderand most of the tropical land is farmed by smallholders-the farm and the household are closely connected. Equally as important as, or even more important than, saleable produce on such a farm is the provision of food for the household, together with goods for personal consumption like tobacco, or raw materials like wool or materials for building huts. Other important production aims are the safeguarding of the future by accumulation of capital in the form of animals or plantations, and the increasing of social status by accumulation ofwealth or by special technical, social, and economic achievements. Although agricultural enterprises are usually tied to a clearly defined place, this is not always the case. For example, production by nomads is based on herds that trek from grazing area to grazing area. In shifting cultivation, the fields are shifted over an extensive area. Nomads and shifting cultivators produce for the purposes of gain, but they operate without a stationary habitat. Holdings are usually, but not always, under the management control of one person. In some cases, there are holdings where a group of people operate on a communal basis without electing or nominating a manager. In smallholder farming in Africa south of the Sahara, it is usual for individual members of a household to cultivate small plots as they think fit and with differing economic aims. In these cases, the holding comprises the various fields belonging to the family as a whole. Where production for the market has priority, there is usually no difficulty in identifying the holdings and the holding management, but where subsistence farming predominates it is not always easy to recognize and to define
7 ngle holdings, and judgement has to be arbitrary in many ways. Because of ie dependence of cropping and use of labour on the requirements of the ousehold, it seems advisable to include in one holding the land and animals bat belong to a household at a given time, iousehold have the right to use. The communalor which the members of the kitchen is usually the defining :haracteristic of the household itself. Within a farm, the various activities are connected functionally in ways :hat can vary from the simple to the very complicated, but in all of them there is a system. Most of the crop and animal activities on a farm are closely related to each other by the common employment of labour, land, and capital. Cultural practices are also closely livestock combinations. ChanL.-s in any linked with the various crop and one practice usually require a reappraisal of all the others, since the former balance will have been upset. Each farm is thus a cohesive, adjusted system in itself. 'Men wish to make their activities "efficient" rather than wasteful' (Herskovits 1964, p. 512). Consequently, 'it can be taken as cross-culturally acceptable that, on the whole, the individual tends to maximize his satisfactions in terms of the choices he makes. Where the gap between utility and disutility is appreciable and the producer. . . is free to make his choice, then, other things being equal, he will make his choice in terms of utility rather than disutility' (Herskovits 1965, p. 18). In describing farming systems and their characteristics, we start therefore with the assumption that they did not come about by chance and that there is always a reason why farming in a specific case is carried out in one way rather than another. Having thus defined the geographical region with which we are concerned, the farming systems, and the meaning of the wordfarm or holding, we proceed to a consideration of some of the more important in the tropical environment and their implications characteristics of farming for farm management.
INTRODUCTION
FARMS in the tropics, like farms everywhere, are adjusted to the circumstances of the environment in which farmers find themselves. This is particularly true of smallholder agriculture, which predominates in most tropical areas. The levels of production technology and the resources available to smallholders are such that they have only a limited ability to change or adapt their practices to the rigours and pussibilities of each situation. The growing literature on agricultural systems underlines the remarkable ingenuity with which farmers and pastoralists have devised workable systems of production for almost all of the many climatic and social situations that occur within the tropical zone. A full appreciation of the extent and nature of producers' adaptation is best obtained through consideration of the main features both of the natural environment and of some critical aspects of tropical farming. which partly arise from the natural conditions of production.
2.1.
In this section, attention is drawn to some of the most important features of the natural environment, which can be classified into three main groups covering climatic, soil, and biological conditions. Whilst some valid generalizations can be made about them, it is important to stress at the outset that the tropical areas are characterized by enormous variability, which stems mainly from the different manifestations of climatic effects on the one hand and altitude on th2 other. Large stretches of uniform ecological conditions, like the North American prairies, or the Black Soil belt of continental U.S.S.R., a.e not found in the tropics. Consequently, generalizations about tropical conditions such as are made here are inevitably beset with oversimplifications. Interpretation of their importance in particular circumstances requires great care.
SOME GENERAL CHARACTERISTICS OF FARMING 2.1.1. Climatic influences Climate is the main distinguishing feature of those areas that are usually considered as falling within the tropical zone, a term which, as Fig. 1.1 shows, generally refers more to places with particular types of climate than to those areas that fall between the two confining lines of latitude. Such areas are normally hot by comparison with other parts of the world. Except where high altitudes modify the effect substantially, coldness is seldom or never a limiting factor of agricultural production. In some parts of the tropics, especially the humid forest zones, and in irrigation areas, crop production is possible at every time of the year. Where the water supply and the quality of the soils are sufficient to make full use of the advantages created by an ample and even supply of solar energy, the production potential of tropical farming is extraordinarily high, as testified by record harvests, which according to Bradfield (1968) attain the following levels: (1) Rice in Madras, India: 13 400 kg per hectare in one harvest. (2) Rice and three sorghum crops per hectare in twelve months in the Philippines: total yield 5000 kg of rice, 16 500 kg of sorghum. (3) Sweet potatoes in the Philippines: 30 tons in 90 days. Most of the tropics, how aver, experience a seasonal pattern of rainfall, with distinct periods of high and low levels of precipitation. This means that soil moisture levels at certain periods are not sufficiently high to sustain plant growth. Moreover, effective rainfall in the tropics isoften less than in temperate areas, since evaporation and transpiration losses are much more severe in the heat. Apart from giving rise to a seasonal shortage of soil moisture in some areas, rainfall patterns are also characterized by their great variation between one year and another, and between adjacent places. The degree of this variation can be greatest in those places where mean annual rainfall islowest. However, in large parts of the tropical world, aconsiderable degree of uncertainty attaches to the level of rain that will fall in any one season, with both drought and flood as serious possibilities that occur all too frequently. Moreover, when the rains do come, they often occur in a series of very heavy storms, in which a lot of rain falls within a short time. Wherever it occurs, this pattern of precipitation, with its unpredictability and severity, increases the problems and uncertainties of farming. Another main feature of the tropical climate is the importance of wind. The impact on temperature and rainfall of some of the main rain-bearing winds, especially the monsoons that are associated with the large continental land masses, is well known, as are the more sensational hurricanes and typhoons. However, winds blowing off the hotter land masses-especially the larger deserts--can have dramatic consequences of their own because of their intense drying effect. In many tropical areas the dry season is a time of
SOME GENERAL CHARACTERISTICS OF FARMING fairly high winds, which not only reduce the opportunities for plant growth, but also bring a threat of wind-induced erosion. A number of other effects of climate on tropical agriculture can be important. Within the tropical latitudes, there is relatively little variation in day length between one part of the year and another, and on the equator the length of all days of the year is virtually the same. Within the day, the intensity of solar radiation and the air temperature may undergo fairly rapid changes, although both effects are less pronounced in places where and at times when the relative humidity of the atmosphere is high, and when there is a high incidence of cloudiness. This means that, whilst the hours of daylight are limited to something like the length of a typical day in early spring or autumn in temperate latitudes (i.e. 13-15 hours), the heat and intensity of radiation may at times be so great that animals and workers seek the shade, and can pursue productive activities for only a limited period. Altitude reduces the temperature effect, but can intensify the radiation problem beyond comfortable levels. Climate exerts its influence on farming not only through the control that it imposes on the timing and volume of plant growth, but also through its effect on livestock production. Excessive heat and humidity can upset animal physiology, causing distress, reducing appetites, and limiting productivity. At the same time, high temperature and/or humidity can make the conservation of harvested produce very difficult, leading to high losses amongst crop reserves kept for use outside the productive season. 2.1.2. Soil conditions Tropical soils vary enormously in type and suitability for farming. Some are derived from materials that are the products of fairly recent geological events, being volcanic or sedimentary in origin, whilst others are derived from very old igneous material. Despite their great variation, however., a few useful generalizations about tropical soils can be made. Perhaps their most widespread feature is the extent to which the pattern of precipitation in sudden heavy rainstorms has led to extensive leaching, so that most of the soluble plant nutrients have been carried away. Consequently many soils, particularly in high-rainfall areas, have a low level of natural potential fertility. Also detracting from their fertility status is the low level of organic material that is typical of most of them. Under warm tropical conditions, most organic matter breaks down quickly whenever it is moist, resulting in a rapid but short-term release of nutrients in areas of alternating wet and dry seasons. Another common characteristic that detracts from their fertility is the poor structure of many tropical soils. Because of this feature, they may not be resilient under intensive cultivation, and this consequently increases the high
10
IN A TROPICAL ENVIRONMENT 11 risk of water and wind erosion. Erosion by surface water is a constant danger, especially because of the intensity of tropical storms. 2.1.3. Biological environmeni Partly within the sphere of human control is the b.o!egcal environment: the natural or introduced species of all kinds, which, in tropical conditions, are particularly numerous and varied. In the warmth, many kinds of weeds, fungi, and parasites survive very well wherever moisture levels are high enough, so that domesticated plants and livestock are constantly subject to competition or parasitic infection and disease. This presents many problems. Not only iscropping beset by the ever present need for clearing and weeding, and the removal of secondary vegetation, but all farm production processes are exposed to high risks from pests and diseases of many kinds. Some of the principle scourges, like the tsetse fly and ticks, are constantly present, as are some of the virus and bacterial agents that cause serious crop and livestock diseases. Others are not a constan, threat, but may build up to plague proportions in a very short time. This is especially true of insect pests, but applies also to many other species, incli.ding rodents and even small birds, which can wreak sudden and tremendous havoc amongst growing and harvested crops. In a pre-technical society, the farmers' main palliatives for problems of this kind are adaptation or avoidance. Many of the effects of other biological agents cannot be avoided, but traditional husbandry methods contain many risk-spreading features that are designed particularly to overcome problems of this kind. 2.2. Some critical aspects of tropical farming Farmers who work under the circumstances depicted in the l&st section are faced with management problems and dilemmas of four main kinds, which derive from certain features of their natural environment and are reflected in the systems of farming that they follow. 2.2.1. Problems of soilfertility First, the numerous problems of maintaining soil fertility in a humid tropical environment have to be considered. The importance of soil fertility and conservation is clearly recognized by most tropical farmers, and practically every system of farming involves measures that are specially adopted to increase the fertility of land that is being cropped. Essentially, the fertility problem is one of making available in the upper layers of the soil sufficient nutrients (in the right condition of solubility and availability) to allow plants to grow. Under natural conditions, an equilibrium tends to develop in which losses through leaching and surface run-off are made up by the chemical
12
breakdown of soil minerals and the supply of available nutrients from lower soil layers that deep-rooted plants maintain. In contrast to this natural situation, cropping or other forms of farming are almost always exploitive, in that they involve the removal of nutrient-bearing product materials from the places where they were produced and use of them elsewhere. Often the residues remaining after their use by humans are not returned to the land, but are disposed of so as to be carried away in drainage water. Natural replenishment may in some c-rcumstances keep pace with this loss, whilst farmers may 'concentrate' fertility by gathering crop residues, manure, or natural vegetation in those places where they intend to undertake cropping. However, all of these systems are wasteful in the sense that through the deliberate removal and transplantation of produce, and by the involuntary disturbance of the natural order in ways that can increase the rate of loss, nutrients are removed from the soil and are not replaced. Sooner or later it is inevitable that such a system must lead to either temporary or permanent reductions in soil fertility. Problems of this nature are much more acute in the tropics than elsewhere, because in the warm climate, especially where it is combined with long periods of high humidity, the chemical and biological processes tak: place very rapidly. The higher the loss of nutrients through leaching, the more easily are the structural properties of the soil destroyed. Farmers in almost all kinds of situation pay some attention to the problems of soil fertility. Under some conditions, indeed, the problem can become such a major preoccupation that whole farming systems and practices are substantially modified to balance the conflicting and imperative short-term needs of maintaining both production and residual fertility levels. Generally, however, the kinds of measure that can be adopted to protect soil fertility and prevent erosion require a great deal of labour for their construction and development. Obviously, this can impede their progress. 2.2.2. Coping with risk and uncertainty Additional major farm-management problems derive from the high levels of production risk and uncertainty that surround agricultural production in the tropics, especially at low levels of technology. In smallholder farming, the objective of every household, wh:ther commercialized or not, is to produce most of the basic food and other natural-product needs of the family. It is therefore essential that a minimum amount of food of different kinds should be available throughout the year. In meeting this objective, farmers have not only to consider the production risks that may arise from climatic variation and the effects of pests and diseases; they have to allow also for the effect of these same influences on stored produce, where destruction and deterioration can be very rapid. The particularly dramatic nature ofvariations of this kind that are associated
IN A TROPICAL ENVIRONMENT
13
vith tropical production causes farmers to adopt a number of devices to pread their risks and ensure a constant supply of produce, which, because )f consumer preference, must be as fresh as possible. Diversification of pro. luction, so as to grow a range of crops, is a typical risk-spreading device Yellowed by farmers the world over. The planting of a particular crop atL different times over an extended planting period, which will allow some pro.. duce to be obtained in either an early or a late season, whichever should occur, and the combination of different species in crop mixtures are both practices with an element of risk-avoidance in their adoption. From the point of view of farm operation and resource allocation, the need to provide for a minimal production in the very worst possible season can lead to an apparent misallocation of resources, in that the target level:; of production for a particular commodity may, in a normal or good year, be such as to provide a surplus, which will not always find a market. At the same time, some crops may be grown that are not really the best for a particular production situation. Thus, to overcome the risk problem, farmers may need to employ more resources for subsistence production than would normally be necessary, and they may be obliged to grow for themselves some products for which their land or skills are not well adapted. 2.2.3. Low labour productivity Closely related to the problems just referred to is the low productivity of agricultural labour in the tropics, whether measured in hours per available worker, or by the physical performance per hour of work, or by the return per hour of work. In the majority of situations, even where land is limited in supply and advanced material inputs like mineral fertilizer and pesticides are not used, the most critical farming input is the labour of the farmer and .his family. Most of the work that they have to do is both tedious and strenuous, so not surprisingly farmers show a marked preference for leisure: this has often been demo-strated by the extent to which hired workers are employed to replace them cnce incomes have exceeded very modest levels. Often, however, despite the use of seasonal labour, farmers are severely limited in their scope for increasing output by making higher labour inputs, through the shortness of the growing season during which useful farm work can be done, the low levels of technology at which they typically operate, and the heavy demands in both time and effort that domestic duties make upon certain members of the household. Where, as very often is the case, labour is the main variable input into farming, its availability can be the limiting factor to production, especially at certain critical times of the year. Consequently farmers seek-and usually quickly take up-new crops, production methods, and product combinations that offer higher returns to labour and are also acceptably free from risks. It
14
is, therefore, those processes that offer higher returns on labour that many farmers actively seek, and the development of small-scale production must largely depend on the discovery and adaptation of practices that will substantially improve returns on work. From the farmer's point of view, labour productivity and work rationalization are considerations of great importance. 2.2.4. The problem of seasonality Farm operation and labour productivity a7e further hindered by the acute seasonality of many climates, in which wide differences exist between the wet and dry seasons, or seasons with and without irrigation water. Where the water is available in only a few months, all of the crop work needs to be carried out very quickly, beginning often only when the rain softens the soil, and continuing with cultivation, planting, and weeding for as long as crop growth is possible. The urgent need for high labour inputs at this time coincides often with the period when fresh food supplies are most scarce, and low nutritional levels inhibit workers and draught animals from making as great exertions as they might wish. Certainly this season contrasts strongly with the other months, in which little work to assist crop production can be carried out, and harvesting and maintenance may be the main activities. At such times, a lot of labour cannot find productive work. Similar problems arise in livestock management. During the rainy season, fodder is abundant and of high quality, and animals can grow, produce milk and other products, recover condition and health, and are capable of sustained work. In the dry weather, the grazing deteriorates. Animals may lose condition, whilst their productivity declines, often to nil, as lactating animals dry off. One consequence of the shortness of the cropping season, and the low productivity of manual and animal power at this time, is that the mechanization of cropping can bring high returns, if propeely managed. Usually, where they can be kept, draught animals can bring ar. improvement over manual labour. Draught animals, however, suffer in conditions of high humidity and heat and the accompanying diseases. They also find it much more difficult than tractors to cope with rapidly hardening soils. The greater speed with which operations can be performed by tractors, and the fact that they can work at times of the yerr when soil conditions prevent hand workers or draught animals from carrying out such tasks as cultivation are of crucial importance to the volume of production where the best use must be made of a small amount of rain. It must not be assumed from this that the use of tractors in the tropics is always more practicable than use of the hand hoe or draught animals, but in the context of farm-management characteristics it is important to note that the physical rate of substitution of traction animals by tractors is comparatively high.
IN A TROPICAL ENVIRONMENT
15
cultivation is the name we use for agricultural systems that involve an alternation between cropping for a few years on selected and cleared plots and a lengthy period when the soil is rested. Cultivation consequently shifts within an area that is otherwise covered by natural vegetation. The intensity of shifting cultivation varies widely. A relatively simple and appropriate criterion ofland-use intensity is the relation between the period of cultivation and the period of fallow (see 1.2.2). The shorter the time between the cultivation periods, or the greater the annually cultivated area in relation to the total area, the more stationary does farming become. In the course of this development, short-fallow systems replace long-fallow systems. If the characteristic R is more than 30, i.e. if 30 per cent of the arable and temporarily used land is cultivated annually, we no longer speak of shifting cultivation but of semi-permanent farming.
SHIFTING
17
to several criteria: here a distinction is made between vegetation systems, migration systems, rotation systems, clearance systems, cropping systems, and tool systems. 3.2.1. Vegetation systems We may distinguish on the basis of vegetation between the shifting cultivation of the forest, of the bush savannas, and of the grassland areas. Shifting cultivation in the rainforests is typical of rain-fed farming in humid areas of low population density. Rotations with bush fallowing predominate in the African savannas. Alternation between cultivation and wild grass vegetation is found in some tropical high-altitude areas, in semi-dry climates, and not least where the reduction of the fallow period has led to the growth, while the ground is resting, of largely useless grasses (frequently Imperata species) instead of bush fororest. Spencer's classification (1966, p. 181) goes even further. He distinguishes between shifting cultivation in (1) virgin forests, (2) mature regenerated forests, frequently called secondary forests, (3) immature regenerated forests, (4) scrub tree growth, and (5) grassland. 3.2.2. Migration systems The continual movement of cropping results, in most systems with shifting cultivation, in a slow migration of the population. The cultivated plots move slowly away from the previous clearing and the vicinity of the hut. At the same time the cost of transporting the harvest increases, especially where root crops are grown. At a certain distance, it becomes advantageous to build a new hut near the field instead of carrying the harvest such a long way. Shifting of plots is made easier by the very nature of tropical housing; it is often more economic to build a new hut than to repair an old one. The frequency of hut changing varies greatly. In the rainforest, in some cases, a change occurs every two or three years; in the savanna less frequently, sometimes only once every one or two decades. Allan (1967, p. 4) describes the process in central Africa as follows: A village or other group within such a concentration will use up the accessible land in its neighbourhood before the area first cultivated has had time to regenerate and a move must then be made to regenerated land... A series of pictures or aerial photographs made at suitable intervals would show a gradual movement of the larger community across the countryside in a manner suggestivc of the progress of an amoeba on a microscope slide. By the direction of the migration we can distinguish, according to Spencer (1966, p. 177), between (1) random shifts, (2) linear shifts onto newly cleared adjoining lands, which result in the steady and progressive clearing of mature forests, (3) cyclic shifts within a given area claimed by the group or family,
3
TABLE
3.1
Sarawak Guatemala Liberia Sierra Leone Assam Sumatra Philippines Nigeria (a) Umuahia (b) Alayi Central Congo
Forest
Moist evergreen forest zone 1 >12 1 1-2 11 2 2 2-4 11 I1 2-3 >4 8-15 8 10-12 10-16 8-10 4-7 7 10-15
Early abandonment of land necessary to prevent invasion of Imperata 'Ando' type soi" Grasses (especially Chamwopodium spp.) invade with excessive cultivation
I 2
5 <7
Rice-millet, Forest maize, rice About 2300 Rice, root Forest crops 2500 Rice, root Forest crops, maize About 2300 Yams, maize, Acioa barteri manioc About 2300 Yams, maize, Macrobolobium manioc spp. 1800 Rice, maize, Forest manioc
Moist semi-deciduous and dry forest zone (including humid zone of derived savanna) West Africa North Burma West Nile, Uganda Abeokuta, Nigeria liesha, Nigeria Central Uganda Ivory Coast Northern Rhodesia 1500-2000 1300-1800 1400 About 1300 About About About About 1300 1300 1300 1300 Maize, manioc Hill rice Eleusine, sorghum, simsim, maize Moist semideciduous forest Grassland and pine forest Grass, mainly Selaria spp. Thicket Thicket Elephant grass Elephant grass Thicket 2 3 3 6-12 6-7 8 3 6-12 2-4 C-12 5 2-3 8-15 3 2 I 9 10 3 4-5 2 6 25 33a 18 30 Kochin Hills area at about 2000 m Refers to "outside' fields Soil derived from tertiary ,--nd; zvidence of nitrogen deficiency Soil derived from granite
2-3 1
4-6 <2
grass
Miombo woodland
'Semi-permanent cultivation in our terminology. Source: Nye and Greenland (1961, p. 128).
20
and (4) reduced '.ic shifts, brought about by increasing population densities, which might finally lead to sedentary living. 3.2.3. Rotation systems Shifting farming is practised not only by migrating cultivators but also by scdentary cultivators. In each case, however, cropping and fallowing alternate, and this alternation can have an irregular or regular character. In the case of a regular sequence, a definite number of fallow years follows a definite number of years of cultivation. A further distinguishing characteristic of the rotational pattern is whether land is selected for use in small plots (where, for example, cultivation is restricted to termite mounds, moist valley bottoms, etc.) or whether continuous tracts of land are cultivated. The rotation systems in rainforest agriculture are mostly simple. Two to four years of cultivation are followed by one to three decades of fallowing. Miracle (1967) calls such patterns 'classical long-fallow systems'. In the savannas of Africa, on the other hand, we often find complicated rotation systems in which short-term fallow periods of 1-2 years, medium-term fallow periods of 3-5 years, and long-term fallow periods of 6 years or more alternate in a single cycle of land use. These combinations of classical, longfallow systems and short-fallow systems also predominate in semi-permanent farming. The shcrt-term fallow periods can usually be attributed to unexpected lack of labour during the cultivation period, whereas increasing weed growth and decreasing fertility lead to medium-term and long-term fallow periods (see 4.3.2). An example will illustrate this type of situation. In the Kilombero Valley, Tanzania, where shifting cultivation of rice is pursued in an area where elephant grass dominates the natural vegetation (Baum 1968, pp. 32-4), the following rotation systems have evolved according to the settlement density: I. Rotation Cycle of 45 Years When there is an abundance of land, cultivation takes place within two rotations, one shorter and one longer. As a rule, the shorter rotation runs as follows: 3 years of rice-growing (with one annual harvest) 3 years of grass-fallow. Several families put their rice plots together into a block, in order to obtain better protection against vermin. They alternate between two blocks, one canying rice, the other one lying fallow for about 15 years. Thereafter, the families move into a neighbouring area where the 'short rotation' is likewise practised. In a few out of the way parts of the inner valley, families even move into a third area. Then they move back to the first area, which has then lain fallow for 30 years. The short rotation of 6 years is thus supplemented by the long rotation over a period of 45 years. This results in a total R value of 17, i.e. 17 per cent of the cultivated land is used annually [see Fig. 3.1, part I]. The total area composed of the three zones farmed by the families is occupied by these families in accordance with traditional land rights. Immigrants may not settle there without
SHIFTING CULTIVATION SYSTEMS 21 permission. As a rule, the household members know their plots in each of the three areas and cultivate them in regular cycles. II and Ill. Rotation Cycle of 30 Years When there is not enough land available for the families to move between three areas, the 'long rotation' is limited to two areas. The families live in each
l15 years
ear
year
15 years
15 years
,3ye''rsfa"low
yearsFields
3years cropping
Fi. 3.1. Rotations of shifting cultivators in the Kilombero Valley, Tanzania. I = short rotation: 3 years cropping, 3 years fallow long rotation: 15 years occupation, 30 years unoccupied total duration of rotation c 'cle: 45 years, frequent changing of huts 11 -total duration of rotation: .)0 years, housing III -:otal duration of rotation: 30 years, semi-stationary stationary housing IV = short rotation only: 3 years cropping, 3 years fallow one for 15 years and rotate 2-3 years of rice in a total R value of 25, i.e. 25 per cent ofwith 2-3 years of fallow. This results the occupied land is cultivated per annum in contrast to 17 per cent of the land with three cropping areas [see Fig. 3.1, part II].
22
SHIFTING CULTIVATION SYSTEMS Fig. 3.1, part III represents a modification of this type of shifting cultivation in view of a higher degree of stationariness. The settlers have a permanent dwelling-place, as, for example, near the road. The fields, however, are still rotated in 15 year cycles, i.e. the land on one side of the settlement is used for 15 years, alternating 3 years of cultivation and 3 years of fallow. Subsequently the fields on the other side of the settlement are cultivated. IV. Semi-permanent Farming The growing population density is conducive to the 'long rotations' being abandoned. Where land becomes increasingly scarce, two to three years of cultivation are followed by the same number of fallow years. The dwelling-place is permanent, either next to the road or in some other favourable location. Fixed and recognized boundaries are established for holdings and fields. Permanent and generally acknowledged rights of use are derived from the traditional land laws. The typical value for this intensity of land use is R = 50, which results when the years of cultivation are followed by the same number of fallow
years.
3.2.4. Clearancesystems The type of clearance work and the order it takes vary according to the vegetation that is to be cleared, the distribution of rainfall, the crops grown, the available tools, and the cultural background of the population. Following Miracle (1967, pp. 41-7), we classify the forms clearance takes as follows: (a) Burn andplant. One of the simplest forms merely involves burning off a thick and dry secondary vegetation. Immediately after burning, maize is planted, and matures before the secondary vegetation has recovered from the burning sufficiently to produce brushwood and leaves. Cook (1921) describes shifting cultivation of this kind in his 'Milpa agriculture' in Central America. (b) Burn, hoe and cut,plant. The burn, hoe and cut, plant system is reported only in savanna areas, where fire effectively eliminates most of the vegetation. Cutting is limited to the remaining trees and bushes. Except for the reversal of the burning and hoeing operations, this method of agriculture is much like the cut, burn, plant system. (c) Cut, burn, plant. The cut, burn, plant method of agriculture is the most common one reported. Typically, vegetation is cut-usually towards the end of the dry season-allowed to dry for a while, and then is burnt as the rainy season approaches and the soil must be prepared for planting. The cut, burn, plant method is divided by Miracle into ten major subdivisions, according to tl'e amount of cutting and burning done and whether further cultivation operatns are employed. (d) Cut, plant, burn. The cut, plant, burn technique is the only system in which the vegetation residue left in the fields is burned after crops are growing, and is reported only in forest areas, or areas near the forest belt which
would possibly still be forest except for the activities of man. Usually the crop occupying the field when it is fired is bananas, plantain, or manioc. In such systems, clearing is often done over a relatively long period, and crops are planted as soon as the smaller vegetation has been cut. By the time all trees to be removed have been felled, the planted crops have established themselves, and the field is given a burning, which apparently inflicts little or no damage upon the growing crops. (e) Cut, bury refuse in mounds, plant. Various sorts of composting operations are reported as an integral part of field preparation in some of the southern African savannas. After vegetation is cut, it is buried in mounds in the field, or perhaps the debris is first burnt and the ashes are buried. In other cases, the debris is buried in mounds and then ignited, possibly because of the beneficial effect of heat on soil fertility (see Nye and Greenland 1961, p. 71). (f) Cut, add extra wood, burn, plant, hoe. The cut, add extra wood, burn plant, hoe method, generally referred to as chitemene in Zambia, involves cutting bush from a greater area than is to be planted, in order to obtain a hotter fire and a large amount of ash on the land that is to be sown. Among the tribes practising chitemene, the amount of land cleared may be from five to twenty times as much as that cultivated, and the cut debris may be piled in one area forming a large circle or distributed into two or more smaller circles. (g) Cut, ivait one season, plant (forest). Several tribes in the Congo Basin partially cut vegetation, plant bananas, and then wait a season before completing the clearing operations and planting other crops among the established bananas. (h) Killing trees by ringing, ridging, planting. This series of operations, additional to those mentioned by Miracle, is practised in the dry forests of southern Tanzania. Ringing replaces clearance work. When the trees have lost their leaves the light can reach the ground, and the leaves that have fallen off and vegetation on the ground are dug into the ridges. The trees are never felled; they are left as dead, standing timber. Shifting systems may also be classified according to how complete the removal of the vegetation has been (Spencer 1966, p. 181). Occasionally the farmers restrict themselves to ringing or creating clearings by felling some trees. If fire does not consume the trunks and the larger branches, these are left lying on the ground, and cultivation is carried on among the debris. In other cases, however, the vegetation is thoroughly burnt off, and in some cases a second fire may be necessary. Differences in clearing technique appear to be fairly well explained by variations in ecological conditions. The simplest and least intensive techniques are found in the humid forest belt. The number of operations and amount of labour required are greater in the savanna areas (Miracle 1967, p. 159).
24
as farming with annual Shifting cultivation is almost exclusively carried on all rice economies or above find and biennial crops. In humid climates we sweet potatoes, yams, manioc, root-crop economies where the emphasis is on of the savannas, climates and taro. Widespread, and extending into the drier economies, mixed numerous are maize and millet economies. In between are evidence ample is There which are based on root crops as well as grain crops. in their fixed traditionally to show that shifting cultivators are not static or of the Most 1966). Spencer cropping patterns (Miracle 1967; Baum 1968; introduced have been crops or varieties grown by African shifting cultivators According to administrators. and settlers, in the last 300 years by traders, cropping with root crops to Spencer (1966, p. 169), a cbange from vegetative Where shifting cultivation grain cropping with rice can be observed in Asia. complementary homesteads, permanent is practised by sedentary farmers with a permanent trees, fruit of activities are usually to be found: the growing garden, plots with irrigated rice, a fish-pond. 3.2.6. Tool systems practised without Shifting cultivation in the rainforest is still occasionally ashes. The axe the in sown cultivation implements; after burning off, seed is as is prepared, is ground and the matchet are the main tools. Where the suggestion the Following usually the case, tools for cultivation are required. digging-stick systems, between distinguish can of Spencer (1966, p. 131), we systems are found in the hoe systems, and plough systems. Digging-stick the savannas, particularly humid rainforests. Hoe systems predominate in years, since in these cirseveral over where the cultivation period stretches are necessary. The plough cumstances soil cultivation and hoeing of weeds where an unused grass system can be employed only in shifting cultivation vegetation grows in the fallow periods.
3.3.
to the next, are characterThe various shifting systemq, from one continent The relative importance of ized by some common organizational features. conditions, but they are these features changes according to the local land special interest in this respect recognizable as tendencies in almost all cases. Of principles, and the cropping are the spatial organization of cropping, the and of the husbandry, organization of the fertilizer economy, of animal labour economy. 3.3.1. The spatial organization of cropping is permanently cropped The farmer with a stationary home and land that for his crops; that is, he tries tries to create favourable growing conditions
to control nature. Shifting cultivators, on the other hand, are usually highly skilled at adapting their cropping practices to the environment in which they are working. Four aspects of adaptation are important: (1) The choice of the plot of land to be cultivated. The main determining factors are soil fertility, the effort of clearing the vegetation, the amount of weed growth, the danger of vermin, and location in relation to tile road and a source of water. On fertile soils, a smaller plot is more able to cover food needs than on poor soils. Fertile soils, however, are by no means preferred in all cases. Expenditure of labour per unit of production is a much more decisive factor. If, with the same effort, poor land as good land can be cleared and cultivated,three times as much even with only half the yield per unit, there is still a net gain (Allan 1967, p. 94). (2) The choice of crops. (3) The organization of cropping in mixed cropping, phased planting, and crop rotations. (4) The arrangement of short-term, middle-term, and long-term fallows. The decisive factor in balancing these requirements is in most cases the very precise knowledge of soils and plants held by the shifting cultivators. Allan (1967, p. 5) writes on this subject, about central Africa: The 'shifting' cultivator has an understanding of needs. He can rate the fertility of a piece of land andhis environment suited to his of his crops by the vegetation which covers it and its suitability for one or other of the soil; and he can assess the 'staying power' by the physical characteristics of a soil, the number of seasons for which it can be cropped with results, and the number of seasons for which it must be rested before satisfactory such results can be obtained again. His indicator of initial fertility is the climax vegetation and his index of returning fertility is the succession of vegetational phases that follows his knowledge is precise and remarkably complete. cultivation. In many cases hundreds of names of trees, grasses, and other plants He has a vocabulary of vegetation associations by specific terms. This fund and he identifies particular of ecological knowledge is the basis of 'shifting' cultivation. Conklin (1957) tells us the same about the rainforest in the Philippines, where the Hanun6o of Mindoro distinguish more than 1600 different plant types, which is a finer classification than that employed by systematic botanists, and includes the astounding number of 430 vivid description of what a Hanun6o swidden cultivated species. Conklin's in full swing looks like (pp. 18-19) gives an excellent picture of the degree :o which this agriculture apes the generalized diversity of the jungle that it temporarily replaces: Hanun6o agriculture emphasises the intercropping of many types of domesticated plants. During the late rice-growing seasons, a cross-section view of a new [plot] illustrates the complexity of this type of swidden cropping (which contrasts remarkably with the type of field cropping more familiar to temperate zone farmers). At the sides and against the swidden fences there isfound an association
25
26
dominated by low-climbing or sprawling legumes (asparagus beans, sieva beans, hyacinth beans, string beans and cowpeas). As one goes out into the center of the swidden, one passes through an association dominated by ripening grain crops, but also including numerous maturing root crops, shrub legumes and tree crops. Poleclimbing yam vines, heart-shaped taro leaves, ground-hugging sweet potato vines, and slhublike manioc stems are the only visible signs of the large store of starch staples which is building up underground, while the grain crops fruit a metre or so above the swidden floor before giving way to the more widelyspaced and less-rapidly maturing tree crops. Besides the systematic adaptation to variations of soil within a plot described by Conklin, we find in most shifting systems that cultivation is often different on those plots that are especially suited to individual crops. Miracle (1957, p. 13) describes this type of cultivation outside the rotation systems on the main piots: Cultivated fruits, vegetables, rice, yams, taro, condiments and tobacco are outside the crop sequences and each of these are in a small environment especially favourable to them. Oil-seed, watermelons, tomatoes, mock tomatoes (Solanurn spp., grown only for their leaves), roselle, egg plant, red pepper, mangoes, and papayas are grown in the courtyard where they are free from weeds and fertilized by ash and refuse. Bananas are grown on small mounds; and okra, yams, peanuts (as a vegetable), and various vegetables are grown on broken ridges in the court. yard. Rice, bananas, taro, sesame js a vegetable), maize and other, are grown on old refuse heaps. Moist, shady spots under the kitchen eaves are used as tobacco nurseries. Ash accumulations found outside the courtyard, especially those resulting from the burning of stumps during the clearing of a field, are used for tobacco and various vegetables. At the base of trees that have been killed, climbing varieties of cowpeas, lima beans, and oil-seed gourds plus bottle gourds and calabashes, are planted. Living trees are used to support yam. Uninhabited term.e mounds are flattened and planted with white sorghum, rice and cowpe as The attempt to combine cash cropping with a regular supply of foods for household use leads, in most shifting systems, to diversified production units. A shifting holding usually consists of a number of distinct components: (1) The dwelling-place consists of the huts and the surrounding yard. The materials used in constructing the huts depend on the degree ofpermanency in housing. If the hut is to be used for less than about six years, then wood and grass are used. If it is to be used for more than six years, then clay is the main building material. Farmers who expect to live permanently in a given place usually endeavour to build stone houses. (2) Usually a garden plot near the dwellings is cultivated with vegetables, gourds, bananas, etc., as long as the family lives in one place. The longer the family lives in one place, the greater is the number of fruit trees grown near the yard. (3) The differentiation of production on the outlying fields ensures that each is cultivated with the crop best suited to it, partly by mixed cropping and rotations, and partly by having several plots on different soils.
27
(4) In addition to cropping, shifting cultivators gather 'wild' products from the surrounding fallow. These are usually not truly wild plants, but the remains of domestic plants within a 'tumbledown' fallow. (5) If the farmers live in permanent dwellings, or move only within a limited area, the shifting cultivation is generally supplemented by other major food-producing activities, including planting of tree crops, permanent gardens, permanent irrigation fields, and fish-ponds.
Two long lines of huts on either side of a track Protected' vegetable g ptthe tos on T -on p atied O el ld e Forest follows l 1 .. s tt lo trets Permanent forest b on e thillsop
bananas
ss tr
dAlluviumi
s//4e~_,e~tor-taros
Fic. 3.2. Diagrammatic cross-section of Pene Niolo, Congo-Kinshasa, and its vicinity (1949) (from Dumont 1966, p. 28).
t eeome i Pen o o haeatteiae subisenened. ulars An example of land use in shifting holdings is shown in Fig. 3.2. A crosssection of Pene Niolo, Congo-Kinshasa, and its vicinity shows a typical ut, some bananas catena: huts on the ridge, vegetable gardens close to t and oil-palms near to the garden, field crops and forest fallows on the slope, and some continuous cropping on the alluvial soils in the valley bottom. The oil-palms on the opposite slope indicate that cash cropping with permanent crops is practised in addition to the production of field crops mainly for subsistence needs. Cultivators in Pene Niolo have at the same time become sedentary, so that any change of dwelling-place is limited to moving from one ridge to the other. 3.3.2. Croppingprinciples The adaptability of shifting systems is closely related to three principles of cultivation, which are linked in varioueways. These are: mixed cropping, phased planting, and crop rotationr', i.e. alternating crop mixtures which are regarded by de Schlippe (1956) as pseudo-rotations.The three principles of cultivation also apply in the semi-permanent or permanent farming of indigenous cultivators, but not in anything like as much variety as is found in shifting systems, which are stimulated by the need to adapt. (a) Mixed cropping. Mixed cropping, i.e. the simultaneous growth of two
28
or more useful plants on the same plot, is an important characteristic of shifting cultivation.' The technique of row cultivation is only rarely applied: the plot is usually uneven, and tree stumps, roots, termite mounds, and low areas with poor drainage would interfere with the rows. A worthwhile labour economy is not achieved, since the farmers in any case rarely work with equipment drawn by animals. Row cultivation, moreover, would not allow the mixture of plants grown on each piece of land to be adapted to the changing soil conditions. In shifting cultivation, therefore, we find predominantly mixed cropping, which is not organized with the help of a formal principle, but is based on the best possible adaptation to the surface condition. Mixed cropping has a number of advantages: (1) It reduces susceptibility to disease and pests, and therefore reduces the yield risk. (2) It allows the adaptation of planting to changing soil conditions. Thus, for example, different crops are grown where there is much ash left after fire clearance from those where there is no ash. The plants grown close round a tree stump differ from those where no stumps are left. (3) It allows cultivation to be adapted to the light and shade requirements of individual crops. Both vertical and horizontal variations of cultivation are facilitated. Thus, for example, medium-height manioc plants grow under bananas, and on the ground pumpkins or beans are found. (4) In smallholder farming, a varied food supply is usually sought. Over and above this, the household relies on a continuous supply of fresh food, since there are insufficient storage facilities and the storage losses are high. Both aims naturally lead to phased planting and mixed cropping. Table 3.2 shows, with an example of the consumption of a shifting cultivator's family, the variety of products that a tropical smallholding produces, and the frequency of consumption, which should coincide where possible with the harvest, because of the limited storage facilities. (5) A further advantage of mixed cropping, which it shares with phased planting, is the soil cover provided. Where there is a vertical arrangement, the rain, for example, falls from the bananas onto the manioc and then onto the beans, and only then does it reach the soil. Mixed cropping may
Some important definitions: Mixed cropping: two or more crops grown simultaneously and intermingled; no row cultivation. Inteiplanting: long-term annual or biennial crops interplanted with short-term annual crops during the early stages of plant development. Infercropping: two or more crops grown simultaneously in alternate rows in the same ground. Interculture: arable crops grown under perennial crops. Relay.planting: the maturing annual crop interplanted with seedlings or seeds of the
following crop.
29
TABLE 3.2
x x x x x x x x x x x x x
Beans (jugo)
Beans (manteiga) Green beans (boer boer) Green beans (nyemba) Green beans (jugo) Manioc leaves Sweet potato leaves Bean leaves of all kinds x x x x x x x x x x x x x x x x x x x x x x x x
x
x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x
Onions Tomatoes Gherkins Egg-plant (2 kinds) Quiabo (Hibiscus esculentus) Groundnuts Sugar-cane Pumpkins Sorghum-cane (ecununga) Collected food (4) Staple foods (animal protein) Grasshoppers Mice and other rodents Caterpillars Larvae of butterflies (5) Additional foods Fungi (9 kinds) Wild fruits (7 kinds) Honey Source: Pbssinger (1967, p. 199).
30
have the same protective effect, therefore, as the original forest or bush vegetation. By virtue of these advantages, mixed cropping often produces higher yields than monocropping. Webster and Wilson (1967, p. 114), on the basis of extensive field research, write the following: In the majority of the above experiments more than one acre of crops in pure
stand was required to produce the yield of one acre of intcrcropping. Since the amount of land that can be cultivated by farmers equipped with hand tools is strictly limited, it seem unlikoly that there would be any general advantage in replacing the traditional practice of mixed cropping by rotations of pure-stand crops, although in certain cases the latter might prove advantageous by reducing the labour of harvesting, or by making it easier to control pests and diseases.
(b) Phasedplanting.It follows necessarily from mixed cropping that crops cover the soil for varying lengths of time. The systematic arrangement of planting dates means that cultivation is phased. The result of this again is a continuous sequence of growth and hp.;vesting. A number of benefits arise from phased planting: (1) The phasing of labour operations allows the work to be distributed over a longer period. (2) The combination of periodic planting and weeding may be labour saving. (3) Planting is not necessarily dependent on renewed soil cultivation. (4) The risk of failure is distributed. The effects of weather variations are counterbalanced, and pest damage is reduced. (5) The phasing of harvests ensures a regular supply of food for the household and minimizes the storage losses. Over a period of several months, a single plot may produce a steady stream of harvestable food, for example, maize, beans, peas, and sweet potatoes. (6) The ground is covered by growing plants for almost the whole year.
Consequently, the soil is protected from water and wind erosion.
(c) Rotations. Shifting cultivation with short-term cycles is rarely connected with crop rotations. Usually one crop or a given mixture of crops is grown for several years. However, when shifting cultivation is carried on with cultivating periods lasting for several years, or even a decade, crop rotations are evolved-a logical continuation of the tendencies that can already be seen in phased cropping. De Schlippe (1956, p. 207) calls the change in the composition of mixed cropping pseudo-rotation, but as a rule it is a matter of carefully considered cropping sequences, which can justifiably be considered as actual rotations. Amongst the great number of forms rotations take, several important features can be observed: (1) In the crop rotations of shifting cultivators, there is rarely a sequence of monocropping. Usually, various mixed crops follow one another, with
31
often only a gradual change in the composition of the mixture. Immediately after clearing, crops that require more fertile soils are chiefly planted, like yams in the rainforest, and cotton, maize, groundnuts, and sesame in the savannas. As the cultivation period proceeds, crops that make little demand on soil fertility are preferred. Biennial or semi-perennial crops like manioc or bananas are frequently grown last of all, because these can survive longer than other crops against the competing weed and bush, on account of their height and the shade they create, before gradually disappearing in the tumbledown fallow. (2) Nearly every plot is different from the point of view of soil type, degree of incline, and distance from the hut, and nearly every field is suited, therefore, to a different crop rotation. The various crop sequences that Richards (1961) found in tile fields and gardens and on old hut sites of the Bemba demonstrate this point (Fig. 3.3). (3) It is rare for one rotation to be applied persistently on a big plot. Division of the plot into several smaller sections is more typical. Each small piece of land then supports a different sequence of crops or mixture of crops. This again is clearly demonstrated in Fig. 3.3. (4) In shifting systems, it is common for the same mixture of crop types or the same crop to be grown on a suitable plot for several consecutive years. When the cultivation period increases, growing 'at intervals' sometimes arises, a system typical of the African savannas, when, for instance, two to three: years of grain crops follow two to three years of cotton. 'Much of the most effective human utilization of the natural habitat', writes Geertz (1963, p. 17), 'consists of changing generalized communities
into more specialized ones
. .
cultivators' farming principles. As land use becomes increasingly intensified and commercial, organizational principles like monocropping and row cultivation are increasingly preferred to the various types of mixed cropping. Agricultural developments and increasing land shortage gradually lead to more systematic rotations and less varied cropping patterns. Only where there is a high degree of intensity of agriculture, as for example in the irrigation agriculture of Taiwan, do mixed cropping and phased cultivation again become preferable, although under the new conditions of rigid organization that occur when they are combined with row cultivation. 3.3.3. Characteristics of the fertilizer economy (a) The regeneration of soil fertility in long-fallow systems. The various features of the organization ofcropping that are applied by shifting cultivators aim primarily at an optimum use of the existing soil fertility. The basic feature of shifting cultivation is the cost-free, effortless regeneration of soil productivity during the fallow period, especially when the fallow consists of
32
forest or bush vegetation. The damper and warmer the climate, and the poorer the soils, the more rapidly the organic substances break down, and the more rapidly the crop yields tend to sink. The first crops are comparatively Fields
Finger millet
and manioc-
1-l
peanuts "- finger
sorghum
sweet potatoes--------fallow
(a)
Finger millet'--
legumes
-manioc
sorghum --. o
(a)
Gardens
sweet potatoes
---
pumpkins
(a)
fallow.----.sorghum
(oneyear
(a)
sweet potatoes
maize and
sweet potatoer
maize and
sweet potatoes
Manioc and
-
sweet potatoes
manioc and
-
sweet potatoes
.manioc
-
maize and
(b)
sweet potatoes
Sweet potatoes
-
sweet potatoes
.sweet potatoes
-
maize
(c)
and maize Old hut sites Maize and pumpkins la) No information maize and pumpkins k tobacco
sweet potatoes
given on what follows.
(b) Maize and sweet potatoes planted year alter year until soil isexhausted. then sweet potatoes alone as the land crop. soil is fertile. then tobacco orsweet potatoes. Maize and pumpkins sown so long as (r)
Fio. 3.3. Crop sequences of the Bemba, Rhodesia (Richards 1961; Miracle 1957, p. 110).
ample, but the harvests in the subsequent years become more and more meagre. On unstable soils of the rainforest, for example, the third crop will usually yield only half, or less than that, of the first. A drastic example of
33
this is provided by Nye and Greenland (1961) (see Fig. 3.4). If further utilization of the plot no longer promises to be rewarding, the shifting cultivator allows it to lie fallow for a fairly long period, in the course of which the soil fertility regenerates. Shifting cultivation thus allows soils poor in nutrients regularly to produce relatively high and certain yields in a climate in which arable farming is always a difficult struggle against nature. This is usually achieved without manure, terracing, or any other means of maintaining fertility that involve a high expenditure of labour. If the fallow periods are
Relative yield
Trial plobs
100 90
B0 70
60 -
20
1 2 3 Maize
Maniac
12 3 Rice
1 2
1 2
Maize
Rice
(Malaya) (Guatemalal
(Congo) [Congo)
Fio. 3.4. Decline in yields under prolonged cropping without fertilizer in tropical rainfrrest areas (data from Nye and Greenland 1961, pp. 734).
long enough, this type of land-use system represents a balanced exploitation of the resources available to farming. The regeneration of soil fertility in the upper soil layers is largely due to the effects of a forest or tree fallow. The deeper tree and scrub roots ensure that nutrients are brought up from the subsoil. The fallen leaves enrich the surface, and the minerals that the rain-water has leached down are brought up again. Thus Beinaert (1941) established in the rainforest of the Congo (Yangambi) that the supply of organic material in the form of leaves, twigs, etc., amounted to 60 tons per hectare per year. Fallow vegetation covering the soil and the comparatively cool temperatures of the covered soil facilitate the regeneration of all that has been taken out of the soil in the years of cultivation. The forest and bush vegetation during the fallow period also protect the soil from erosion
4
34
by heavy rainfall. In addition, the rapid evaporation from the fallow vegetation lessens, or even stops, the leaching process. Provided the fallow periods are long enough, a slash and burn system proves to be in no way harmful to the soil. In fact, the yields after a fallow period with secondary forest are in some cases higher than those after the first clearing. The smallness of the plots, the considerable shading of the soil during the cultivation period that arises from mixed cropping and phased planting, the presence of tree stumps in the field, and other factors mean in addition that balanced shifting cultivation involves little risk of erosion damage. Another advantage of field shifting is the fact that losses through plant diseases remain comparatively slight. (b) Fertilizer application in shifting systems. Since shifting cultivation is associated with a low degree of commercialization, the export of crops over and above the local sale of produce is not a major cause ofthe loss of nutrients. There is, however, a movement of nutrients from the field to the dwellingplace, and therefore some change in the location of soil fertility. Secondly, nutrients are lost during the years of cultivation because of washing out and erosion. Particularly important is the loss of the nutrients that are washed away together with the ashes. The long-term fallow period is the most common method of regenerating the fertility ofsoil. In addition, however, there are various forms of fertilizing: (1) Utilization of hut refuse. The first step in the development of a fertilizer economy consists usually of applying household refuse. It is normally put where it can be useful to plant growth. Thus, for example, fertilization of bananas near a hut with household refuse is common among shifting cultivators. (2) Establishment of gardens on old hut sides. A general practice when the hut has been shifted is to use the old hut site in order to grow crops that demand fertile soil. The movement of the huts means, in fact, that heavily fertilized plots are systematically obtained. This applies particularly where cattle are reared and kept overnight in a kraal near the hut. This practice may be considered as the first step in the development of folding systems (see 4.3.3). (3) Green manuring and use of compost. Where ridge or mound cultivation is practised, it is common to clear tl'e vegetation and to allow it to dry before the soil is worked. It is then Jug right into the centre of the ridge or the mound as a green manure. In the Porka system in Ghana, the cleared vegetation is thrown on heaps and later used as compost (Manshard 1968, p. 90). (4) Working with pit systems. In the pit systems, and the hole system in Tanzania, the fallow vegetation and weeds are collected in pits or in heaps. These are covered with earth, cultivated, and dug over several times. In
35
these shifting systems, the cropping period lasts for up to twelve years, and is followed by a long-term fallow period. (5) Working with heated mounds. With a number of African tribes, the fallow vegetation ;s covered with earth and then set alight and charred, which results in cultivation taking place on little mounds of carbon. One assumes that the motive behind this practice is the change in yield caused by heating the soil, which may convert nutrients into a more freely available form (Allan 1967, pp. 73-4; Brouwer 1965, p. 17; Miracle 1964, p. 84). (6) Ashes as fertilizer. Where fire-clearance methods are used, ash in any case serves as fertilizer. In some regions an ash concentration on small plots is achieved with considerable labour cost. In the chitemene system, as practised in some parts of Zambia and southern Tanzania, branches are transported from the forests in the vicinity to the farmland, piled up a metre thick, and burnt to increase the ash fertilizer. According to Allan (1967, p. 86), for the cultivation of one hectare of finger millet tree ash from 2-5 hectares has to be used. Millet is sown in the ash. This method produces comparatively high and certain yields of 15 quintals per hectare on very poor soils. The main disadvantage is the size of the area from which the branches are taken. The forest requires fifteen to twenty years to regenerate. For the annual cultivation of one hectare of finger millet, a nutrient supply of 30-100 hectares of forest is required. Richards (1961), reporting on the ash system practised by the Bemtna, writes: Cut branches are trimmed enough so that they will lie flat in a pile for burning, and are laid in rows, each man taking care to keep his branches separate l from those of his neighbours. Women later carry the branches to the plot to be cultivated and place them radially, with the smallest diameter to the centre. At the time of Richards's study, branches were cut from an area up to six times as large as that to be planted. Pieces of wood are piled one on the other until the pile is about 65 centimetres high, care being taken to keep the thickness of the pile uniform. The Bemba are said to often single out barren patches of a field as spots where branches were not properly stacked before burning. The last task in piling branches isto - itline the edge of the circle with small branches, in order to keep the fire from spreading into the bush, and to provide an extra thickness of ashes at the edge for --owing gourds. (Taken from Miracle 1964, pp. 13-18.) Whether, or to what extent, the yields of these systems are to be attributed to the effect of fertilizer or to the yield-promoting effect of heat is an open question: It may be that this is the mechanism through which the heat of the chitemene burn exercises its main effect, but there is as yet no direct proof. Whatever the explanation, it seems clear that the ash-planting peoples, with no livestock and only the simplest equipment, have devised an effective means of conferring a high iftransient fertility on soils of very low intrinsic value. (Allan 1967, p. 74.)
36
(7) Application of mineral fertilizer. The use of mineral fertilizer in shifting systems is rare, limited to those with a higher degre of cash cropping. Mineral fertilizer has been adopted especially where tobacco and cotton are grown. The majority of shifting systems do not involve the importation of purchased nutrients. They depend on the nutrients present in the soil. This is a precarious basis, and the main reason for the continual movement of shifting cultivators. 3.3.4. Characteristics of animal husbandry The kinds and extent of animal-rearing among shifting cultivators are as varied as the forms of their plant cultivation. In the rainforests, the cultivators have few domestic animals. Mostly there are a few goats, sheep, or chickens. In south-east Asia and West Africa, pig-keeping is often practised. As a supplement we find fishing and collecting caterpillars and other minor sources of animal food. In the humid African savanna, in which there is often plenty of grass for fodder, the tsetse fly has hitherto prevented cattle-rearing, apart from in limited areas in West Africa and the Congo where tsetseresistant cattle breeds are kept. The situation is different in Latin America, particularly in Brazil, where shifting cultivation and a large herd of cattle are often combined, and the cattle eat the grass on the cleared open spaces. In the drier savannas, animal-rearing among the shifting cultivators is usually on a large scale, and is closely connected with arable cropping. Under these conditions slash and burn agriculture may convert bush-covered land into grazing. The aims and characteristics of cattle-rearing, as they are connected with slash and bum agriculture, are discussed in more detail in Chapters 4 and 9. 3.3.5. Characteristics of the labour economy Shifting cultivation is a system of production, almost without capital, based on labour and the periodic consumption of soil fertility. Practically the whole input consists of manual work. The operations of shifting cultivators can conveniently be classified as follows: (a) Clearance of wild vegetation. From the standpoint of the labour economy, clearance is the most strenuous activity. Clearing is frequently, but by no means in all cases, carried out within a system of fire-farming. Table 3.3 shows that the input into clearance varies widely (179-1450 hours per hectare) according to the vegetation and the thoroughness of the clearing. (b) Land preparation and planting. Immediately after burning, planting can often be carried out without cultivating the soil. However, most shifting cultivators do in fact disturb, loosen, and move small amounts of surface soil as part of the operation of planting (Spencer 1966, p. 31). In the savanna, or on grassland, or where cultivation lasts for several years, the soil has to be thoroughly loosened, because it is usually much harder, but on the whole the
37
shifting cultivator needs to prepare the soil comparatively little. He replaces, as it were, cultivation with the hoe by clearance work with the axe and matchet. Thorough deep-soil preparation with the hoe is less characteristic of shifting cultivators than of semi-permanent cultivators. (c) Weeding. The fallow suppresses the growth of weeds, and after fire clearance the soil is often weed free. Many shifting systems manage, therefore, with little weeding. Only as the period of cultivation increases does the effort spent on hoeing and weeding increase. Scaring away birds may, however, demand much time and may compete as an activity for children with going to school. (d) Harvest, transport of harvest, andprocessing. Together with clearing, harvesting is the most labour-demanding task. This applies particularly to root economies, in which the transportation of harvested produce-there are usually no roads or carts-and the processing of the crop tend to be time consuming, as for example with mai'ioc. 3.3 Labour input into land clearing in shifting s',stems (man-hoursper hectare)
TABLE
Virgin forest
Guatemala' Milpa
Congoc.d Rainforest
Congod Savanna
Slashing and 293-326 felling Burning and 32-82 clearing Total 325-408 'Clearing for hill rice.
Man-days for supervised work, assuming 6 hours per man-day. * Careful clearing with two burnings.
In addition, various other activities have to be considered. Tending livestock may be very time consuming where shifting farming is practised by people with a pasturalist background, and where men spend most of their time with their cattle. Getting water often takes much of the wives' time. Visiting local markets is also very time consuming. Work in the fields usually absorbs not more than half the working time. It is important in this connection to realize that the seasonality of labour demand in shifting systems is usually less pronounced than in more intensive arable systems. Clearing,
38
If
II .
e,
i: fI
iIfi
'--,
IeI
Fildboundary
~Boundary
C
of holding
-.-.
lll
Boundary of realm
0:20
40
1ODm
Fxa. 3.5. The responsibilities of different members of a household for the individual plots in a Zande holding (from de Schlippe 1956, p. 109).
(a) courtyard; (c) maize-sweet potatoes; (d) maize-oil seed-gourds; (e) groundnutsEleusine; (f) mainarea; Eleusine plot; patch; (g) Eleusine-grass; (g ) sesame-grass; on fire-preserved same-g~rass (h) bean (hf) beans-Eleusine or sorghum (ga) tield; (i)cotton (j) ianioc fallow.
39
harvesting, and processing are the major time-consuming activities, and there is more latitude for doing these chores than there is with cultivation or weeding. In the traditional forms of shifting cultivation, the execution of the labour operations is usually governed by the customs of labour division. A number of main types of labour division can be recognized: (a) Division of labour with respect to different jobs. A clear division of jobs between men and women is usually found wherever forest and bush vegetation must be cleared. In fire-farming in the forest, clearing is the man's job, while planting, weeding-if done at all-and harvesting are carried out by the women. (b) Division of labour with respect to plots. Short-term cycles of cropping and fallowing give rise to a fallow vegetation consisting of grass and bush. The input for clearing is comparatively low. Consequently, work is divided between men and women in terms of areas. Both men and women tend to cultivate their own plots. Men do most of the heavy work (clearing) and the women perform the tasks requiring more dexterity (weeding, harvesting). Fig. 3.5 is taken from de Schlippe's book (1956) about the Zande, who live on the borders of the Sudan and the Congo. It provides an example of the distribution of plots to the members of the family. (c) Division of labour with respect to specific crops. When larger plots are planted with cash crops, then the women are usually responsible for cultivating the food crops, whereas the men farm the plots with cash crops. Housework and collecting firewood are almoit always women's and children's tasks, while the men are responsible for cutting the wood used for building and for erecting the houses. TABr. 3.4 Labour input in shifting and semi-permanent holdings, Kiberege Strip, Kilombero Valley, Tanzania
Shifting cultivation
Shifinghe ltvart
Semi-permanent
farming in the
on the escarpment
Number of holdings R value Crop area (ha ME-') 6 5 0.607
valley
14 55 0.930
Rice yield (kg ha -) Labour input " b (man-days per ha) Labour input as %of the total:
Land cultivation Planting Weeding Harvesting
1683 272
31 19 20 30
1189 247
18 15 31 36
Data taken from only 4 and 8 holdings respectively. The number of hours worked per day is different for the individual jobs. Thus, clearing takes an average of 6 hours and weeding and harvesting 4 hours. Source: Baum (1967, p. 37).
40
Although the implements of shifting cultivators as a rule are limited to hand-tools unassisted by domestic animals, the productivity of labour is by no means lower than in more intensive forms oftraditional land use. Conklin's data from shifting cultivation with rice in the Philippines (1957) indicates that the labour input per quintal of rice compares favourably with the labour cost figures for rice production in irrigated paddies. Information collected by Baum (1968) in the Kilombero Valley indicates that shifting cultivators need 30 per cent less labour per unit of rice than do semi-permanent cultivators (see Table 3.4). There is, however, a change in the timing of labour demand. Shifting cultivators put most of the effort into land clearance and cultivation, while semi-permanent cu!tivators devote most of their time to weeding. 3.4. Three case studies 3.4.1. Extensive shifting cultivation with monocropping of the Iban, Sarawak The shifting cultivation of the Iban on the island of Borneo is limited to the cultivation of a single crop: upland rice. Where primary forest has been cleared, rice is grown for two years. Where there is secondary forest, the Iban restrict themselves usually to a single harvest and then the ground is rested for fifteen to twenty-five years. The R value is very low, about 5-6. The Iban combine shifting cultivation with stationary housing in solid 'long houses'. When the fields are too distant from the main homestead, smaller groups build huts away from the long house and use the land around them. When the soil fertility has regenerated near the main house, they return. Stationary housing is made possible particularly by the fact that in rice the Iban have a product that is relatively easy to transport. From the staadpoint of household and labour economies, there are disadvantages that arise because the Iban grow only one crop, and have only one harvest a year, but these are partly compensated for by the fact that the families, who live together, help each other out, although they work separately. Over nnd above this, the growing of the rice crop is phased. The main crop of slow-maturing varieties is supplemented by quick-maturing varieties, so that the household can be supplied soon after the beginning of each new growing season. The rice harvest is thus spread over three months. According to the material collected by Freeman (1955), shown in Table 3.5, a family of six people grows, on average, 2.2 hectares of rice (0.37 hectares per person), and harvested in the unfavourable test year (1959-60) only 16.1 quintals of paddy (7.3 quintals per hectare). The object of production isto supply domestic needs. However, when there are good harvests the surplus is sold. Measured in American dollars the value of the rice yield, and therefore the income from agriculture, amounts to $80.50, i.e. $0.24 per working day. Families are inclined more and more to supplement their income from arable farming by paid work in the plantations and by planting rubber.
41
3.4.2. Shifting cultivation with mixed cropping in Congo-Kinshasa As a rule, shifting systems are not organized as simply as that of the Iban, but are characterized by a variety of crops. A description of a shifting system with diversified cropping in the rainforests of Congo-Kinshasa is given by Tondeur (1956). The head of the family and the tribal chiefs cooperate in selecting the land that is to be cleared, and in allotting the plots to the
TABLE
3.5
Gross returns
($)b
80.50
330 0.24
families. The clearance work consists of(l) chopping down the undergrowth, (2) selective felling of the bigger trees, and (3) burning off. To save labour, since the trunks tend to get thicker at the bottom, the trees are felled about one metre up. Hardwood trees are killed by fire or ringing. Trees with edible fruits, or with leaves that support edible caterpillars, are left untouched. Clearing work is frequently undertaken communally by the men of the village. Burning occurs several weeks after felling, when the timber is dry. There is no need for the soil to be cultivated immediately after the forest vegetation has been burnt off, since the fire leaves it in a loosened condition. When the rains begin, sowing and planting take place. Rice and millet are sown broadcast, while maize and manioc are planted in holes, which used to be made with a digging-stick, but are usually made nowadays with a hoe. Initially the main crop is rice. When the rice has sprouted, manioc is planted in between. A few weeks later bananas are planted into the mixture of rice and manioc. When the rice has been harvested, manioc becomes the main crop. The stand of manioc and bananas gradually turns into a tumbledown fallow. The land is then allowed to rest for eight to eighteen years. The R value is thus 11-25. The harvest, transport to the hut, drying of manioc, and transport to the
42
market are the principal labour tasks. Although rice has to be plucked ear by ear, manioc and bananas involve an especially high expenditure of labour in transport and preparation. On the other hand, they have advantages from the standpoint of both the labour and household economies. The banana harvest and the supply to the household stretches over a long period of time.
TABLE 3.6
Economic return of shifting cultivation in the Congo-Kinshasa (model calculation per hectare, 1956a)
Yield Product Rice Manioc (dry) Bananas Gross returnb Purchased inputs' Expenditure on tools Income of holdingb Productivity Labour expenditure Clearing and burning, 160 work-days/ha Seeding and planting Rice harvest, including transport to the hut' (800 kg, 500 m) Preparation of produce Transport to market (400 kg, 5 km) Manioc harvest, 15 t, including transport to the hut Preparation of produce and drying Transport to market (5 t, 5 km) Banana harvest, 10 t, including transport to the hut Transport to market (5 t, 5 km) Total labour Gross return Gross return Gross return
a
Harvest
(q)
8 50 100
(S q-1)
3.15 1.26 0.42
Price
(S)
13020
1.00 129.20 Work-daysd 80 30 45 5 7 50 120 42 16 83 478 027 130.20 13.02
expenditure ($ per work-day) (S per ha cultivated) ($ per ha of the total area of land)
2 years of cultivation, 18 fallow years, 0.5 ha annual clearance, I ha annual .ultivation. Area requirement the holding: 10 ha. b Not including the of yield from hunting, collecting, and fishing. Domestic consumption: 4 q rice, 50 q bananas is included. c Not including farm-produced seeds and planting material. d Tondeur does not supply the number of work-hours per day. Pinxten (1954) estimates in these conditions 4 hours' work a day. The labour expenditure of 160 days per ha is very high indeed. Samuel (1950, p. 589) calculated 90 days per ha for similar conditions. The labour expenditure on harvest given by Tondeur corresponds with that of other data. Transport for carrying. Data from: Tondeur (1956, p. 43).
43 SHIFTING CULTIVATION SYSTEMS when time is Manioc can stay in the soil and can be dug out when required or used stays not is available. Manioc is often grown in excess of need, and what another when in the soil as a safeguard against hunger, being harvested only crop unexpectedly fails. of such The calculations in Table 3.6 give an idea of the labour input and yield cropping. of years two follow a shifting-cultivation holding. Eighteen fallow years hectares and Correspondingly the area requirement of a holding amounts to 10 hectare as 8 I of yield the gives Tondeur hectare. 1 to cultivation the annual of fresh quintals of rice (paddy), 50 quintals of dried manioc, and 100 quintals input The $130.20. of value money a to corresponds production bananas. This which tools, on expenditure the to is it of material resources is small, limited as the holding. accounts for $1 annually. Young plants and seeds are taken from work-days 478 against set be Thus, a holding income of $129.20 remains, to processing.' expended. 75 per cent of these are devoted to harvest, transport, and work-hours. 4-5 with work-day per $0.27 to amounts work-hour per yield The of the The yield per hectare, taking into account the total area requirement considerable in often produce, other this, to addition In $13.02. to holding, comes hut, and from quantities, is obtained from the garden that is developed near the hunting. and collecting, fishing, 3.4.3. Shifting cultivation with rotation among the Bemba, Rhodesia those of the The shifting systems of the dry savannas are different from which of forest because of the prevalence of long-term cropping periods, cultivation in necessity lead to intensive hoe-cultivation practices. Whereas it lasts for up the forest seldom lasts longer than three years, in the savanna by a oneto ten years, and sometimes even longer, occasionally interrupted decades. two to one for rested land the is year or two-year fallow. Only then described Typical of this type of farming is the Bemba system, which has been fields, cleared the are hut the from distance Some 3.3). Fig. by Richards (1961) (see from year to year. or in other words small, irregular plots, the size of which varies a monocrop, partly Immediately after clearing, finger millet is sown, partly as The rotation of the interplanted with manioc, and partly mixed with sorghum. farming, but we can permanent in usual is as individual fields is not as systematic from field to different are rotations the that examples Richards's well see from household the by demanded is that grown field, and that in each case the crop is holding. the to well-suited and is relatively and in Near the hut is the garden, which is fertilized with ash and refuse, manioc and potatoes, sweet maize, pumpkins, which are found plants-legumes, as different crops -that require a better soil than those grown in the fields. Just sections, each several garden the in also are are grown in different fields, there crops. different supporting of transport for According to Jurion and Henry (1967, p. 320), the time requirements and bananas, groundnuts, rice, maize, of a family of four, planting annually 0.5 hectares distance between amount to 120-40 man-days annually, under the condition of an average 4-5 journeys per hut and plot of 1.5 kilometres, an average load of 35-45 kilograms, and man-day.
I
44
An earlier hut site forms a separate plot, which becomes available when a new hut is built, and it is characterized by a comparatively high fertility. Maize, pumpkins, tobacco, and sweet potatoes are grown here.
45
savannas, there is a lack of In the forest regions and in the tsetse-infested fertilizer is usually not worth livestock to provide animal manure. Mineral ratios, which can hard!," be while, because of unfavourable cost-return is limited to the hne; rioughs avoided in shifting systems. Shifting cultivation are cleared, and that would stumps cannot be used unless roots and tree vegetation. In shifting cultivation fallow desired prevent the growth of theunknown; tihire are only foot-paths. The frequent proper tracks are almost large network of tracks. changes of field would necessitate an exceptionally the introduction of carts. The The absence of tracks, however, prevents or transported by bicycle, and those harvest must for the most part be carried carry it to the field in baskets. people who have animal manure have to of hut sites, and much time is lost The shifting of fields leads to the change and huts are moved at intervals in building huts. Those people whose fields permanent improvement of the land, have little inclination to invest work in is therefore distinguishable by the and a countryside with shifting cultivation as irrigation, drainage, and tree lack of permanent improvements, such are scattered. The lack of stationary plantations. In general, the family units fields create difficulties in storing housing and the long distances between division of labour between farming food, collecting surplus produce, and cultivation is thus a hindrance to the and non-agricultural pursuits. Shifting political bodies, and advanced development of villages and towns, central the economic atti.ade of the farmers societies. The land-use system moulds in the habit of regarding the soil as who practise it. Shifting cultivators are has to be maintained. Where soil a free gift of nature, not as capital that of being eroded, shifting cultivafertility decreases and land is in the process land and to look for virgin territory tors are much more inclined to mine the than to invest labour in soil conservation. shifting systems 3.5.2. The tendency towards 'degraded' production for sale, a number of With increasing population and growing distinguished: distinct stages of development can be shifting of cultivation and of the (1) The oldest form is probably the a primary vegetation. farming group in one direction within of other claimants to land, the (2) According to the growing number and then each family shifts groups share the total area among themselves, by secondary vegetation, in a circle in the area allotted, which is covered intensive cultivation Relatively and moves on after one or several decades. (see Fig. 3.2). 30-50) = (R is temporarily practised near the dwelling shifting with circular of instead (3) With a further increase in population, there occurs often the shifting long-term intervals of fallow and cropping, with the tendency to stationary of fields within a fairly limited area,
46
housing, where possible along a road. On the permanent site the hut position is sometimes changed. (4) A continuous increase in population and the development of cash cropping involve an intensification of shifting cultivation by shortening the fallow (R rises to 30-60). At this stage, the basis of land utilization has changed to such an extent that the system is beet regarded as semipermanent farming. Such systems are considered in the next chapter. These processes usually lead to a situation of relative overpopulation on land which thcn loses soil fertility. Without changes in technology and inputs, only a small number of people can be supported compared with othe, landuse systems. Naturally the crop yields of shifting systems vary greatly according to climate and soil type. For example, whereas on the typical light soils of the rainforest in the Congo twelve fallow years have to follow three years of cultivation, with the same rainfall on volcanic soils there are only two-year or three-year fallows for the same period of cultivation. In any case, however, the shortening of the fallow or the lengthening of the cultivation period beyond a certain point disturbs the equilibrium of the land-use system. Shifting cultivation, which can be called 'balanced explotation' where the fallow period is sufficiently long, becomes sheer exploitation of the soil fertility. Geertz (1963, p. 26) argues that in the rainforests of south-east Asia, in an ecologically balanced shifting economy, no more than twenty to fifty people can live per square kilometre with guaranteed subsistence. Figure 3.6 shows these developments in the form of a graph. yield of the soil drops with the number of cultivation years and recovers in The the In (a) the fallow period lasts longer than the regeneration of fallow period. the soil requires. R is low (11). This is shifting cultivation with production reserves. The situation in (b) corresponds to that of shifting cultivation without production reserves (R = 29). The fallow period is,however, long enough to restore soil fertility to its original level. (c) shows what happens as the fallow is shortened (R = 46). The fallow is no longer sufficient to restore soil productivity and the yields per hectare fall, though since the shortening of the fallow period means that a greater part of the total area is cultivated, the fall in yields per hectare accompanied by a rising total production. However, the result may well be is a continuous degrading process. The increase in cultivation involves a more thorough clearing, more intense hoeing, and, at the same time, the death cf roots and tree stumps. The fallow ground does not produce bush quickly enough, especially on soils with low fertility, after the years of cultivation, so that only arable weeds, bamboo, or wild bananas establish themselves. Bush regeneration grass, is made difficult, and the clearing of bamboo and many types of grass (above all Imperata cylindrica) involves a high labour cost. Where cattle are ,ept, grazing is an additional impediment to forest or bush regeneration. Spencer
cultivator groups are conscious of p. 127) reports that most shifting as shifting systems are intensified. seat of the spread of grassland with shifting cultivation, the burning theless, almost everywhere in areas in the dry season, partly from habit, .t bush and forest areas takes place to get early fresh grass for the i to make hunting easier, and partly
47
I
.
I ti it
Cultivation Fallow years years ,5 cultivation shifting soil productivity in a(from fallow and1956). Fio. 3.6. The relation between length of Guillemin
10
15
20
25
30
35
40O
rainy season. Regular grass fires also cattle before the beginning of the next bush. help to prevent the growth of forest andfire transforms the countryside of and grass between The interaction Latin America and into a of Africa areas vegetation or bush by man. forest created natural a large vegetation fromIn grass systems shifting is cattle unbalanced the for grazing this process is deliberately encouraged because as do not have the same ability desired. Grasses in the tropics, however, to accumulat organic matter. bushes or trees to remobilize nutrients and of rootig, the capacity to depth Trees and shrubs alone have the necessary
48
make good use of a large part of the rainfall throughout the whole year, and the ability to retain, in their own structure or in their leaf litter, a substantial portion of the nutrient material available. This involution, therefore, leads to a countryside which is in grave danger because of loss of fertility and the effects of erosion.'
3.6.
Shifting cultivation is obviously not essential in the tropics, but is rather an expression of a certain stage in population density, technology, and price relations. However, in circumstances of relatively mild population pressure, shifting cultivation has been and still is widely suited, ecologically and socioeconomically, to the simple needs of man and the potentialities of the environment. Its replacement by some more satisfactory land-use system is therefore by no means easy in the light of our present knowledge. The spread of more productive systems of land use is, as a rule, the result of necessity and incentives. When shifting systems are modified, the incentives come from cash cropping, and the main pressure tends to be the relativ shortage of land. Increasing cash production, as well as the increasing need for subsistence crops, and both together in interaction with the introduction of technical advances, lead to efforts to regulate and improve shifting systems or to replace them by more efficient patterns of land use. 3.6.1. Improved shifting systems (a) Shifting systems with plantedfallow vegetation. Experiments to replace wild vegetation in the form of forest or bush, by planting quick-growing tree or shrub legumes, have as yet not been able to make much headway. De Kroon (1967) reports, however, that planting them in Surinam has proved a technical success. The fallow was shortened by several years. The soil under cover was in excellent condition, easily tilled, and the yield of the succeeding crops higher, but this process has been adopted no more widely in Surinam than in West and East Africa, where similar efforts were made. The main reason is that it is very expensive, i.e. it costs a great deal in terms of labour, in relation to the extra yield, to produce a planted but unused fallow vegetation. (b) Shifting systems inforestry reserves. The people responsible for forestry in several areas (India, Burma, Indonesia, East Pakistan, East Africa) successfully combine shifting cultivation with planned reforestation. They allow, direct, and control the cultivation of food crops in the forest reserves by a limited number of shifting cultivators. The cultivators are allowed to cut IInvolution is the attainment of higher total income in the area accompanied by a lower income per head, because of population increase. Evolution is the attainment of higher income per head.
49
down, burn, and farm a certain area of unimproved forest allotted to them, under the condition that the land is vacated after two or three years of cropping and that the forest trees (mahogany, teak, pines, etc.) which are planted in the second year are properly weeded. Then they receive a new plot. This procedure is advantageous to both sides: the farmers have the chance of farming within a restricted area, while there is no danger of forest destruction; the forestry department saves the clearing costs of reforestation, because these are borne by the shifting cultivator. The only, but considerable, disadvantage in this process is that relative!y few shifting cultivators can be absorbed. The felling cycle of the timber is much longer than the usual fallow of fifteen years, reaching eighty years with teak and mahogany and about thirty years with pines. This method is thus well adapted to forest areas with little population pressure, but it is unsuitable for more densely populated areas that are inhabited by shifting cultivators. (c) Shifting systems in plantations. In Sumatra an ingenious system has been devised which permits the combination of large-scale commercial tobacco production with subsistence production within a shifting system. It is described by Geertz (1963, p. 107) as follows: ... the famous Deli wrapper tobacco is grown only on land which has been lying under forest fallow for at least seven or eight years ... The tobacco planter of East Sumatra is thus a shifting cultivator who raises tobacco only once in a cycle of eight years, and each tobacco plantation therefore requires about eight times as much land as is planted in any one year ... When the tobacco companies originally applied to the sultans for agricultural concessions, no attempt was made to set aside sufficient land for the peasantry, although the contracts stipulated that the companies had to give each peasant family 4 bouw (2.9 hectares) of land, so that it could carry on its own traditional form of agriculture. Eut instead of reducing the size of their concessions by turning over to each family within the concession the stipulated amount of land, the planters preferred to have the use of all the land and to lend the harvested tobacco land to the peasants raising a single rice crop. This meant that the planter had access to all but the actual village land within his concession, while the East Coast peasant no longer had to clear land for his own swidden. Instead, he received from the estate each year in May or June the same amount of land ready for the sowing of upland rice that he would have had to clear had he not lived within a tobacco concession. Advantageous as this system may be to both sides concerned, it still suffers from the same disadvantage as do most shifting cultivation systems, namely that further intensification is scarcely possible. (d) The 'couloir' system. A large-scale experiment to regulate and improve shifting systems took place under the Belgian colonial administration in the Congo. In 1960, about two million hectares belonging to 200 000 shifting cultivators were cultivated under the couloir system. The system consisted of a sufficiently large and fertile area being selected, declared a paysannat in agreement with the population, and divided into straight strips. The
5
50
individual families each received a strip of forest of 8-12 hectares, of which 1-2 hectares were cleared and cultivated whilst the remainder was fallow.1 A rotation of cultivation years and fallow years was established on the plot. The plots cleared by each family were next to each other, so that normally a cleared corridor ran through the forest. As clearance continued, this corridor shifted in one direction and left behind incipient secondary forest (Fig. 3.7).
The state of the various strips in 1958 Clearing schedule 1953-71
/19"01 1960 1969,
Clearingin November -December
19I',l67:"
Mamo:mize,JulybananosriceSept monoc Harvest of monioc and bananas May-maize.Sept.A hroms (fibre), groundnuits ,
1958
, , J v, ,,_
1
,,,,1 1968%
f
i
.. 1956
1 19651 ;. 1964.; . _
,
,954
//
9ez~95
1963
I
1953
and
192tL't~-~.j.J
1972
FIG. 3.7. Controlled shifting cultivation in the couloir system (from Dumont 1957, p. 40).
Experience has shown that the best system consists in the opening of east-west strips in the forest, alternating with strips of untouched forest. The strips left unfelled provide shade and seeds Pir the regeneration of the felled areas once they have been left to revert to fallow. The optimum width of such a couloir has been established at 100 metres. 2 Cultivation in the corridor depends on the cultivators' willingness to follow the obligatory rotation, i.e. to grow particular crops according to whether it is the first, second, or third cultivation year. The whole corridor is thus divided up into strips of the same crops. At Yangambi the rotation was practised that is illustrated in Fig. 3.7. The strip
for cropping in a particular year-say 1958-is felled and prepared for cultivation. The following year (1959) the next strip is felled and prepared, and the year after that (1960) the next strip, and so on. When the strip for 1961 has been opened up,
I An example: 1600 hectares are available for 200 families. The rotation consists of 3 cultivation years and 16 fallow years. The annual clearance amounts therefore to 0'4 hectares, the area of cultivation to 1.2 hectares (Tondeur 0J56, p. 77). 1 For the numerous variations in the organization of the paysannats, se Pinxten (1954).
51
the one to be exploited next is at the other end of the block, next to the strip that was cleared for 1953. The clearing of alternate strips then proceeds through the block for ten years, each strip being cropp.d for only one year. When the twentieth year after the start of the work is reached, felling takes place again on the strip first cleared in 1953, where a nineteen-year-old regenerated forest exists. All subsequent strips have also had a nineteen years' fallow before they are cleared again for cropping.
The regulation of land use by the paysannat system has a number of advantages: (1) The ratio of cultivated land to fallow land, and thus the preservation of soil fertility, is under control. (2) The concentrated cultivation of particular crops in one belt, where each family has a section, facilitates the introduction of improvements like rotations, cultivation of cash crops such as cotton, better varieties, row cultivation, planting at optimum times, denser planting, weed control, burning off old cotton stalks, plant protection, etc. (3) The concentration of cultivation reduces the expenditure of labour on transport. Building tracks becomes an attractive proposition. Marketing and processing can be organized cooperatively. Difficulties resulting from varying soil fertility and the varying work capacity of families can be largely overcome in more flexible paysannats. As Fig. 3.8 shows, it is possible under the couloir system to give the participants plots with different soils and of different sizes. Each number designates the plots of one family. No matter how beneficial some features of the couloir system may have been, it has not survived in the Congo after independence. The shifting cultivator who is used to freedom of movement obviously finds the compulsory rotation of land oppressive. Maintenance of the rotation probably required the administration of the paysannat by an official from an outside institution, with sufficient 'power keys' (like project rules, control of credit and sales) to be able to make the farmers adhere to a system that they would not otherwise have preferred. The planning and supervision of a paysannat thus makes considerable staffing demands. Undoubtedly the couloir system brings considerable increase in productivity per hectare and per man-unit, but it remains in principle a land-use system that demands a great deal of land, the labour productivity remains at a low level, and it is a system that the participants often feel to be too rigid. In short, it is an attempt at improving and organizing a system which, in the context of agricultural development, will sooner or later disappear anyhow. Newton (1960) reports that at the end of the 1950s the interest in the couloir system in the Congo declined, and that experiments with field-grass systems were increasingly being made.
SHIFTING CULTIVATION SYSTEMS 3.6.2. The changefrom shifting systems to more intensive types offarming (a) From shifting to permanent rain-fed farming. The obvious development of shifting systems lies in a progressive shortening of the fallow, which leads finally to permanent agriculture after a period of semi-permanent agriculture
52
I cous
pf:
ing
Good soil
'- Less
good Soil
2nd
,-
Good soil
lit Vear
4th
gear
ED
3rd
year
5th gear Under 71fallow FiG. 3.8. Couloir system with varying size of individual plots (from Brouwer 1965). or unregulated ley systems. In tropical high-altitude areas, on especially fertile soils and in the dry savannas, the development toward permanent farming has taken place to a considerable degree in a relatively short period. In the rainforest climate, on the other hand, the introduction of systems with
gear
53
permanent arable farming proves to be a difficult problem from the standpoint of agronomy and farm economy. The first experiments with this kind of system, which were carried out in Yangambi, were an unequivocal failure. Dumont (1966, p. 37) describes them: On the first experimental plots at Yangambi, the forest was cut down, burnt, and uprooted. The soil was ploughed deep and, as a precaution, sown with Pueraria, Calopogonium, and other leguminous crops which had proved their worth in plantations. Thes: were used as green manure and various grasses (Pennisetum purpureum) were sometimes sown as well. The land was ploughed a second time and then cultivated on a two-year rotation. As the region has rain throughout the year, with only a slight abatement from December to April, it was possible to introduce four breaks into this rotation. A crop of upland rice was followed by two of manioc and one of groundnuts. This was followed by a leguminous cover crop. In this manner it was hoped to begin continuous cultivation on the pattern of temperate agriculture. The ant-hills, which occupied about a third of the area, were not cultivated. The soil was exposed to the harmful effects of chemical photo-oxidation for more than half the time, for thirteen months out of twentyfour, to be exact. Deep ploughing, expensive as it was, impaired the soil structure. The absence of a vegetation cover for so much of the period increased the amount of percolition, and the removal of plant foods from the topsoil was thus accelerated. The cover crops themselves did not have any apparent effect in improving the soil and their woody stems decomposed too slowly. Bacterial activity was hindered by excessive insolation and was unable to keep up an adequate supply of humus. The yield of every crop fell rapidly. The ears of rice did not swell. The 150 acres on which the experiment had been conducted had to be abandoned after a few years. Too far from the edge of the forest to be reseeded quickly and with the micro-climate of its soil reduced almost to aridity, the area was still only thinly wooded ten years later. The Belgians frankly recognized their mistakes. Their notions of farming in temperate regions were obviously out of place in the Congo. In 1940 they decided to start again, but this time they took the Bantu system as their only sure starting-point and tried to improve it. Recent experience shows that permanent cultivation on the light soils that predominate in the humid tropics is possible with heavy applications of mineral fertilizer and the maintenance of sufficient organic matter in the soil. These conditions are scarcely practicable tinder the input-output conditions of most tropical countries, even where agronomically satisfactory solutions have been worked out. They prove mostly to be too complicated or too unrewarding to be adopted on a large scale by indigenous farmers (see 6.6). (b) From shifting to regulated ley farming. An alternative possibility of evolution is the replacement of the bush fallow that occupies the land for several years by plants that are themselves usable, regenerate the soil productivity, and allow use of the plough. The obvious way to achieve this is by transition to the Icy system. However, arable crops, grasses, and leguminous crops in a rainy climate do not have the capacity of the forest or of tree crops to keep a reasonable balance between leaching and the remobilization of
54
nutrients. The experience in Yangambi at the end of the 1950s shows that in a rainy climate ley systems can take the place of forest and bush fallow only with the help of heavy applications of mineral fertilizer. However, the right economic conditions for ley systems demanding much fertilizer are found only rarely (see 4.6.2). (c) From shifting cultivation to perennial crops. The humid tropics are by nature forested, and the natural environment is well suited to perennial crops like bananas and to tree crops, a fact that is widely recognized and used by the shifting cultivators. The slowing down of the shifting cycle leads to stationary housing with more or less permanently cultivated gardens. Permanent banana groves develop, offering soil protection as well as high yields of calories per hectare. Wherever remunerative markets are available, shifting cultivators tend to plant tree crops like rubber, coconuts, and cocoa in their plots, since these are less choked by weeds and demand less of the soil than annual crops. The cultivation years are thus not followed by regeneration of fallow vegetation, but by a stand of tree crops whose effects on the soil resemble those of forest and bush vegetation. Evolution of this kind has taken place on a large scale. Nowadays in tropical Africa and Asia thousands of hectares that were previously utilized by shifting cultivators support cocoa, oil-palms, and coffee. The evolution from shifting systems to perennial crops is obviously suitable for the natural conditions of the wet tropics in particular: it is also a change that is relatively simple and cheap. (d) From shifting cultivation to irrigationfarming, in particular to paddy rice. The experience of tropical Asia shows that the best system of arable cropping in a warm, damp climate is the concentration of production in the relatively fertile, irrigated valley bottoms, i.e. replacement of shifting cultivation over large areas by irrigation farming, especially with rice, in small, intensively cultivated plots. The transition from the shifting system to the paddy-rice system, or the combination of both methods, can be observed in large areas of the tropics (Madagascar, East Africa, Ceylon, Indonesia). It depends, as a rule, on !he initiative of the indigenous population, and obviously represents an evolution that is ideally adapted to the special environmental conditions (see 6.4.2). (e) From shifting cultivation to ranching. In some areas in Africa and to a greater degree in Latin America, especially Brazil, shifting cultivation serves as a method of obtaining cleared grazing land. The shifting cultivator obtains the right to clear the land and to cultivate it for several years. Thereafter a new plot is allotted to him. The period of crop farming, however, is not followed by the regeneration of a fallow vegetation of bush, but by the transition into grassland, facilitated by heavy grazing, supplementary bush clearing, and planned burning. Similar methods of using shifting cultivation in order to turn bush into grazing land are widely practised by cattle-owning cultivators in the African savannas.
cash production and the growing subsistence needs of an increasing population lead to a gradual extensi.., of arable farming at the expense of the fallow, and short-fallow systems replace long-fallow systems. Cultivateurs avec jachres (Faucher 1949, p. 9) cultivate areas with alternating cropping and bush or grass fallows. Brinkmann (1924, p. 959) used the term Umlagewirtschaft to indicate that stationary farms change the cultivated area more often and more regularly than shifting cultivators. In this type of land use the value of R is between 30 and 70. An R value of 50 can be regarded as typical, and arises when, for example three, five, or ten fallow years follow three, five, or ten years of cropping. In such a situation 50 per cent of the arable land is cultivated annually.' The length of fallow is mostly insufficient for fallow vegetation of forest or dense bush to regenerate. Grass or light bush is typical. In areas not infested by the tsetse fly, the fallow vegetation is grazed by cattle, sheep, and goats. The characteristics of these farming systems are distinct from those of shifting cultivation. We classify them as semi-permanent systems. Where the fallow vegetation is systematically grazed we speak of unregulated ley systems. Table 4.1 illustrates this development amongst the Bailundu of Angola. Within a few decades, the reduction of fallows has increased the R value from 17 in the traditional system to 31 in hoe systems and 54 in plough systems. Semi-permanent farming is usually characterized by clcarly defined holdings with largely permanent field divisions. Quasi-stationary housing predominates, since the changing of hut sites is a matter of moving short distances only. Families generally have defacto or registered ownership of the land, in contrast with most shifting systems, in which the holding boundaries are not usually clearly defined, housing is more or less of a migratory nature, and land rights are even less precisely defined.
EXPANDING
'A further criterion isthat several fallow years succeed several cultivation years. In scmidry climates of the subtropics the continuous rotation grain-fallow is widespread. lron the point of vi :m'of farm management, these dry-farming systems are fundamentally different from semi.pcrmanent farming. 55
56
Most semi-permanent farmers practise hoe cultivation in the form of a soil-conserving and yield-maintaining, but labour-demanding, ridge or mound culture. They usually cultivate a larger area than shifting cultivators in the same environment, with a fairly large cash-crop area, which may take half of the total land cropped. In Africa in particular, the economic importance of semi-permanent farming is far greater than that of shifting systems. Most of the population and areas that would normally be associated with shifting cultivation are in fact part of semi-permanent cultivation systems. 4.1 Development of the land-use system of the Bailundu, Angola
TABLE Traditional farming system, which has almost disappeared
R= 17
These systems are chiefly to be found where the soils and climate allow a relatively short crop-fallow cycle, i.e. a high R value, withcut rapidly decreasing yields. In general terms, a reduction of the fallow period causes a reduction of the yields per hectare, unless there is fertilizer application or manuring, which is normally not the case in these systems. The warmer and the rainier the climate, and the poorer the soils, the more necessary is the long-term forest or bush fallow, and the less likelihood there is of semi-permanent farming being successfully developed. It is therefore not a typical system in the humid forest zone. The main regions where it is found ate the dry and humid savanna climates and at high altitudes, where there is less leaching and the soils tend to be richer. Under these conditions, extension of cropping at the expense of the fallow may lead to a rising overall production from the total area available, in spite of decreasing yields per hectare cropped. The lower yield per hectare cultivated is usually more than balanced by cropping a larger area. Semi-permanent cultivation is found also on high-fertility soils, especially on alluvial and colluvial soils in river valleys, and at the more fertile bases of slopes, i.e. where the motive for fallowing is less the regeneration of soil fertility than the suppression of weed growth.
TABLE 4.2 Intensity offallow systems andfarming patterns in the Kilombero Valley, Tanzania System of land use R value Permanency of the dwelling-place Housing material Main input Typical agricultural innovations Extensive shifting cultivation 15 Frequent shifting of dwelling-place and garden Tree poles High input for clearing; low input for weeding New annual crops and new crop varieties Intensive shifting cultivation 30 Infrequent shifting of dwelling-place and garden; sometimes stationary housing Tree poles and clay High input for clearing; low inpu: for weeding New annual crops and varieties; some new permanent crops Semi-permanent farming 50 Stationary housing and garden Permanent farming 70 Stationary housing and garden
Clay/tin roof Stone/tin roof High input for weeding; low input for clearing; increasing application of agricultural innovations; increasing investments Introduction of insecticides; introduction of mineral fertilizer; increasing planting of perennial crops
58
59 SYSTEMS SEMI-PERMANENT CULTIVATION forest kind can be seen particularly in the of semi-permanent land use of this in Chapter 8. areas of West Africa. They are discussed in connection with irrigation ftrming 4.2.3. Semi-permanent cultivation irrigation farming, especially in Similar conditions occur in permanent requires permanent homes and wet-rice systems. Paddy-rice cultivation land in the vicinity of irrigated fields. creates a relative shortage of rain-fed to paddy-rice cultivation in irrigated Where shifting cultivators change in usually replace long-fallow systems valley bottoms, short-fallow systems clear examples of developments of the surrounding hill country. Especially and Indonesia 'ee Chapter 7). this kind can be seen in Madagascar drier savannas 4.2.4. Unregulated ley systems in the 4ks systems occur in the dry savannas. More important semi-permanent gives of fallows, forest or bush vegetation cultivation spreads at the expense often is which fallow, large areas of grass way to a cultivation steppe with fly this process is causing the tsetse Africa In dotted with patches of bush. harvest and fallows be kept, and the grass to be displaced, so that cattle can kraels concentration of cattle dung in the residues can be used as fodder. The organ"ed an provides the beginnings of where the animals spend the night a stage in this nature may be considered as folding system. Fallow systems of ley and are therefore called unregulated the transition to ley farming proper farming. kind of land-use system is considerable, The economic importance of this and holdings of the savannas in western especially in Africa. The cotton be holdings in western Africa, may eastern Africa, and the groundnut-millet farming. form of unregulated ley classified wholly or partly as a high-altitude areas 4.2.5. Unregulated ley systems in of the fallow is the outstanding feature Greatly reduced dependence on and Asia, Africa, high-altitude areas of agricultural systems in temperate, values R with 143). Short-fallow systems, Latin America (Miracle 1967, p. shifting systems, are consequently that are much higher than in long-fallow a rule The fallow vegetation consists as widespread in tropical high altitudes. in other words we find wild, unregulated of grass which is used for grazing; in, This kind of farming is widespread ley systems, as in the drier savannas. Ethiopia. Burundi, Tanzania, Kenya, and for example, the highlands of mountainous the in exist ley systems Comparatively large areas of unregulated and Bolivia, where the grass Venezuela, parts of Peru, Equador, Colombia, by and alpacas, and, to a lesser extent, fallow is grazed by sheep, llamas, unregulated Some 1955; Sick 1969, p. 164). cattle (Dion 1950; Fals-Borda India in mountainous areas of northern ley farming is still to be found
60
(Himachal Pradesh). In these areas, double cropping with rice and wheat is carried out for some years, and followed by several years of grazing. In most cases we still do not find the permanent field divisions that characterize regular ley systems. Instead, suitable parts of the grazing area are cropped for a few years and then left to go back to grass.
4.3.
Semi-permanent systems are of particular importance in Africa south of the Sahara. The description of general characteristics, therefore, refers principally to African conditions, especially those prevailing in the African savannas. 4.3. 1. Spatial organization of cropping Semi-permanent systems, like the shifting systems from which they have usually developed, are characterized less by attempts at controlling nature than by complex and skilful adaptation to the natural conditions of production. A typical feature is the juxtaposition of plots used for different crops, especially where the soils change over a short distance.
Type of soil
Rock zone Luguru
Prevailing use
Grazing in the wet season Maize, sorghum, groundnuts.
:."ib am bas i. ..
~ ~44~i'.hardpan
*..
soils
soils
__
._,,_,._
FIG. 4.1. The catena and the principal uses of the different types of soil in Sukumaland, Tanzania (from Rotcnhan 1968, p. 58). Figure 4.1 shows the catena of various soils in Sukumaland, Tanzania. The most suitable crops are grown on each type of soil. A Sukuma holding consists, where possible, of a strip of land running through all the soil levels. At the top of the slope are the fields with maize and sorghum, followed by plots chiefly planted with cotton and manioc. The hardpan soils serve principally for grazing. Finally, in the valley bottom, we find plots with rice and sweet potatoes. Figure 4.2 illustrates, with an example from Senegal, the horizontal differentiation that has developed in western and eastern Africa with the introduction of cash crops into arable farming (Pelissier 1966; Dumont 1966, p. 66; Froelich, Alexandre,
61 SEMI-PERMANENT CULTIVATION SYSTEMS circles from and Cornevin 1963). The intensity of cropping decreases in concentric periscattered widely often the to the domestic gardens in the vicinity of the huts pheral fields. subsistence Holdings of this type usually include a great number of various area, so holding the within and cash activities carried out on different plots
...... .. ... . .i,, . .. u.. ,. . . .
;"'.
...
--
" h
4':
IT
Senegal (from Pelissier 1966, FIG. 4.2. Spatial organization of land use in N'Gayine, (3) semi-perman'nt cultivation; cultivation; (2)permanent gardens; p. 474). (1) Houses and shifting cultivation. extensive and bush (5) (4) intensive shifting cultivation; as one of the most much so that semi-permanent farming may be considered of these holdings diversified farming systems in the tropics. The main features are as follows: like bananas (1) Permanent gardens with fruit trees and perennial crops is either which and papayas are found in the immediate vicinity of the hut, intervals. no longer shifted or is moved only a short distance at long is mostly culti(2) Near the huts or villages, what is called the dung-land soils, and should vated. It is used for crops that require more fertile corresponds to guarantee a good yield. The intensity of land use frequently garden culture. in concentric (3) Adjoining this we find semi-permanently used fields the staple growing for used circles of varying size. These fields are often pasture. as food and cash crops. The fallow is mostly used to the distance (4) The intensity of the crop cycle decreases proportionately
62
from the farmstead. Between the cropping years the ground is left fallow for long periods; in other words, shifting cultivation is practised. (5) In some areas the holdings include plots in valley bottoms, which are used for cropping sweet potatoes, taro, or sugar-cane. On the valley bottoms rice cultivation is also widely practised. It should be noted in addition that on moist plots millet or maize is cropped in the dry season, thus providing the household with an off-season supply of fresh food. (6) Trees are often found scattered over the area: fruit trees (e.g. mangoes in eastern Africa), fodder trees (e.g. Faidherbia albida in Senegal), or sisal hedges. (7) Where cattle are kept, a calf enclosure is occasionally found near the hut or inside the kraal. (8) The fallow and communal pasture are grazed by cattle, sheep, and goats. (9) Livestock-rearing in a stationary homestead is supplemented here and there by a certain amount of nomadic herding: some of the herds belonging to the cultivators graze on remote grazing areas watched over by herdsmen. By no means all holdings that are classified as belonging to semi-permanent systems incorporate all the activities cited above. In most cases, however, there is a catena in land use, or a spatial differentiation within the holding; and there is a combination of intensive agriculture and shifting cultivation, providing as it does for adjustment to differing soil conditions and labour requirements in holdings of partly commercialized farmers with stationary housing. 4.3.2. Croppingprinciples The main source of semi-permanent cultivators' sustenance is arable farming, with animals playing a supplementary role only. In the savannas, crop yields vary considerably from year to year, more so than those obtained by shifting cultivators in the humid forest zones. Consequently, semipermanent cultivators try above all to secure a reliable supply of food. Another aim is to obtain tasty food: millets and maize are usually preferred to manioc or sweet potatoes. In addition, there is as a rule a dominating cash crop like cotton, groundnuts, or tobacco in each farming system. These are the usual objectives on which the organization of cropping is based, and they require some basically different cropping principles from those that are typical of shifting cultivation. (a) Pure stands and mixed cropping. Mixed cropping usually still predominates in semi-permanent systems. The cultivation of cash crops, however, has led to a considerable number of pure stands. Cotton, groundnuts, rice, and tobacco are usually cultivated only in pure stands. (b) Phasedplanting. A widely practised principle is phased planting of
63
each crop, even though losses due to late planting are considerable. The point is illustrated by the practice followed in Sukumaland, Tanzania (see Fig. 4.3). The cultivators point to the unreliability of the rain, particularly in the drier regions. To allow for this uncertainty, they distribute the planting of their clops over a long period, preferring to spread the risk instead of aiming at maximum yields. They also prefer to spread the demand for labour over several months by phased planting. Household requirements offer an additional explanation, planting being so timed that the farmers can get food from the garden or field for continuous periods of several months. Thus the planting of subsistence crops in particular, such as beans, sweet potatoes, groundnuts, and maize, is spread over several months. (c) Crop rotation. As short-fallow systems replace long-fallow systems, we can see the pseudo-rotation (de Schlippe 1956) being increasingly displaced by proper crop rotations. When compared with the crop mixtures and pseudo-rotations of shifting cultivators, the rotations show several distinct tendencies: (1) Whereas the rotations of shifting cultivators tend to include a wide range of different crops, as cuitivation becomes more permanent cropping is restricted to fewer crops on each type of soil. (2) There is a tendency to use different rotations for the particular soils of different plots. (3) Fixed and regular sequences of crops like rotations of the kind found in permanent cropping have by no means yet evolved, but crops demanding fertile soil are increasingly being planted at the beginning of the period of cultivation, and crops demanding less fertility are planted at the end. (d) Fallow systems. The organization of fallowing in semi-permanent cultivation is as a rule especially varied. According to Rotenhan (1968, p. 64), the variety of forms that it takes can be classified in the following way: (1) There is a half-year'sfallow if the land is not used during a six-month period between the harvest and the following planting. In most wet-dry climates, the six-month fallow is the rule on all arable land. (2) There is a properfallow, if no crops at all are grown on the land during the whole year. The ground is left to be covered naturally by grass or bush. Subject to the influence of climate and soil in most African savanna areas, a short-grasssteppe develops. With increasing length of the fallow, this short-grass steppe changes into bush vegetation. A distinction can be made between different forms of proper fallow: (1) Permanentfallow exists on those arable plots that are not likely to be used for cultivation in the foreseeable future. They usually serve as grazing areas.
64
o/.
100 80 60
10 80 60
40
20 0. 0 N DJ F M % 100 80 60 40Rice ON D J FM Groundnuts ON D J F M Manioc
40
20 0 % 100 80 60 10
20
20
0.
% 100 80 60
N DJ F M
0ON D J F M
ON D J F M Cow-beans
0
% 100 80 60
Bambara nuts
Sweet potatoes
40, 20
0 ND JFM 0
40 N L20
O ND J FM 0
O ND J FM
FxG. 4.3. Phased planting in Sukumaland, Tanzania. Seve~ity-five holdings in three locations were studied, each with several hundred plots, and with only minor differences at the beginning of the rainy season. The percentage of the total area used for each crop sown in different months is shown. (From Rotenhan 1968, 62.)
r.
it is given a much that further cultivation appears uneconomical, then at once long-rest fallow of five to twenty years. Frequently several fields to cease to be cropped, so that we can speak offallow blocks as opposed cultivated blocks. cultivated (3) There is a medium-term fallow if a plot ofarable land within a of fallow form this of aim The years. five to area is left uncultivated for two of elimination the with fertility soil the is to combine the regeneration uf widespread. weeds. Medium-term fallows are particularly the aim of (4) The short-term fallow of one to two years has generally not such circumstances, other restoring soil fertility, but is caused rather by the make may which as the illness of a farmer or unfavourable weather, planting of the field difficult. can be The high degree of diversification of semi-permanent cropping practised as cultivation shifting from regarded partly as the inheritance land use. earlier, and partly as adaptation to a more intensive stage of 4.3.3. Characteristics of the fertilizer economy manuring The higher intensity of land use necessitates more highly evolved techniques, although it must be recognized that in most semi-permanent undeveloped. systems, as in shifting systems, the fertilizer economy is still either: on The continuity of cropping depends cropping, (1) high soil fertility, which is not visibly impaired by frequent regenerate to much in which case fallow periods are not interpolated so fertility as to eradicate weed growth; or period, which (2) the ability of the soil to regenerate during the fallow consists principally of grass; or (3) stabilization of yields at a low level. in the drier In high-altitude tropical areas with a temperate climate, and In these fertility. soil savannas, grass fallows are usually able to regenerate can be cultivation conditions and on very fertile soils, semi-permanent to apply however, regarded as a balanced land-use system. This does not, where savannas, the large areas of semi-permanent cultivation in the humid cropping repeated conditions, these In fertility. the soils are of only moderate deterioration with short, unfertilized grass fallows brings about a long-term gully-erosion particular in erosion, favours of the land. Repeated cultivation less effective are roots shallow with and sheet-erosion. Furthermore, grasses Port~res Thus rainfall. of action than trees or bush in reducing the leaching the soil Senegal of areas some (1956), for example, has demonstrated that in soil the out, washed are eventually loses its structure: clay components
65 SEMI-PERMANENT CULTIVATION SYSTEMS have fallen so (2) If the yields from a plot that has long been cultivated
66
'wecomes increasingly pervious, the proportion of sand in the topsoil increases, and the ability to retain water is reduced. In so far as cultivators have recognized the problem of soil mining and practise manuring, a number of fertility-restoring methods can be identified: (a) Shfting of the hut and kraal. It is in any case traditional for the home and farmstead to be shifted at fairly long intervals. The huts become dilapidated after some years, the site foul and eroded, the kitchen gardens reach the point of exhaustion, and it is better to move to a new site than to attempt to rehabilitate the old dwellings. These rebuilding moves are generally short, sometimes a mile or more but often no more than a few hundred yards... (Allan 1967, p. 7.) The wish to obtain an especially fertile plot that can sorve as a permanent garden for some time also motivates the shifting of the holding and the kraal. (b) Systematic folding. In some cases we also find a systematic folding economy (Froelich et al. 1963, p. 71; Dumont 1966, p. 65; Pelissier 1966). Where semi-permanent cultivation is combined with livestock-rearing, the cultivators can use the manure from the kraals. Transporting the manure to the fields, however, and digging it into the soil present considerable problems where there are no tracks, carts, or draught animals. The cultivators therefore prefer to move the kraal and, in the case of the Wasukuma (see 4.4.2), for example, they do so irregularly, thereby obtaining a series of manured plots. In some areas of Senegal, and on the west coast of Madagascar and Ethiopia, the kraals are moved regularly, so that fairly large plots are systematically manured. In these cases, it is the area around the farmstead which is treated in this way, with the consequence that it can be cropped more or less permanently (see 4.3.1). (c) Green manuring. A widespread practice in hoe cultivation, especially in ridge cultivation, is foi the fallow and weed vegetation to be dug systematically into the soil. Sometimes the gathering of green manure from further afield is organized as well. Pbssinger (1968, p. 117) reports, for example, that farmers in Luanda, Angola, artificially increase the size of termite mounds, which are preferred for cropping. The cultivators pile up maize straw, manioc stalks, and other plant waste in a circle around the mound, and the termites break down this plant material on the spot. (d) Mineral fertilizers. Where cotton, groundnuts, tobacco, and other annuals are grown as cash crops, more and more cultivators apply mineral fertilizers. Mineral fertilizer-when it is economically worth while-can be introduced comparatively easily in semi-permanent farming, because it can be applied efficiently on small plots without requiring tracks, carts, or much labour. At the same time, it can be stored without appreciable detcrioration, is easy to apply, and produces an increased crop after a relatively short period. For these reasons, it is usually easier to introduce the application of
SEMI-PERMANENT CULTIVATION SYSTEMS 67 mineral fertilizers in semi-permanent farming than to introduce manuring with available cattle dung. Even though the various stages of incipient fertilization may be worthy of note, it must nevertheless be remembered that they are mostly inadequate. Semi-permanent land-use systems often represent nothing more than degraded forms of previously balanced systems of shifting cultivation. 4. .4. Characteristics of animal husbandry Whereas shifting cultivation is largely practised with little or no livestock, in semi-permanent systems we often find large stocks of cattle. In extended areas of the African savannas, it is the interaction of arable cropping and cattle-keeping that keeps the tsetse fly at bay. Slash and burn agriculture thins out the forest or bush vegetation and destroys the breeding grounds of the tsetse fly; intensive grazing prevents bush regeneration and therefore pre. vents the insect from re-establishing itself. Arable cropping and animal husbandry in unregulated icy systems are consequently closely dependent on each other. The aims of stock-keeping are varied: (1) Cattle are kept to cover the risk of harvest failure or sickness. When land is not privately owned, cattle are kept as a means of support in old age. (2) In close relation to this there are social functions. Cattle act as brideprice, while a large herd is a status symbol. (3) Farmers want a supply of meat and milk for the household; sale of this produce is not very important. (4) A factor of increasing importance is the provision of the traction power for ox-plough cultivation. (5) Only in a few cases do semi-permanent cultivators consider the contribution of manure as an essential purpose of stock-keeping, although it might be of considerable relevance to their farming. Harvest residues, fallow grazing, and natural grazing provide the fodder. Communal use of grazing land is customary: everybody has the right to allow any number of animals to graze on the fallows, pastures, and stubbles. Fodder cropping is practically non-existent. Occasionally, balanced feeding is achieved by seasonally moving the livestock to grazing areas some distance away, which have not been used earlier in the season. The composition of the livestock herds depends above all on the availability of fodder. Where there is good grassland with sufficient watering places, the number of cattle per family is high, and a high proportion of the stock consists of male animals. With increasing shortage of grazing and higher cattle densitics, primarily female stock is kept, the male calves being
68
slaughtered soon after birth. Where fodder is not sufficient to keep large herds in one place, the lending out of cattle is common.
TABLE
4.3
Management indicators of cattle husbandry in Sukumuland, Tanzania Kwimba Shinyanga Svmi-permanent cultivation Number of cattle surveyed Average age of cattle
over 1 year
472
7.6
296
9.1
Average age of female animals Average age for bearing first calves (months), Calving interval (months)
Calves per 100 female
9.2 38 16 39.3
10.9 37 18 37.4
10.3 54 26 12.0
l0.0 39 18 36.5
animalsb
b Live
Animals with more than two calves. births. Sources: Rotenhan 1966, pp. 56-7. Table 4.3 illustrates the cattle economy in a typical dry savanna in three regions of Sukumaland, Tanzania, namely Shinyanga, where the R value is 39, with a relatively large amount of grazing land; Kwimba, which has an R value of about 33, with less grazing; and Ukerewe, where the R value is about 75, with permanent cultivation. One hundred cows between three and six years old can be expected to produce on average thirty-six calves, and both male and female calves are raised, although only 60 per cent reach their second year. A fully grown new generation of twenty-two animals is therefore produced by one hundred cows. They bear their first calves at thirty-nine months, and the intercalving period is eighteen months. Excluding the milk consumed by the calf, there is a yield of about 600 ki:ograms in a lactation period of 246 days. The number of animals consumed, including calves, constitutes 10 per cent of the total stock annually. There is no selling on a large scale. Of all the animals consumed, 75 per cent are those that die or have to be slaughtered, and only 25 per cent are slaughtered specifically for consumption. Efforts to control losses caused by disease by means of dipping, de-stocking campaigns, and projects to introduce organized use of grassland have not so far been successful. Table 4.3 shows clearly that performance is generally low and decreases with increasing shortage of fallows, :.e. with higher R values. An important sociological feature is the fact that the ownership of livestock is much more concentrated than the cropping of land. In Sukumaland, for example, 25 per cent of the families crop 55 per cent of the cultivation area, but 25 per cent own 80 per cent of the animals (see Fig. 4.4).
SEMI-PERMANENT
CULTIVATION
SYSTEMS
69
economy 3.5. Characteristics of the labourshifting and semi-permancnt systems may between Important differences degree of permanenCY so be found in the labour economy. The increasing induces rather cropping, cash of amount land use, and the increasing riking changes:
9o 80 70 60 50 40 30 20 10 0
Total stock
80
90 100
~1924
Livestock units per holdinq Tanzania (from Rotenhan property in Sukumaland, livestock of FIG. 4.4. Distribution 1966, p. 57).
and clearing in work pattern. Tree fellingmore tedious change a is There the (a) Technical. fire-farming shifting cultivators; the most timeare are the main jobs of weeding hoe and of the with cultivators, particularly tasks of cultivating semi-permanent the of and weeders. demanding activities carried out with ploughs be yet cannot for more where these processes case that people work the always almost more often (b) Economic. It is because semi-permanent cultivation is shifting cultivation, hours each day, firstly per unit of output than working time. demanding time mote not than additional cash cropping requires cultivators in Sukuand secondly because semi-permanent research, the growing season, According to Rotennan's of 6 hours a day during average an for work maland
70
SEMI-PERMANENT CULTIVATION SYSTEMS which lasts for about eight months (1966, p. 70). Haswell (1963) gives a similar figure of 6.4 hours daily for groundnut holdings in Gambia. (c) Sociological. We find a change from the rather rigid traditional division of labour between the sexes, which is still typical of many shifting systems, towards a more flexible use of the available labour force. The demands of cash cropping seem to be the driving force behind this. An important feature is the increasing involvement of men in field work such as weeding and harvesting. This in turn indicates a new tendency in farm management. Traditionally, several persons in the household cultivate the food they need for themselves and their dependants on a plot that might be considered a sub-farm within the family holding. With increasing permanency and commercialization, land and labour use within a holding seem to become more centrally organized.
Typical of semi-permanent land use in the African savannas is the cultivation with hand-tools of comparatively large areas of 2-8 hectares per family. The labour input per hectare is rather high. A lot of hard work has to be done in the growing season. Table 4.4 gives an idea of the size and composition of the labour input in several savanna crops. On the light soils of Senegal, where a deep cultivation of the soil before cropping is usually not practised, the input in land preparation and planting is comparatively low. The labour input is much higher in the classical hoe-farming areas of Sukumaland, where careful ridge cultivation is practised. Almost 400 hours per hectare are required for land preparation under these circumstances. The cultivated area varies accordingly. 6-8 hectares of crops is the average figure per household in Senegal, while 3 hectares is the norm in Sukumaland. A significant change in the work s;tuation is brought about by the introduction of the ox plough, which is technically possible wherever areas with a fallow vegetation of gras; have been de-stumped. First and foremost, it eases the burden of work .y relieving the hard toil of land cultivation. At the same time, fields of mostly regular rectangular shape are formed, the size of fields increases, and the cropped area is expanded at the expense of the fallow (see Table 4.5). Where the ox plough is not combined with the oxdrawn weeder and row cultivation, as, for example, in cotton cropping in Sukumaland, the labour economy in cultivation may easily be cancelled out by the greater work required for weeding. For the full success of animal traction, to obtain the greatest possible reduction in the work-load, the oxdrawn weeder and cart, as well as the plough, are necessary. The great reduction of labour input in groundnut cultivation in Senegal stems from this combination (see Table 4.4). A further feature of this land-use system is the seasonal nature of labour demand. In shifting systems, the task of clearing does not have the same rigid timetable as has cultivation for the semi-permanent cultivator. He must wait
4.4
Gro
u nd n u
ts
'd
456
1520
755
476
27). production agricole (1965, p. Sources: : Bureau pour le d~veloppement de la agricoie (1965, pp. 214, 238). production la de d~veloppment le Bureau pour table 6, p. 75). (1965, p. 35). C Rotenhan (1968, de la production agricole pour le developpement ,Bureau
72
4.5
until the beginning of the rainy season to cultivate, whereas the shifting cultivator can carry out clearance work in the dry season. Over and above this, there are sometimes certain jobs that are fixed to a timetable and require a heavy labour input in weeding and often in harvesting. Such is the case, for example, with groundnuts. As Fig. 4.5 shows, both in Sukumaland and
24 22 20 18
(a)
24 22 20 18
(b)
~16
12
16
12
88
6 -6 4 4
0 NA
El
0
JJ FM JA SA
Cultivating
Weeding
M J JA SO0N DI J FM Harvesting
FIG. 4.5. Seasonal distribution of field work in semi-permanent farming systems of Tanzania and Senegal (from Lacombe-Orlhac 1967, annex 1; Rotenhan 1968, p. 63). (a) Cottonmillet holdings in Sukumaland, Tanzania. (b) Groundnut-millet holdings in Senegal.
SEMI-PERMANENT
CULTIVATION SYSTEMS
73
Senegal 90 per cent of the field work is concentrated into six months, and 50-60 per cent in only three of those months. Since tield work, apart from herding of cattle, is almost the total work of the holding, considerable underemployment occurs during the remaining months. In this connection, we must bear in mind that the number of work-hours per day does not give a satisfactory idea of the different labour demands. In seasons of peak demand on labour, performance per hour is considerably higher. Finally, it should be pointed out that the labour economy of partly commercialized, semi-permanent systems is usually far removed from the uniformity of traditional subsistence farming. Differences in the man to land ratio, differences in the choice of crops and the adoption of innovations, different access to development services, and different opportunities for cash cropping-these factors all combine to bring about a wide variation in the size of enterprises and in the smallholders' performance within a given location. In most semi-permanent systems, only a few hectares are cultivated and a few animals kept, i.e. the size of agricultural enterprise is rather small. This, however, does not mean that the average area cultivated by each family or each worker is about the same. Within each type of farming, even wher-. land is still ample, we find striking differences as to the area cultivated per family and per member of the available labour force. The equality that characterizes traditional shifting systems, in which each family cultivates only what is required to cover subsistence needs and everyone uses the same practices, no longer exists. Within semi-permanent cultivation systems, we find larger and smaller units, richer and poorer families. The differentiation in cropping is the more noticeable the greater the extent of land shortage and the higher the degree of commercialization. Even more pronounced are the variations in performance. On average the hours of field work per available man-equivalent are low. There are, however, wide fluctuations in the input of work per hectare and returns per hour of work within the range of production possibilities offered by a semi-permanent system. Farm-managerial ability, pa, ticularly in cash cropping, is one of the major factors differentiating progressive producers from others. The explanation is an obvious one: semi-permanent cultivators differ widely, much more than shifting cultivators, as to their physical effort, their drive, and their knowledge. 4.4. Two case studies 4.4. 1. Semi-permanentfarming in the subhumid savanna: rice holdings in the Kilombero Valley, Tanzania The Kilombero Valley in Tanzania is warm and humid, receiving 1200year. 2000 millimetres of rain annually. Most of the valley is flooded every consists plain the of The vegetation on the heavy, black-cotton soil ('mbuga')
74
Productivity Hours of field work per ME 1014c Hours uf field work per ha 1250c Hectares cultivated per ME 0.81c Gross return (S/ha cultivated) d 57 1 Gross return ($ ME- ) 46 Value addedo (S/ha cultivated) 50 Value added ($ ME-') 41
Areas over wnich the farmer has a right of usage. Percentage of occupied and arable ;and that is copped. c Four holdings only. d The comparability is limited because of livestock returns which are not derived from cultivated area. Value added = income in smallholdings (gross return minus purchased inputs) = net return + wages, salaries, taxes, etc., on large farms. Sources: Baum (1968, p. 40); Rotenhan (1968, p. 55).
75
mainly of elephant grass. Most of the valley is empty or thinly populated. The population is concentrated in the Kiberege Strip, lying on the western side of the valley, which throughout the year has access to a road, and therefore to a market. Higher population density along the road has given rise to the transition from shifting to semi-permanent cultivation. The concentration of population has brought about a relative shortage of land. The right of use is strictly observed even for fallow land. In a pilot study of fourteen holdings, the total average arable area amounted to 3.8 hectares, of which 2.2 hectares (57 per cent) were cropped each year. Land is brought into cultivation in the sequence grass clearance-burning-'iceing-broadcast sowing of rice. Two or three years of rice, followed by the same period of fallow, is the normal rotation. The encroachment of grass into the field is the main reason for fallowing. Rice cultivation, with yields of about 10 quintals per hectare, is supplemenicd by manioc or wet-season maize, beans, and sweet potatoes. Housing has become stationary. Bananas and papayas are grown near the hut, and the older the settlement the higher and larger are the mango and coconut groves nearby. Livestock is restricted to a few goats, sheep, and chickens, and the presence of the tsetse fly has hitherto prevented cattle-keeping from being introduced. Table 4.6 shows the econoinic returns from this type of farming. The gross return from cropping amounts to $125, with a labour input of 2750 hours of field work per annum, resulting in a gross return per work-hour of about S0.05. Only a small part, 16 per cent of the gross return, is sold. These are therefore typically subsistence holdings. 4.4.2. Unregulated ley farming in the drier savanna: cotton-maize-mille: holdings in Sukumaland, Tanzania An example of arable farming in the drier savanna is the Wasukuma's land use in the cultivation steppe to the south of Lake Victoria which has a rainfall of 600-800 millimetres during the six-month rainy season. A typical Sukuma farm is a strip of land stretching from the top of a hill to the bottom of a valley, which includes different types of soil carrying different crops (see Figs. 4.1 and 4.6). Cropping per holding amounts to 2.3 hectares annually and takes up about half the land available for cultivation. 1-1.5 hectares are cropped with cotton. 1.5-2 hectares carry maize, sorghu.i, legumes, and manioc for personal consumption. Rice and sweet potatoes are cropped on separate plots in the valley bottoms. The different ways in which the land is used are the result less of rotating various crops than of successively cropping the plant or plant mixture that thrives best on the particular soils until the yields decline. The last crops to be grown are manioc and millets, and then the land goes back to grass and the plot becomes grazing fallow for a number of years. Normally six to eight years of fallow succeed six to eight years of cropping. The holdings are defined by clear boundaries, which are frequently sisal hedges, whose leaves are processed in the off-season. In some parts of Sukumaland, dearly recognizable field boundaries have already been developed, in addition to the fixed holding boundaries. The soil is still cultivated for the most part with the hoe. Cotton and crops for home consumption are planted in ridges 1.3-1.5 metres apart.
SEMI-PERMANENT CULTIVATION SYSTEMS In the north-eastern and southern border areas, ox ploughs are increasingly used. In some areas, tractor ploughing has been introduced. A holding normally has 7-8 livestock units the form of II Zebu cattle, 4 sheep, 4 goats, and a number of chickens. Thanks to iin a variety of soil have a natural system of fodder supply. In the dry season, types, the Wasukuma from the end of June to the beginning of Ortober, the wide valley bottoms ('mbuga'), where a Ict of fodder grows, lose their swampy character and can be grazed. In the rainy season the fallow land is grazed. Households that possess a larger number ,f livestock arrtrge
11170M
76
.ISv.
-.
i~
Slp
V
V
:v
- -. -- -- -- -v -,~ ' v v -
1210M"N
. Y
50
lum
0 0
...
.
Footpath Soma
o0
Huts
FIG. 4.6. Land use in a Sukuma holding, Tanzania (from Ludwig 1967, p. 104).
for sc ne of the animals to be grazed in remoter steppe areas. However, arable farming and animal husbandry are not integrated. The livestock is fed on pastures, fallow, and harvest residues. Cattle dung from the kraals is not utilized unless the kraal is siifted and a garden is laid out on the old site. Mineral fertilizer is being applied increasingly to cotton. Rotenhan (1968; see Table 4.6) estimates the average gross return of a holding to be $206-341. TI.e costs for purchased inputs and wages are small; the r(Rmily income comes to only slightly less than the gross value of production. There is an income of $66-93 for 709-857 hours of field labour ?er worker, i.e. $0.09-0.10 per hour of work.
4,5.
Semi-permanent cultivation in partly commercialized holdings is distinguished from shifting systems both by more intensive land use and by the
77 SYSTEMS SEMI-PERMANENT CULTIVATION more easily applied. People live in :t thet technical innovations can be New varieties and production .tionary housing with permanent gardens. a major cash crop is grown, where ,thods can make headway, especially with ploughs is possible where the :e cotton or groundnuts. Cultivation the same time, the first steps are being llow vegetation consists of grass. At of perennial crops, permanent roads, ken in investing labour in plantations nd other long-term capital works. .5.1. The tendency towards soil mining a number of basic drawbacks Desrite these de elopment advantages, are those of maintaining soil fertility. .hould be noted. The main problems and the fodder cropping are exceptional, .anuring with animal dung and usually marginal because of unfavourable ffectiveness of mineral fertilizers is .in husbandry, particularly in weeding. price relations and short-comings per hectare are often less than in shifting Other things being equal, the yields has areas, especially where the plough systems. At the same time, in many semi-permais on the increase. Normally replaced the hoe, erosion damage it a stable system of land use, since nent cultivation does not constitute especially gradual forms of soil mining, generally amounts to either rapid or in the drier savannas. economy 4.5.2. Degeneration of the livestock systems is the degeneration semi-permanent Another conspicuous aspect of grazing becomes scarce. Static animalof the livestock economy once farming system almost inevitably lead husbandry practices within a changing grazing land available, the herds to problems. As long as there is sufficient are well fed. An increase in the cattle per family are of a good size and the a reduction of the grazing area. necessitates farming area needed for arable crop farming are partly being invested Simultaneously, the proceeds from in and status. The results are an increase in cattle, in order to gain security in the grazing area, causing complethe cattle population and a reduction cattle deteriorates and erosion inthe mentary damage. The condition of proportion of the rain runs off as flash high a creases. On over-grazed land, thickening of bush, particularly those floods. Over-grazing encourages the browse. Thus in the long run over-grazing species on which livestock do not the grazing area, reduces the length and reduces the carrying capacity of the clearance of and increases the amount effectiveness of the fallow period, work. grazing is at the root of the probThe practice of unorganized communal to graze all his cattle on the available lem. Usually everybody has the right of the grazing economy is of grazing, including the fallows. No improvement are limited to the carrying capacity value, however, unless the stock numbers
78
of the grazing land, and unless the movement of stock within any given area is controlled by a rotational grazing system. 4.5.3. Bottlenecks in the labour economy In nost semi-permanent farming systems only about one-third to twothirds of the available labour capacity is absorbed in field work. Nevertheless, cultivators almost everywhere, regardless of the degree of underemployment, unanimously feel that labour shortage is the most important factor limiting output. When asked about this, they almost invariably argue that they would like to produce more and earn more, but they are not in a position either to cultivate larger areas, or to change to labour-demanding crops, or to cultivate, plant, weed, and harvest in time. or to increase the labour input per hectare of a given crop, because there is not enough free family labour available during the relevant weeks, and there are not enough funds to pay hired labour. Much, if not most, of the cultivated land is neit'aer planted nor weeded according to a timetable that would maximize yields. The fact that important labour shortages within family holdings and a high degree of underemployment go hand in hand may be traced back to a number of major factors: (1) It is well known that labour peaks in agriculture, particularly with arable crops, occur seasonally. The Wasukuma, for instance, have a great deal of work to do between December and April. In the second half of the summer there is no field work at all. (2) Agricultural work is not the only work that has to be performed in smallholdings. In fourteen cotton-maize holdings in Machakos District, Kenya, Heyer (1965) established that on average throughout the year the relation of work-hours on the land to household work is 1 to 1.7, i.e. general household work is much more time consuming than field work. (3) The labour capacity of the various persons in the household is not utilized equally, because of traditional concepts of the way in which work is divided between men and women. It may be, for instance, that the women cannot weed in time because the work connected with the household demands too much attention. On the other hand, the men who are unemployed at this time do not consider that tending the food crops should be part of their work. (4) Another important factor is the low efficiency per hour of work. This is caused for one thing by unsuitable hand implements. In Sukumaland, for instance, the weeds are hoed with the same hoe that is used for ridging. The usual hoe is well adapted to the latter activity, but not at all to the former. In addition, smallholders are accustomed to making a concentrated, sustained effort only in connection with a few procedures, such as felling, clearing, and hoe cultivation. The other jobs have traditionally been done in a leisurely manner.
79 SEMI-PERMANENT CULTIVATION SYSTEMS Finally, the extent of underemployment is calculated according to ns of working hours per year aid per man-equivalent. The smallJers who refer to difficulties in the labour economy are naturally not diliar with such norms. Many of them do not accomplish more work :ause they regard a five-hour day, even at planting time, as a complete .lization (. .heir working capacity. lany of the problems of the labour economy in semi-permanent land use e because these systems are a transitional phase between shifting and manent land use. The advances in the labour economy that are characterc of permanent arable farming-row cultivation, ploughing, mechanical :eding, proper farm layout, and so on-have not yet been introduced on large scale. Usually the transition to cash cropping, with its consequent bour demands, takes place within a fairly short period, before the farmers re accustomed to longer hours of work per day. On the other hand, the ropping areas are relatively large, and the labour requirement is seasonal. 2onsequently, there is a lively interest among semi-permanent cultivators in Labour-saving innovations, even more than in yield-increasing innovations. Farmers in Sukumaland do not generally change from hoe work to ox ploughing to gain higher returns, but to replace hard work with the hoe by easier work with the plough. Particularly striking is the general interest in tractors. Wherever finance is available, and the return per hectare high enough to justify tractor ploughing, smallholders endeavour to buy tractors or to hire tractor services.
4.6.
Production increase in semi-permanent land-use systems normally results from extending the crop area by clearing the land in neighbouring extension areas and from reducing the fallow area. It is most obvious that measures must be introduced that will lead to higher returns per hectare if involution is not to replace evolution. Classification of the possibilities for higher returns per hectare entails differentiation between innovations that give rise to improved semi-permanent systems, and new combinations that eventually create other, more productive systems. 4.6.1. Improved systems In consideriag field cultivation, we must emphasize the supreme importance of better husbandry. Established routines of work, which might have been the best in shifting systems, are still practised in semi-permanent systems, although the actual farming is more intensive. Consequently, there is much
80
SEMI-PERMANENT CULTIVATION SYSTEMS that scope for better husbandry. The main short-comings in husbandry could be at least partially remedied are usually the following: (1) Bottlenecks in the labour economy often result in late planting. (2) Most farmers tend to use plant spacing that is greater than is advisable. of an They aim at robust individual plants and overlook the advantages optimum plant density. (3) No other factor is as much to blame for poor yields as late weeding. This can be only partly explained by labour problems. Probably farmers would harvest more if they planted smaller plots and devoted more care to weeding. (4) There is further scope for the introduction of new varieties. The cultivation of better-yielding types of cotton and groundnuts brought noticeable successes in a number of cases. (5) The lack of field roads and carts makes the application of kraal the manure difficult. This makes the introduction of mineral fertilizers all more relevant in some locations. (6) Few other measures promise such great increases in returns in relation an oxto the additional costs as the supplementation of the plough by cultivation. row drawn weeder, in conjunction with the introduction of (7) The quality of ploughing with oxen is very poor even where ploughs also are used. Proper ploughing would not only raise yields, but would significantly reduce the weeding task. (8) Where it is felt to be worth-while to keep tractors, very real improvements are possible, especially timely cultivation, the better ploughing-in of weeds, deeper ploughing, and more effective use of mineral fertilizers. of N'dama breeds, (9) In areas infested by the tsetse fly, the introduction in order to considered, be which are resistant to trypanosomes should reap the benefits that cattle may bring. step Semi-permanent land-use systems as a whole represent " promising and fallows, witi. farming into towards introducing technical innovations of app!ication for scope The therefore towards agricultural development. systems. innovations is incomparably greater than in shifting 4.6.2. Change towards more intensive types offarming In most semi-permanent land-use systems, grass or bush vegetation grows seek the in the fallow period and is then grazed. It is therefore natural to solution to the problem caused by soil mining in semi-permanent farming This through the change-over from unregulated to regulated ley systems. erinvolves replacement of the several years of fallow, with their natural regt mixture plant or plant a of cultivation ation of grass and bush, by deliberate which is more productive than the fallow vegetation, regenerates szil fertility, in the and allows plough cultivation. However, the ecosicrnics of Icy farming
81
drier areas are very tropics are not yet proved, and short-term leys in the or manure of unlikely to maintain fertility without liberal additions fertilizers (see 5.6). system should The expense of such a fundamental change in the farming investment in also be borne in mind. Ley systems call for considerable buildings. The cost clearing, livestock, fences, implements, carts, tracks, and amounts to two or of draught oxen and their necessary accessories alone these holdings. three times as much as the gross annual earnings from consolidation. land is farming ley Another pre-condition for regulated cultivaplough for enough large fields Holdings have to be in one block, with tion. further obThe fact that a 'technological jump' must be made presents series of whole above stacles. In the holdings in Sukumaland mentioned and follow, to is success innovations have to be adopted in combination if to enough not simply the same applies to most semi-permanent systems. It is the to kraals the encourage the use of ox-carts to take animal manure from are none, unpaid fields, since the tracks also have to exist. But since there communal work is necessary to build them. seeds and Farmers can scarcely be expected to establish leys and buy economic return. mineral fertilizer for them without obtaining a direct and storage transport, harvest, the Keeping dairy animals again necessitates The healthy. animals the keeping all, of fodder, correct feeding, and, above ecothe but technically, solved be problems that are raised can in general too low to be a nomics are u:ually marginal. The producer's prices are mostly scarcely moreover, can, sufficient incentive to change. The technological jump support active the or be accomplished by an individual without the consent farming dairy of the group, and this further complicates the issue. Profitable involves keeping breeds with a high milk yield, maintaining their health, The effectiveness of and making available suitable male animals or semen. unless neighbouring the necessary ,-:.rinary care on a single farm is doubtful traditional rights, with clashes animals are also treated. Fodder cropping animals for everybody's for available been because the fallow pasturage has the fallow on plants fodder crop to generations. Thus, for anyoae wanting It is rights. grazing their renounce it is necessary for the others to agree to Icy regulated to transfer a generally the case that recommendations for etc., manuring, mineral and systems with dairying, fodder cropping, dung land shortage for a can be expected to s .c:eed only where there has been a is little desire to number of decades. Wnere there is no such pressure, there fertility, farmers change, especially as, even with the continuing loss of soil can still gain a comparatively high income. is through Another way of introducing stable systems of rain-fed farming in a settlement plan settlements under close supervision. Farmers taking part fertilizing, have to obey the project leader in matters concerning rotations,
7
82
animal-keeping, fodder, veterinary supervision, and so on. The lack of knowledge and finance that prevent the individual from making the leap into proper ley farming are counteracted in the settlement by the rules of the supervisor and the credit afforded by the supporters of the settlement. In Africa, south of the Sahara, there are numerous examples of this kind of approach to the modernization of agriculture, but they usually reveal an unsatisfactory ratio between input and yield. Most settlements need continual financial support, or they are abandoned after a few years. The reasons are usually quite straightforward: on the one hand, the cost of personnel for supervision is high; on the other hand, the land that is mostly used is capable of producing only moderate yields. Close supervision, which can be rewarding in irrigation agriculture and with certain perennial crops, is normally not worth while with ley systems, unless cash crops with a high return per unit of area, like tobacco, can be cultivated. Given this situation, for the time being probably nothing can be done but to accept forms of semi-permanent cultivation that are detrimental to soil preservation. The long-term hope is that increasing ..,mercialization will demonstrate more and more clearly the inadequacy of the system and create enough incentive for the farmers to remedy the situation. It is very possible that the judgment of Rounce (1949) applies generally to semi-permanent cultivation. He wrote about Sukumaland: 'A situation of overpopulated villages and impoverished soils must be passed before better husbandry can spread' (quoted from Rotenhan 1966, p. 52).
5. Syst
5.1.
Definition
WE speak of a regulated ley system where several years of arable cropping are followed by several years of ley. In British nomenclature, only those fields that are deliberately planted to grass, clover, or lucerne are normally considered as ley. This, however, is too narrow a definition for the tropics. The German nomenclature, where the word 'ley' is used wherever several years of arable cropping are followed by several years of grass and clover utilized for livestock production, is preferable. A distinction is then made between unregulated and regulated ley systems. Unregulated ley systems in the tropics are characterized by a natural fallow vegetation of various grass species, a certain amount of bush growth on the pasture, communal grazing, and a lack of pasture management, all of which make such systems often more akin to short-term fallow systems than to regulated ILy systems. Individual grazing, fencing, pasture management, and rotational use of the grassland are the usual characteristics of regulated ley systems. The seeding or planting of grass leys, the production of hay or silage, and the application of mineral fertilizer are indicators of intensive types. Ley systems show a more or less regular alternation between arable farming and leys. The sum of the arable years and the ley years tells us the rotation cycle. The leys may be used either for grazing or as meadows producing roughage for stabled cattle. In -1 nost all tropical ley systems both regulated and unregulated grazing are practised. It should be noted, however, that the border line between unregulated and regulated systems is often difficult to draw. 5.2. Types of ley farming and their geographical distribution Regulated ley systems, which are widespread in the subtropics of the northern and southern hemisphere, are rare in the tropics. Those that do occur are more or less restricted to areas that show one or more special features, namely: (1) Those where vigorous growth of palatable grasses is possible.
83
84
(2) Those where the dependence on the long-fallow system is not pronounced, i.e. where relatively few years of grass are sufficient to restore soil fertility. (3) Those where land becomes so scarce that intensification is required beyond permanent grazing or unregulated ley farming, yet is not so scarce that permanent cropping is necessary. (4) Those where the health risks of intensive animal-keeping are comparatively low. (5) Those where price relations favour intensive forms of meat and milk production. (6) Those where sufficient draught animals or tractors are available to plough the leys. (7) Those where farmers have sufficient knowledge and capital to organise a proper fodder economy over the year. These conditions are often to be found in high-altitude large-farm areas. In addition, some types of regulated ley farming have developed in the drier savannas, but ley systems in the humid tropics are still in the experimental stage. The organization of ley systems is closely related to farm size. We can therefore classify the main forms into three principal groups. 5.2. 1. Regulated ley systems in traditional smallholder areas There are but few cases where the evolution of indigenous farming leads to regulated ley systems. A noteworthy exception is the land system of the Serere in Senegal, where cropping and grazing are systematically alternated. Even more important is the ley system in the highlands of Ethiopia, where the vigorous growth of palatable grasses, the use of the horse-drawn or oxdrawn plough, and permanent rights of land-ownership seem to have been the decisive factors that induced the change from unregulated to regulated ley farming. A unique feature or the Ethiopian Icy system is the cultivation of a Festuceae called teff (Eragrostis abyssinica). Teff is a grain crop which supplies a highly appreciated food. Agronomically, however, it is much more a ley grass than a grain crop, combining as it does the soil-restoring and fodder-producing properties of a ley grass with food production.' In other African areas, ley farming has been introduced through the development efforts of various agricultural extension departments. Cases in point are the economically important ley systems in the Kano Plains, Nigeria, in medium-altitude areas of Uganda (Teso system), and in high-altitude areas of Kenya. The following rotation cycle, with six arable and six ley years, which is practised in the
area around P *' i in tho Ethiopian highlands, actually comprises eight years of grass, because of the pr.:tnce of teff in the rotation: teff-maize-nigerseed (Quireta abyssinica)teff-maize-niger:.ed-six years of unimproved leys.
85
activities in these landThe intensification and modernization of the arable by improved animal use systems is, however, not usually accompanied Fodder production on the husbandry. Unimproved local breeds are kept. year. The return of meat leys is rather low and unevenly distributed over the livestock is low as well. and milk per hectare and per inadequately fed head of areas in Kenya smallholder Noteworthy exceptions are some traditional 1964). MacArthur (see cattle with intensive dairy production based on grade 5.2.2. Regulated ley systems in settlement schemes schemes, most The African savannas are literally dotted with settlement animal modern and systems of which aim at the introduction of regulated ley conusually is settlements husbandry. The organization of production within that innovations of set sidered as an instrument for introducing the complete Kenya the In technology. is required for the jump into a new and demanding led to the establishestates and farms ex-European of highlands, settlement and intensive milk proment of numerous smallholdings with ley farming a certain degree of duction as major planning objectives, which achieved schemes, which success. However, few of the numerous smaller settlement of East and West are to be found in medium-altitude and low-altitude areas with leys, farming mixed incorporate Africa, and which were planned to They sense. economic an or have proved to be successful in either a technical as regarded be yet cannot but may be considered worthwhile pilot schemes, economically viable and established farming systems. 5.2.3. Regulated ley systems in largefarms mechanized farms, priMost established ley systems are found on large, a complete clearance with only marily because the use of the tractor is possible grass. Furthermore, of of bush and stumps, i.e. it requires a fallow vegetation animal husbandry, of forms the profitability of the leys depends on modern farms. small in than which are much more easily introduced in large the most among are The following examples of large-farm ley systems important: of Zambia and Rho(1) The tobacco farms in the drier African savanna of the ley is not to desia are largely based on leys. The prime objective produce fodder but to free the soil from nematodes. and beef and dairy (2) In the Kenya highlands, wheat, pyrethrum, sheep, modern ley farms (Maccattle are the main branches of activity in large are to be found in the Arthur and England 1963). Less evolved ley systems 1967). highlands of Angola and Mozambique (PFssinger to be found in South are farm ley (3) Numerous varied types of regulated Sugar-cane growers savannas. and Central America, mainly in the humid rotational reasons for in Cuba put leys between two sequences of sugar-cane in southern growers Coffee and in order to ease peaks in labour demand.
86
SYSTEMS WITH REGULATED LEY FARMING Brazil cultivate some of their crop lands within ley systems, because dairying has become a paying proposition. In addition, there are a great number of ley systems at various altitudes in the Andes. (4) Irrigated land in sparsely populated areas is occasionally farmed along the lines of a ley system. A case in point is the combination of two to three years of rice followed by five to ten years of irrigated pasture in New South Wales, Australia.
5.3. General characteristics of regulated ley systems 5.3.1. Principles of crop and ley management Cropping becomes clearer and more uniform as farming is carried on within clearly demarcated fields. The various distinct patterns in the spatial organization of cropping that we find in shifting and semi-permanent systems are rarely found in proper ley systems. Fields are more uniformly cultivated, and planted predominantly in pure stands. Pseudc-rotations, with overlapping harvests and planting dates, are replaced by clear-cut crop rotations. Nutrient concentration around the homestead is uncommon because of the available transport facilities (carts, tractors, field roads) on the one hand, and because of the preference for night grazing on the other. The relevant farm-management features are rather different from those of unregulated ley systems. In the analysis ofregulated Icy systems, the following aspects are of most importance: (a) The ratio of arable years to ley years or of ley area to crop area is an indicator of the intensity of land use. Tn general, the number of grass years is relatively high and the ley system extensive where: (1) soils are heavy and wet, as they frequently are in some high-altitude areas; (2) crop yields are uncertain, as in the drier savannas; (3) the land has steep slopes; (4) labour is expensive; and (5) price conditions are more favourable to stock-keeping than to arable farming. (b) The cycle of arable and grass years. Long-term cycles between cropping and grass years are indicative of extensively organized ley systems. Shortterm cycles occur principally where the holding has a relatively large amount of permanent grazing and where, therefore, the livestock economy is not absolutely dependent on the fodder contribution from the leys. On the other hand, long-term cycles are preferred: (1) on heavy soils; (2) where traction power for ploughing the icy is scarce; and (3) where establishing leys is expensive.
87 The number of ley years depends, moreover, on the choice of the grasses and legumes grown. (c) The intensity of cropping. In temperate climates the management of leys and stock exercises a great influence on the pattern of arable farming within the holding. In most tropical ley systems these interactions are less pronounced. The main emphasis is usually on cash crops during the arable years. In addition, some arable fodder is grown to bridge the dry season. Otherwise, cropping and the use of the leys are two distinct activities. (d) The intensity of ley management. This varies widely. As far as ley establishment is concerned, the following systems are distinguishable: (1) Often a sign of extensive Icy farming is a natural grass regeneration. This, however, may not be true where grass growth is prolific. In the highlands of Kenya and Ethiopia, naturally regenerated grass is found even in comparatively intensively farmed holdings. Smallholders rely almost entirely on natural grass regeneration. (2) Ley establishment by the sowing of grass seed is in general restricted to large holdings. (3) A number of especially productive grasses, e.g. ,'kuyu grass and elephant grass, are always established by the planting of vegetation material, not seeds. Planting is also the usual practice where (for example in the highlands of Cenya) the change is made in smallholdings from naturally to artificially established swards. In any case, planting depends on the availability of ample labour. (4) Ley systems achieve the highest productivity where it is possible to sow clover-grass mixtures. Leys can be distinguished on the basis of management according to: (1) whether the leys are fenced; (2) whether rotational grazing is practised; and (3) the intensity of labour input in grassland management. The usual way in which leys are used is for grazing. In some cases grass is also cut to make hay or silage. In the tropical highlands, however, haymaking suffers because of the heavy dew, or because it coincides with the rainy season. Thus, it is usually more advantageous to meet the fodder requirement with roughage and silage. 5.3.2. Characteristicsof thefertilizer economy The continuity of cropping in ley systems is usually built on the soilconserving and fertility-regenerating capacity ofgrass leys. Regulated systems fulfil these functions better than unregulated systems for a number of reasons: (1) The fencing of leys and the systematic rotation of cropping and grass reduce the langer of over-grazing and the resulting erosion damage.
88
SYSTEMS WITH REGULATED LEY FARMING (2) The systematic establishment, management, and use of the leys bring about a better snil coverage and allow the development of a dense root system. (3) Regulated ley systems are occasionally combined with folding systems, in which there is more manure. because of the more intensive stocking. (4) Where ley farming is practised with clover-grass mixtures, as for instance in the highlands of Kenya, not only is there a significant increase in the quantity and quality of fodder, but an increase in nitrogen fixation is also brought about.
It should be stressed, however, that the soil-preserving characteristics of leys as grown in temperate climates or at high altitudes cannot be simply repeated in medium-altitude and low-altitude tropics. Webster and Wilson (1967, p. 318) write in this connection: The grass Icy probably has a modest, direct, beneficial effect on the nutrient and physical status of the surface soil, and, compared with continuous cultivation, a rotation including a grass break is likely to be advantageous in reducing erosion and decreasing the incidence of weeds, pests and diseases in the arable crops. However, unfertilized grass leys are no more effective in maintaining fertility than are fallows of natural regeneration. A large number of well-tried clover-grass mixtures have recently been developed for the subtropics and some tropicai high-altitude areas. In Australia and Kenya they are already being extensively used. However, there are to date no correspondingly suitable legumes for incorporation in leys in low-altitude or medium-altitude tropical areas. Mineral fertilizing is becoming increasingly important in commercialized ley holdings with extensive cash cropping. However, in these holdings mineral fertilizer is used specifically for the cash crops. Only in exceptional cases so far has mineral fertilizer been used on grass leys. 5.3.3. Characteristics of animal husbandry The obstacles preventing rationalization of the animal economy that are found in unregulated ley systems with communal grazing exist in regulated Icy systems only where communal grazing is not replaced by definite grazing rights on each holding. In so far as this has occurred, there is a tendency for more intensive forms of animal htsbandry to be employed, whereby the production of beef and heifers tends to be of primary importance on large holdings, and milk production on small holdings. Regulated ley systems frequently show advanced husbandry practices. On settlement schemes and large farms, we usually find improved or grade cattle and planned breeding programmes. Whereas semi-permanent cultivators keep their animals overnight in a kraal with fencing, in regulated Icy systems night grazing is possible, and the need to walk long distances from
89 LEY FARMING SYSTEMS WITH REGULATED Furthermore, and water is thus obviated. the kraal to the grazing areas separately. cows graze calves, heifers, and fodder fencing makes it possible to intensive ley systems is the more Another characteristic of regulated standard are and arable fodder production in tropical economy: rotational grazing concentrates, which is important procedures. However, the use of systems. ley farming, plays a lesser role in systems with permanent arable favourable especially afford rainfall The tropical highlands with sufficient in regulated ley systems. It is much conditions for low-cost animal production healthy in the cool mountain areas animals may easier and cheaper to keep grade is evenly distributed, the cattle rainfall the When climates. hotter than in fodder rmght provide some additional in graze almost all the year. Arable example of low-cost animal production the of roughage in a dry spell. A good parts some of the large-farm economy regulated ley systems occurs in Kenya Highlands. economy 5.3.4. Characteristics of the labour are mainly hoe economies, in regulated power Whereas unregulated ley systems largely because of the traction systems we find plough cultivation, holdings are too small to keep draught the for required to plough the leys. Where undertaken by a contractor, as usually is ploughing tractors, or animals example is widely practised in Kenya. a low labour input per unit of output land Ley systems as a whole may require The weeds that grow on arable in relation to other farming systems. In fact, the labour input in weeding years. are suppressed during the grass Whereas the in systems with permanent farming. with the is very often less than largely done be to still clearance work in unregulated systems has areas most In sward is broken with the plough. hoe, in regulated systems the proposition. becomes a financially viable the use of tractors in ley systems involves the that work in animal husbandry A further important point is labour. relatively regular employment of
90
SYSTEMS WITH REGULATED LEY FARMING difficult to produce, but it yields harvest of 1000 kilograms, worth $1000, of dried tobacco per hectare. In the sample case, tobacco constitutes practically the only cash product. Maize, beans, and cattle are used mainly for pyments in kind to farm workers.
TABLE 5.1
900 Tobacco-holding 1962-3 Model 48Ab 850 600 62 35 27 90 1 000 1000 1200 38 000 92% 19950 36% 19%
18050
Less wages Profit Productivity Gross return Gross return Gross return Value added Value added Value added
a
8 550 9 500 (S/ha potential arable land) (S/ha cultivated land) ($ ME-) (S/ha potential arable land) (S/ha cultivated land) ($ ME-') 63 613 792 30 291 376
Key: man 19-60 years = 1.00 ME man 15-19 years = 0.67 ME woman 19-60 years = 0.67 ME woman 15-19 years = 0.50 ME child 10-15 years = 0.25 ME b One-third of the ME are women and children. Livestock produce, maize, and beans are mainly used for payments in kind. Sources: Haviland (1952); Rhodesia Ministry of Agriculture (1963, p. 12).
The chief problem in the farm economy of these holdings is that of achieving high yields of good-quality tobacco per hectare. For this, not only must
91 LEY FARMING SYSTEMS WITH REGULATED be correct way, but fertile soils must cropping be carried out in a technically than nutrients of case less the presence prepared-fertility meaning in this a cropping shifted originally within the absence of nematodes. Tobacco cleared, selected, Suitable plots were holding covered by *Miombo' forest. abandoned to secondary vegetation. then used for a number of years, and or either ten years of bush fallow, Elimination of nematodes necessitates In that are inimical to nematodes. three to four years of ley with grasses stage a as semi-permanent land use serves production conditions like these, ley systems. In large holdings that in the developrment of fully integrated Tobacco includes removal of tree stumps. operate tractors, forest clearing cultivation The use. further ploughing and cropping leaves the soil ready for because maize produces low harvests however, of maize presents a problem, 2nd a host plant for nematodes. Tobacco on tobacco soils, and it is also the within soils separately on compatible maize, therefore, must be grown too be to prove the years of tobacco holding. Green-manure crops between an therefore is of cultivation. Planted grass costly and entail a high amount following The alternation with tobacco. attractive and practical crop for rotations have been developed in consequence: and (1) tobacco-tobacco-leys (3-5 years); yeais). (3-5 (2) tobacco-maize-tobacco-leys cattle, of grasses that are not eaten by In some holdings, the leys consist other in But them is to eliminate nematodes. and the only reason for growing productive grazing. In these cases the cases the pastures are used as cattle planted fallow vegetation. The combinaley replaces a system of deliberately of ping is characterized by a number tion of tobacco cropping and cattle-i, advantages: partly of large areas of non-arable (1) The holding in any case consists grazing. land, which can be used only for steady returns, whereas the comparatively (2) Cattle husbandry produces widely. yields from tobacco cropping fluctuate payment in kind reduce the wage as use (3) The production of beef and its cost of tobacco growing. can be used only seasonally for tobacco, (4) The labour force, which is for work connected with the livestock used in the out-of-season period activities. from easy to carry out the change-over In large holdings, it is relatively systematicwith to semi.permanent farming shifting cultivation with tobacco and finally to a ley system. Large-scale ally established fallow vegetation, and the necessary capital or credit, tobacco growers can usually command major problems for smaUholders. the knowledge, both of which present
92
5.4.2. Largefarms in the Kenya highlands where the transiThe Kenya highlands are among the few tropical regions with accomplished being is systems ley tion from unregulated to regulated and grasses of growth the for relative ease. The ,climate is naturally suitable on developed first was potential for keeping grade cattle. The production smallholder the into spreading large farms, but improved practices are now farms that are typical areas. Table 5.2 contains the data for large and small of the area.
TABLE 5.2
in the Kenya Farmn-managenlent data from large and small le' farms highlands
Location (m) Altitude Rainfall (mm) Type of farming Year9 Year Number of farms Method Size of farm (ha usable land) Labour force per farm (ME) Labour force per 100 ha usable land Land use (ha) Arable Small grain Maize Pyrethrum Other arable crops Permanent crops Grassland, including leys Livestock (LU) Dairy animals Young stock Sheep Molo.Mau-Narok 2500 1250 Wheat-sheep 1962 20 Book-keeping 435 52 12 149 Nandi
2000
1500 Milk-maize 1962 9 Case studies 14.4 1-61 11 1.15
2 285 263 3!% 13% 51% 54000 31% 38% 12% 19%
9 96% 4%
Pigs
5%
Pyrethrum (q1ha- ) Maize (q ha- ) 1 Wheat/barley (q ha- ) year) and cow Milk (kg per LU100 ha grassland ha') Pyrethrum (S 1 Maize ($ ha ) Wheat/barley ($ ha- 1) Livestock products per ha grassland
16 16
1100, 113
49 18
SYSTEMS WITH REGULATED LEY FARMING Economic return ($ per holding) Gross return Pyrethrurn Grain and maize Milk Cattle Sales as % of gross return Costs of material inputs Implements Fertilizers Concentrates Income Location Wages for hired labour, including payment in kind Family income' Profit Productivity Gross return (ha- 1) 1 ) Gross return ($ ME" Value added ($ha - ) Value added ($ ME-')
Sheep
93
29 838
355
94 16 184
62 128
n.a. 2% 22%
Most farmers work elsewhere some of the time, and get an income from asource outside the farm-for the most part they have key positions in local governmcnt, local cooperatives, etc. Seasonal labourers are employed in addition to the above-mentioned permanent labour force. Young plantations of tea and coffee. o Of these, 5.8 LU are grade cattle (Guernsey. Ayrshire, Sahiwal), the remainder being improved native Zebu cattle.
d Young
plantations.
* Refers only to grade milk cows. Only from farming. Data from: MacArthur and England (1963); MacArthur, 1962-63. On the large farms depicted in Table 5.2, a third of the land is used for pyrethrum,' wheat, and barley. Pyrethrum, with gross yields of more than $250 per hectare on the one hand, and requiring a high manual labour input on the other, is the most intensive crop. In terms of area, wheat and barley are the most important, with yields of 15-20 quintals per hectare and a gross income of $100-150. Grain cropping is fully mechanized. Four to six years of cultivation with wheat, pyrethrum, fodder oats, barley, etc., are followed sometimes by uaturally established leys, but normally by planted or sown leys. There are five to ten years of ley before the land isploughed again and used for arable cropping. The grass ismainly used for woolled sheep and dairy cattle. We can see from Table 5.2 that the number of livestock units per 100 hectares of grassland is lower in the large farms. They nevertheless attain a much higher output per hectare of grassland, because of improved husbandry techniques and more economic management. A daisy-like flower whose blossom isharvested and dried. Pyrethrin, an insecticide, is extracted from the dried flowers.
94
5.4.3. Sinallholdings in the Ken*ra hhlan/s Tile returns from the holdings in Nandi District apply to a region where a traditionally pastoralist tribe has been settled since the turn of the century. In the 1950s, tile Nandi divided up the land, which had hitherto been communally owned and mainly used for grazing, into individual holdings. Since then, clearly demarcated and fenced holdings of 10-20 hectares have existed. Cattle-keeping and maize cropping form the basis of their economy. The cattle are still predominantly the indigenous Zebus. A great number of lowquality cattle are kept, which is typical of unimproved cattle husbandry in Africa. The cattle density, at 113 head per 100 hectares, is far above the optimum capacity of unfertilized and untended pastures. The Nandi prefer maize cropping to wheat cropping. The maize is largely used for home consumption. The yield is more certain than that of wheat, which may be attacked by rust. Maize can be cropped without costly machinery, and theft can be controlled-which is not always easy in large holdings. Maize cropping rotates in the grassland. It is ploughed with oxen, and cultivated, weeded, and harvested with hand-tools. In recent years, tractor ploughing has gained ground rapidly. Owing to the combined effect of the cool climate and the short days that are typical of areas on the equator, the maize grows slowly and is harvested as much as eight to ten months after planting. The yields, at 6-16 quintals per hectare, are very low. The land is used for only three maize crops, after which the plot is abandoned to natural grass. Cultivation follows again after an interval of fifteen to twenty years. Occasionally, when the kraal is shifted, maize is planted on manured land. An increasing number of Nandi farmers are becoming commercially active. More and more hybrid maize is being planted. Some farmers plant tea and sell the leaves to neighbouring tea factories. Attempts to improve animal husbandry are especially widespread and important. Guernsey, Jersey, and Ayrshire cows are bought to produce milk, which is sold through cooperative dairies. Weekly dipping of all cattle as protection against the tick that carries east coast fever is a usual practice throughout the area. The cows no longer spend the night in the kraal but on the pastures, where they are able to graze continuously, especially in the cooler morning and evening hours. Cows producing a high yield of milk are given concentrates. Artificial insemination has been introduced. The improvement of the fodder economy goes hand in hand with improvement of animal husbandry. Use of fencing and rotational use of the grazing areas are already practised by some farmers. Here and there, natural grass growth is already being replaced by the sowing of Rhodes grass or the planting of Kikuyu grass. In some cases, elephant grass is planted. Already a few farmers produce maize silage as a fodder reserve for the three-month dry season.
95
5.G.
We can see from the examples in Kenya that the tropical highlands, where soil and climate allow good fodder grasses to grow the whole year round, and where there is organized marketing, providc favourable conditions for Icy farming supporting intensive beef or dairy production. As soon as we leave the tropical altitudes and enter a warmer climate, however, ley systems meet with an increasing number of cbstacles: (1) The fodder value of tropical grasses is usually low. (2) The obstacles to the keeping of grade cattle multiply. Problems arise both in keeping the animals healthy and in feeding them. The alternation of wet and dry seasons means that a lot of attention must be paid to fodder production and conservation for the dry period. (3) Comparatively high yields of maize, manioc, sorghum, and other crops may be set against the increasing difficulties of ley cultivation and cattle husbandry. In other words, arable farming has a high relative advantage compared with Iey farming. The warmer the climate becomes, the greater is the technological jump required by a regulated Icy system. In the highlands of Kenya, at an altitude of 2000 metres, and with a rainfall of more than 1000 millimetres, we find many examples of a gradual change, even among the smallholders. But as the altitude decreases, or as the rainfall becomes less and is limited to the rainy seasons, we find scarcely any Icy systems. A number of reasons account for this situation. 5.5.1. The low fertility effect of tropicalgrass leys Tropical cultivated pastures are generally of grass alone, and frequently of only one species of grass. Their effect on the yie!d of arable crops is by no means unequivocally better than that of natural regeneration. With regard to their value to the farm economy as a whole, Webster and Wilson (1967, p. 318) have concluded: Hence it is unlikely to be worthwhile expending labour and money on planting grass unless an appreciably better return can be obtained from animal production on the Icy than from natural regeneration. Furthermore, although increased crop yields are normally obtained from a grass Icy, these increases do not usually compensate for the loss of cropping during the Icy period, so that the maintenance of fertility is,in part, achieved at the expense of a reduction in the total amount of crop products removed from the land. Clearly, alternate husbandry will only be satisfactory if crop yields during the arable break and animal production from the Icy can be raised to levels that together more than compensate for the loss of cropping during the grass break. It seems unlikely that this can be achieved without a use of fertilizers which most tropical farmers cannot at present afford, or without a much greater ability in husbandry than the majority at present possess.
96
It might be expected that the inclusion of legumes in tropical leys would improve the fertility-restoring properties of the leys. This, however, is a particularly difficult problem. With certain local exceptions, no satisfactory legumes have been found for inclusion in tropical leys. There is also insufficient information yet on the extent to which nitrogen is fixcd by pasture legumes in the tropics (Webster and Wilson 1967, pp. 298,.311). 5.5.2. Unfavourable capital-output ratio Regulated ley systems, in contrast with unregulated ley systems, require the following investments: (1) complete land clearance, i.e. the removal of trees, bush, and roots; (2) fencing; and (3) sufficient traction power-tractors or oxen-and the necessary implements for mechanical cultivation, because regulated ley farming and plough cultivation are highly complementary to each other. In addition, it is frequently necessary to inest in a number of improvements, including: (4) seeding or planting of leys; (5) purchase of improved cattle; (6) cattle pens, dips, etc.; (7) water supplies; (8) carts for the transport of manure; (9) roads; (10) drainage and the levelling of termite hills; (I I) tsetse control; and (12) land consolidation and the demarcation of fields. Some of these improvements can be carried out by the available labour force in the slack months, but as a rule they involve considerable financial outlays: ploughing the leys requires a good deal of traction power, and to use this traction power economically in a holding requires a holding size that is rarely found in smallholder agriculture in the tropics. Alkali (1967) quotes, for example, for northern Nigeria minimum sizes of 8-16 hectares for ley farms with ox-plough cultivation. In this context, Webster and Wilson (1967, p. 297) write, 'Land tenure systems under which the land is communally owned, or held by individuals on short-term, insecure leases, may make farmers uninterested in establishing leys. Fragmentation of holdings, resulting in small, scattered and unmanageable plots of land, has a like effect.' 5.5.3. Dependence on improved animal husbandry The return from animal husbandry determines to a large extent the economic practicability of grass leys. Consequently, regulated ley systems remain
97
more or less uneconomic wherever animal husbandry is organized on traditional lines, because under these conditions the farmer does not receive a sufficient return to justify any appreciable expenditure on the establishment and improvement of leys. The introduction of improved animal husbandry,
however, depends not only on better feeding and better care of the animals, but above all on the provision of better communications, abattoirs, organized markets, and prices conducive to the production of quality meat, milk, or other livestock products. 5.6. DOvoloprnent paths of regulated lay systems 5.6.1. Improved icy systems Where the transition to regulated Icy systems has been accomplished, there is, as a rule, much scope for increased production, provided that price relations are favourable. The first steps that are taken are usually the following: (a) Intensification of arable cropping is often cf primary importance. The use of new crop varieties, of mineral fertilizer and herbicides, the investment in traction power and implements, and in particular the introduction of better husbandry continually offer scope for greater production. (b) Intensified use of grassland,requiring more time and substantial investments, usually follows after a considerable time lag. Seeding or planting of the leys, intensive rotations of the grazing, and the application of mineral fertilizer are important indicators of the improved use of the leys. The most important step in Icy improvement is the inclusion of legumes wherever feasible. (c) Improt'ements in animal husbandry are usually of key importance; including disease control, artificial insemination, adequate feeding all the year round, reduction of the calving interval, proper calf-rearing, and so on. (d) Interactions between arable cropping and aninal production. Night bedding in cattle pens will result in the production of more manure, the application of which may well increase arable yields significantly. Another important aspect is the substitution of tractors for draught animals. The replacement of oxen by a tractor usually offers more scope for meat and milk production. 5.6.2. Changefrom ley systems to other types offarming A Icy system must show itself to be better than tumbledown grassland or other fallow systems in (1) having a better fertility-restoring capacity, (2) supporting more livestock production, and (3) allowing a more efficient use of the farm labour; and the combined effect of these advantages must be higher by quite a substantial margin than the costs of establishing a full leyfarming system. In many parts of the tropics these essential advantages cannot yet be obtained.
SYSTEMS WITH REGULATED LEY FARMING The fact that regulated ley systems have spread and been able to hold their own in the tropics only in exceptional cases indicates that the expansion path of tropical rain-fed farming usually leads directly from shifting or semipermanent systems to permanent farming. Only occasionally are Icy systems being intensified. A more noticeable tendency is for the area of unregulated leys in each holding to be progressively reduced, until finally permanent farming is practised. Usually, it is economically more profitable in smallholdings to expand the cropping of high-yielding cash crops, and if necessary to practice arable fodder cropping, than to intensify the ley system. Alkali (1967), for example, with reference to northern Nigeria, describes the policy of agricultural advisers as follows: 'The system of aiternate husbandry was scrapped mainly because the current design of ax-drawn implements was incapable of ploughing up leys, and also there was no proof that a system of leys was better than permanent pasture or permanent arable cropping.' The tendency for farmers to prefer permanent cropping systems to ley systems stems not least from basic technical advances. New varieties of maize, wheat, sorghum, cotton, groundnuts, etc. have made arable farming more profitable. With the increased availability and use of mineral fertilizers, the farmers are less dependent than before on the fertility-restoring properties of leys. Thanks to herbicides and mechanical weeders, they no longer need leys to save labour in weed control. The tendency to expand arable farming at the expense of the leys will probably be reversed only when the rotation valuz and the fodder value of the leys are raised decisively by inclusion of legumes, and when at the same time the price conditions begin to favour milk and meat.
98
at the expense of the fallow or ley A CONTINUAL expansion of arable farming into systcn.3 with permanent transforms semi-permanent and ley systems systems in which fallows or cultivation. This class includes those farming interpolated between the cultivation leys are only rarely, and for a short term, value exceeds 70. In contrast with of arable crops, with the result that the R this kind are normally choracterfallow or ley systems, farming systems of holding between arable land and ized by (1) a permanent division within the (2) clearly demarcated fields; grassland, which is seldom or never cultivated; crops. biannual and (3) a predominance of annual and plots are a supplementary activity rain-fcd cultivated Often, permanently to irrigation farming or tree in holdings that are predominantly devoted valley bottoms is often comrice in crops. In particular, the growing of wet cropped rain-fed plots on the permanently of bined with the cultivation of fruit trees has led to stationary slopes. The same applies where the planting in the vicinity of the tree-crop areas housing and where the shortage of land accounts are restricted, howhas led to permanent cropping. The following permanent rain-fed cropping is the ever, to those farming systems in which main activity.
systems towards permanent The evolution from shifting or semi-permanent the development of agriculture arable cropping on rain-fed land characterizes climates. In the tropics, perin densely populated regions with temperate in its occurrence. Where increasing manent cultivation is still fairly limited the intensification of land population density and cash cropping necessitate preferred: use, other possibilities are as a rule irrigated rice proves to (1) Irrigation farming is spreading; in particular, use of land in tropical be a simple and highly productive permanent of soil fertility in rain-fed climates. With the problem of conservation 99
100
SYSTEMS
farming, and as the pressure of population increases, there is an obvious tendency to concentrate production on what is topographically the lowest land, because that is where the most fertile soils occur, and often irrigation is possible in these valley bottoms. (2) The planting of perennial crops, in particular of tree crops, is increasing. By this means, arable farming, which is 'alien' to nature in the tropics, is replaced by crops whose effect on the soil is similar to that of forest or bush vegetation. Systems with permanent rain-fed arable farming are thus no more typical of tropical land use than are regulated ley systems. They are found where the following conditions apply: (1) The population density is very high and irrigation farming is either not possible or has not yet been adopted. There are numerous areas where a Malthusian process of impoverishment has compelled intensification. (2) The soils allow permanent rain-fed cropping without a drastic decline in yields. Cases of this kind are usually found where the c!imate has an extended dry period, during which leaching is not taking place. (3) An intensive fertilizer economy is practised. Most traditional types of permanent rain-fed farming in the tropics are relatively poor forms of agriculture. The main forms are outlined below. 6.2.1. Permanentcultivation in tropicalhighlands The evolution path of arable land use in densely populated tropical highlands is similar to that of moderate climates: shifting cultivation --,. unregulated ley systems - permanent arable- cropping. The decline in soil fertility that accompanies more intensive land use witiout fertilization is less pronounced than in a hot and humid cimate. Yields usually stabilize at a level sufficiently high to make permanent cultivation advantageous, even without heavy inputs of fertilizer. The main problem encountered in permanent cropping in high altitudes is not so much the constant reduction of soil fertility as the loss of soil through erosion. Permanent rain-fed cultivation with maize and beans is widely practised in the Mexican highlands and in several smallholder areas of the Andes. Potatoes, vegetables, and millets are the main crops of permanent cultivators in the Nilgiris Mountains of southern India. In parts of Ethiopia, ley systems have been replaced by permanent cropping with wheat, teff, millet, and numerous other crops. Maize, beans, manioc, and millets are the predominant crops grown by permanent cultivators in the highland areas of East Africa.
ON RAIN-FED LAND
101
6.2.2. Permanent cultivation in the African savannas Table 6.1 provides information about areas in Africa where permanent cropping is traditional and where improved techniques are applied. Systems of great economic importance are located in the East African highlands. Only a few cases are found in high-rainfall, low-altitude areas. The savannas, however, owing to higher population densities and increasing cash cropping, are already dotted with small but expanding systems withpermanent cultivation. Where the process of ensuring the food supply for a slowly expanding for population has led to the gradual spread of permanent cropping, as methods sophisticated often 6.4.2), (see example in Ukara, Lake Victoria of manuring have spread. The traditional methods of maintaining fertility, however, require a high labour input. The holdings are small and their yield in relation to the labour input is low. By contrast, in those areas where permanent cropping has developed through increased cash cropping by a local peasantry, as for example in the groundnut areas of Senegal and northern Nigeria, the situation is quite different. Incomes are higher, but soils are mined with little regard for future returns. Technical and biological innovations that can give stability to permanent cropping on soils of moderate fertility have been introduced only in exceptional cases. The more rapidly the commercialization of production has been achieved, the less farmers usually worry about conserving the soil. This applies at least until the possibilities of finding new land for cropping are exhausted. 6.2.3. Permanent cultivation in monsoon Asia Permanent rain-fed farming prevails over large parts of India. The predominance of this relatively poor form of agriculture reflects the handicap from which much Indian development suffers. The farming is usually mixed, often in combination with irrigated plots. Fodder production, manuring, are terracing, and other soil-preserving measures are applied, but they Consequentlevel. original its at normally insuflicient to maintain soil fertility ly, yields persist at a low level. Ceertz (1963, p. 29), taking as an example rain-fed farming in Java, describes this process as follows: On virgin soils a rapid decline in yield usually takes place, in the absence of fertilization, within the first two or three years, but after ten or twenty years the yield tends to remain stable more or less indefinitely. This has been borne out by experiments in various parts of tropical Asia ... On infertile soils and with inadequate fertilization the yield stabilizes at a very low level, as is the case now in Ceylon and most of South Asia, but it stabilizes. Why this should be so is not yet entirely understood. 6.2.4. Permanent cultivation in the hot, humid tropics The permanent cultivation of rain-fed land in a hot, humid climate presents some of the most troublesome problems of tropical agriculture. The
TABLE 6.1 Permanentfarming systems with soil-conserving practices in the African tropics Population
Tribe Malinke Baule Kita Dogon Bobo Gurensi Nunuma Mamprusi Losso Kabre Kam u ku , Ka n u r!, ' Mandara Chamba Bauchi, Berron Sokoto, Kano
Country Senegal, Guinea Ivory Coast Mali Dahomey Dahomey Dahomey Dahomey Ghana Togoland Togoland Nigeria Nigeria Nigeria Nigeria
Average altitude fet 1600-3000 1600 1600-3300 2000 1600 1600-3300 1600-3300 1600-3300 1600--2600 2600-3300 1600 1600-3300 1600-3300 1600-3300
density, inhabit, per sq. mile 26 26-51 26 51-129 51-129 51-129 5 1-129 77-129 129-258 568 129-181 129-258 129-258 310-516
bd crops
x x
x x x x x x x
Millet, rice, maize Yams, banana, taro Millet Millet, yams, banana Millet, yams, banana Millet, yams, banana Millet, yams, banana Millet, yams, banana x Millet, groundnut, yams Millet, yams, rice i l t beans e n Millet, Millet, yams, banana Millet Millet, groundnut, manioc
x x x x
x
x x x x x x
Batta, Mundang, Mandji, Barnum, Dama, Musgu Bana, Adamawa Kuru, Bari Konso Tigre Kips;gi, Kikuyu, Nandi, Suk, Keyu Rundi Ruanda Kiga Matengo, Makonde Kinga Sandawe, Iraque, Fipa, Turu, Gogo Mbugu, Shambala, Pare, Meru, Teita Wakara
b
Carneroons Cameroons Sudan Ethiopia Ethiopia Kenya Burundi Ruanda Uganda Tanzania Tanzania Tanzania Tanzania Tanzania
2600-5000 5000-6600 1600-3300 5000 5000-6600 5000-5600 5000-6600 5000-6600 5000-6600 3300-5000 1600-5000 2600-5000 5000-6600 400
31-9 31-59 39-55 39-47 24-39 55-71 39-55 39-55 31-59 39-47 39-55 31-47 59-79 63 Manuring.
129-258 258-387 '77-129 490 258-387 129-387 258-387 387-516 129-258 77-258 51-258 26-258 129-258 542 Stabling.
x x x x x x x x x x x x x x x x
X x x x x x x x
x x
Millet, yams, banana Millet, beans, Millet Millet, coiton, maize Millet Maize, manioc Maize, banana, sweet potatoes Banana, millet, sweet potatoes Maize, banana, beans Maize, millet, manioc Maize, millet Millet, maize, beans
89 4
x
X x X x x x
Terracing.
c Irrigation farming.
104
traditional solution is gardening,1 especially wiere the population density is very high, where labour is cheap, and where profitable markets are at hand. Permanent gardens can already be found in shifting systems where the hut change is infrequent. In permanent farming, they develop into an important branch of the holding, and one which can become dominant where there is a town in the vicinity or where there is a marked shortage of land. Thus, in densely populated Java, one-fifth of the land used for agriculture is reported as being plots that are more like gardens than fields. Examples of rain-fed permanent cropping based on vegetables, and which preserve the soil, and are economically successful, can be found in Chinese agriculture, for example in Taiwan. Here the permanence of cropping is based on an intensive fertilizing with organic and mineral fertilizers, supplemented by measures for erosion control, and careful cultivation. In other densely populated areas, for example in Java and in the West Indies, root-crop systems (manioc, sweec potatoes) have principally developed, and are associated with various tree crops. There are as yet few examples of economically promising systems of lowintensity permanent rain-fed farming in the humid tropics. Research stations provide us with an increasing number of rotations that would maintain soil fertility, but these rotations usually depend on traction power and mineral fertilizer. They include several green-manure or fodder crops, which have a low economic return. Rain-fed cultivation in the humid tropics frequently lacks a lucrative cash crop. In short, there may be technically feasible solutions to the problems of permanent cultivation, but their economic returns are as yet still marginal.
6.3.
Even though the forms of permanent rain-fed cropping in the tropics may be varied, a number of characteristics are typical of the system. 6.3.1. Spatialorganization of cropping A basic difference between permanent cultivation systems and fallow or Icy systems is the fact that the cultivator in a permanent system can no longer select a plot that is suited to the particular crop; instead, he must decide which crop will grow best on the given plot. We commonly find permanent IGarden cropping is distinguished from arable cropping by the following features, which are usually, but by no means in all cases, found simultaneously: (I) cropping those plants for personal consumption that cannot be collected nor supplied by arable farming, (2) small plots, (3) proximity to the house, (4) fencing, (5) mixed or dense planting of a great number of annual, semi-permanent, and perennial crops, (6) a high intensity of land use, (7) land cultivation several times a year, (8) permanence of cultivation, and (9) cultivation with hand implements.
ON RAIN-FED LAND
105
land-use systems with distinct spatial differentiations of cropping, and with rotations including a great number of crops. There are three basic types of spatial organization: (a) Organization of cropping according to different soils within a catena. Where the types of soils vary within a catena, the differentiation of cropping mainly arises from the planting of each type of soil with the most suitable crop or mixture of crops. A case in point is land use in a valley on Ukara, Lake Victoria (Fig. 6.1). Rice, sweet potatoes, vegetables, and sorghum predominate in the valley bottoms, manioc (cassava) and Leguminosae are the 'pediment crops' and the slopes of the 'lnselberg' serve as grazing. (b) Organizationof cropping in concentric belts. The tendency to cropping in concentric belts is caused by stationary housing. The older the farmyard and the greater the difficulties of transporting manure, the more markedly does cropping change as the distance from the farmstead increases. In the farmstead or village we almost always find fruit trees, papayas, bananas, etc. Usually there is an adjoining belt of heavily fertilized garden. As the distance increases, the fertilizer input declines and the more frequent are the fallows. (c) Organizationofcropping according to the micro-climate. The differences in the micro-climate (the amount of rain, sunlight, land gradiert, and danger of wind damage or flooding) help to diversify cropping. In the valley bottoms and flooded plains we frequently find some rice, occasionally surrounded by embankments, indicating the tendency to supplement permanent rain-fed farming by primitive types of irrigation (see 7.1). Such first steps towards a differentiation of cropping within the holding may be found in shifting systems (see 3.3.1) and are common in semipermanent systems (see 4.3.1). In permanent systems, however, they are considerably more pronounced because of the stationary nature of the housing. 6.3.2. Croppingprinciples A further feature of permanent cultivation systems is the variety of forms of mixed cropping, of relay-planting, and of rotations that occur. (a) Mixed cropping and relay-planting. Provided land is scarce and labour ample, mixed cropping is as common in permanent systems as in shifting and semi-permanent systems (see 3.3.2 (a)). Relay-planting is the usual practice in intensively cropped areas with extended rainy seasons, or where several rainy seasons overlap (see 2.3.2 (b)). beans, This is exemplified in Kerya by the Kikuyu practice of cropping maize, end of the at begins, season rainy the When peas. pigeon and sweet potatoes, broadcast peas and maize the plant womea March or the beginning of April, the on the plot, which has been prepared with the hoe by the men. When the seed has germinated, the women go over the field again and plant two different sorts of bean, an early-maturing and a late-maturing variety, among the growing maize and
1228 1208
"
E.S.E.
W.N.W.
8, 6_
Chala
40
20-
1148
S
Manioc ,illet I a
1
Sm
0
A I r m
Inselberg"
'Upper pediment"
Settlement
"tnselherg" 2o00(m)
FiG. 6.1. Profile of a valley on Ukara Island, Lake Victoria, showing land use within a catena (from Ludwig 1967, p. 100).
107 SYSTEMS WITH PERMANENT CULTIVATION pigeon peas. When the bean seed has sprouted, a third planting takes place. The same time, sweet-potato seedlings are planted in the growing stands of maize, peas, and beans. The early-maturing beans produce the first yield from tile middle of June. Maize and late-maturing beans are harvested from the middle of August. Sweet potatoes and pigeon peas are harvested from the plot as required from September onwards. In regions with several overlapping rainy seasons, relay-planting can become continuous cropping, i.e. when one crop is harvested the plot is already being used by the next interplanted crop. (b) Crop rotations. Whereas it is often difficult in semi-permanent systems to recognize a systematic sequence of crops (it is for this reason that de Schlippe (1956) speaks of 'pseudo-rotations'), in permanent cultivation systems there are generally quite definite rotations. The rotation patterns of tropical permanent farming systems are of confusing variety. The great number of forms they take may be classified according to the following criteria: (1) In regions with distinct and short rainy seasons, one crop a year is sown usually, occasionally supplemented by a catch crop. In the intervening dry period, the plot lies fallow. These are, therefore, systems with seasonal cropping. In contrast, continuous cropping (where planting takes place before or immediately after the harvest) in rain-fed farming can be practised only where the rainfall is fairly high and evenly distributed. The fertilizer and labour requirements per hectare of continuous cropping are naturally considerably higher than for seasonal cropping. (2) A distinction must be made between annual, biennial, triennial, and other rotations of different lengths. Annual rotations with two or more crops a year may be found in regions with extended rainy seasons or continuous cultivation. Biennial rotations in these rainfall areas often comprise three crops, one of which, for example sugar-cane or manioc, occupies the plot for a particularly long time. In general, rotations in areas with seasonal cropping occupy more years than in areas with continuous cultivation. (3) The length of the vegetation cycle of the crops cultivated is an important structural element of rotations. Arable farming in the tropics is in this respect much more varied than in temperate climates. The rotation may be composed of annual short-term crops (e.g. beans, cabbages, or sweet potatoes), annual long-term crops (e.g. cotton or maize), and biennial crops (e.g. manioc, sugar-cane). The cropping index (see Table 6.3 on p. 120) (the cropped area as a percentage of the arable land) is consequently of only limited application as a measure of intensity in permanent farming systems. The input in the case of a biennial arable crop that occupies the
field, which has produced a great deal of weed in the meantime, is weeded. At the
108
plot for sixteen months is sometirves higher than that of several annual short-term crops, which can be cultivated in the same time. (4) Apart from this, the farm-economic aspects of crop rotations depend on whether soil-preserving or soil-exhausting crops are cultivated. In general, fodder crops, especially legumes, are considered to have a beneficial effect on the subsequent crops, while crops like maize or cotton are considered to put a strain on soil fertility (5) The rotations, and the crop elements of the rotation, may be arranged according to crop yields and input requirement. When millet is displaced by maize in a rotation, or maize by a root crop like manioc, production measured in kilocalories per hectare increases considerably, but at the same time labour and fertilizer requirements usually increase also. 6.3.3. Characteristics of the fertilizer economy' The methods of fertilizing in tropical permanent rain-fed farming are as varied as the rotations. In most cases, the fertilization is insufficient, and we find soil mining, and stagnation at a low yield per year. Generally speaking, however, permanent cultivators on rain-fed land practise some methods of fertilization, and the more so where the original fertility was low and the longer ago the change occurred from fallow systems to permanent cropping. The principal fertilizing methods of semi-permanent cultivation or of unregulated ley systems are scarcely practicable in these cases, since permanent cropping requires stationary housing. The moving of the hut sites, and the consequent creation of fertile gardens, can take place only within a restr;cted area, and is hardly equivalent to the 'human folding' which is so common among semi-permanent cultivators. The cattle numbers are usually insufficient for folding systems, unless extensive grazing areas are available. Also, permanent cropping has to forgo the erosion protection that ley systems provide in the years between cultivation. Permanent cultivators who do practise fertilization and erosion control depend on other methods, with a higher labour input. Various traditional methods of erosion control have been employed, from ridge and mound cultivation to terracing and the use of baskets to carry back the soil that has been washed away. In particular, stone terraces in traditional, rain-fed farming systems are an indication of a long-term, acute land shortage, which has helped to make erosion control part of the cultural heritage of the local people. Although these practices are traditionally known, they are rarely employed in farming systems where cash cropping has been introduced and where land shortage is a recent phenomenon. In many of these situations, particularly in the drier savannas, gullies increase rapidly in number and size, soil conservation usually being neglected as cash cropping and incomes per head increase, mainly because of the unfavourable short-term input-output relationship of the labour invested. The
ON RAIN-FED LAND
109
way out of this undesirable situation probaktly does not lie in a return to traditional agricultural methods, but in additional cash cropping, which, by changing the economic setting, can make s:ail conservation economically worth while. The tractor-drawn plough becom es very important in this connection, because it enables the farmer to carry out soii-con.;erva.tion measures like contour ploughing, mechanized idging,, anc mechanized terracing, which cannot be done with an ox pleagh. tiiisr economy in perOf special interest is the dcve'ipment of the manent rain-fed farming. The primary task i's the rep.c.ment of nutrients and organic matter. From '!,.e standpoint of the evolution of farming systems, a number of stages may be distinguished: (1) Manuring begins as a rule with the collection and transport of household refuse. (2) Next comes the application of animal manure, which is sometimes increased by stabling and the provision of bedding. (3) Cultivation of green-manure crops is an indication of a more advanced fertilizer economy. (4) Processing of household refuse, harvest residues, and farmstead earth in compost is characteristic of a higher level of intensity. (5) The use of night soil s tr.,.ditional with Chinese cultivators. It is rare in Africa and only practised by -ome tribes with a long-term land shortage (in Ethiopia (Nowak 1954) and Mali (Dumont 1966, p. 90;). Increase in the intensity of cronp.in.,-, entails the irnpOrt of nutrients and organic matter. Traditionally, this cou ,, 4'ok. one of the following forms: (1) Areas outside the holding are gr..ed, nd '~'m manure collects in the farmyard, mainly at night. In tijs way/ nutr~ints are concentrated on the permanently cropped land. This is a prirziplp: that is practised generally. (2) Nutrients are imported in fuel. The se,.rcn for fuel has stripped exten6 . as1 (needles, leaves, sive bush or forest land near densely popu L a'nu twigs). The ashes serve as fertilizer, either directly or as a component of compost. (3) An indication of a special lack of nutrients is the collection of leaves and branches as green manure, which is applied directly to the land (see
6.4.2).
(4) In traditional Chinese agriculture, oil-cake, fish, and other organic materials were bought as manures. A general feature of traditional forms of intensive fertilization is the high labour input; this is the case both in preparation oF compost and in collection of animal manure, and lies especially in the difficulties of transport that are part and parcel of most forms of organic manuring. Whereas in shifting and semi-permanent systems the fertilizer is produced where it is used, i.e. it
110
scarcely needs to be transported, the permanent cultivator with stationary housing must transport the material. Consequently, we find the tendency to concentrate the fertilizer on the gardens near the farmyard, particularly where the lack of transport is acute and there are no carts, draught animals, or field tracks. Bottlenecks in transport are the reason why old Chinese towns are often surrounded by a belt of black soils, which are enriched with nutrients and humus. Elsewhere, the saying that 'those fields are most fertile which can hear the cock crowing' reflects the same situation. A further feature of the fertilizer economy in densely populated areas is the burning of straw and cow manure. Since there is no fallow vegetation to provide fuel, as is the case in shifting and semi-permanent systems, straw and cow manure are frequently the only available and economic types of traditional fuel. Technical progress has scarcely altered the production conditions in any branch of agriculture so much as in the fertilizer economy of holdings with permanent cropping: (1) Mineral fertilizer is of primary importance; compared with organic fertilizers it proves to be labour saving because of its high nutrient concentration and the relative ease with which it can be transported. (2) New crops and cultivation methods for green-mnure crops are available. The use of tractors can provide more scope for increased green manuring, since the work is carried out more quickly. (3) Where intensive livestock-keeping is practised, the fertilizer cycle is extended by the import of nutrients as purchased concentrates. (4) Progress in transport vehicles like ox-carts, tractors with trailers, or lorries and the development towards a comprehensive road network facilitate the transport of manure, night soil, etc. Thus we may say that agricultural innovations have drastically changed the fertility problem h) permanent rain-fed farming in the tropics. Technical solutions are available. Unfortunately, however, the price relations are often too unfavourable to make intensive fertilizing economic at the present stage. of animal husbandry 6.3.4. Characieristics Traditional faming on permanently cropped, rain-fed land in the tropics is usually cornbimcd with small herds per family, but high cattle densities per 100 liectarcs. The role of livestock in these conditions differs from that of stock in semi-permanent cultivation or ley farming. The aim is not only to produce milk and meat, but increasingly to provide traction power and manure. The hoarding of the cattle is not important, because the available grazing is limited, and also because permanent cropping is usually coupled with private land rights, and the acquisition of land is an alternative way of gaining status and security.
ON RAIN-FED LAND
III
The fodder basis primarily determines which forms of animal husbandry predominate in permanent cropping systems, ed to what extent they are carried on. In this connection we can distinguish two principal forms. (a) Animal husbandry in systems with extensive grazing areas. Where extensive areas are still available, cattle husbandry is mostly organized on lines similar to those in semi-permanent systems. The animals are communally grazed and in the charge of herdsmen by day, and they return to the farm at night. Emphasis is put on cows, female calves, and heifers. The husbandry practices are more or less the same as those of semi-permanent cultivators, but there are fundamental differences in the fodder economy, which is based increasingly on the productivity of arable farming, e.g. on the seasonal distribution of grazing between the arable land and the more distant grazing
areas.
A typical example of animal husbandry in an African farming system which passed recently from semi-permanent to permanent farming issupplied by Guillard (1965) in his monograph about the village of Golonpoui in northern Cameroon (see Fig. 6.2). At the beginning of the rainy period in June or July, the animals graze on the arable land that is not yet cultivated or is lying fallow. As cropping progresses, the animals are concentrated on the fallows of the semi-permarently used land, which are 2-3 kilometres from the village. In October, millet straw from the permanent gardens provides additional fodder. At the beginning of the dry season, the animals are driven to bush grazing further afield. This is interrupted in January and February, when harvest residues are eaten. The animals spend the rest of the dry season, from February to May, in the more distant grazing areas. In addition, the leaves and fruits of odd acacias (Fatdheribaalbida) scattered in the arable land are used. (b) Animal husbandry in systems without extensive grazingareas.As arable cultivation spreads at the expense of the grazing area, there is a tendency for farmers to reduce the number of animals kept, or to hire out the cattle to nomads who are hardly ever near the village. Roadside grazing, seasonal fallows, and harvest residues constitute the remaining fodder basis. The lower limit to the number of cattle kept is determined by the number of traction animals required. Hoe cultivators may come to have no cattle at all. With the reduction of the grazing areas, the rearing of goats, pigs, and fowls increases in relative importance. In requiring less grazing land and providing smaller units for sale or consumption, this kind of stock is better suited to the economic condition of smallholdings. The reduction in the importance of livestock that occurs as land becomes increasingly scarce can be halted only by the introduction of fodder crops, which are practically unknown in fallow systems. The need for traction animals is usually the first cause for a step in this direction, since the feeding of traction animals can scarcely be avoided. Where subsections of the holding can be irrigated, they are in many cases used chiefly to supply the traction animals with fodder. Beyond this, traditional farming systems on permanently
CULTIVATION SYSTEMS WITH PERMANENT volume a position to support a substantial cropped rain-fed land are rarely in importalthough the livestock economy, of fodder production. In most cases, of productivity (see also 6.6.1). level low a ant in numbers, stagnates at 112
December Jcnuary
February November
-.
; April
aMa
Septmber
July
June
Village
.
seasons in Golonpoui, the village land during the different 305). FiG. 6.2. Distribution of cattle on p. 1965, Guillard northern Cameroon (from
economy 6.3.5. Characteristics of the labour rely mostly on hoe cultivation, although African permanent cultivators still hoe the compelling technical reasons for forest and bush fallows, which are size farm because usually remains farming, no longer exist. Hoe cultivation This livestock. that there is no room for in relation to family size is so small like the Kisii and Kikuyu Districts is true for densely populated highlands
ON RAIN-FED LAND
113
in Kenya. In Asia and Latin America, however, we find a different and varied
tool system, which involves plough cultivation, carts, and animal traction. The plough makes it possible to cultivate larger areas and to improve the timing of cultivation. On the other hand, disadvantages are (1) the fodder requirement of the draught animals, (2) the poorer quality of animal ploughing compared with hoe cultivation, and (3) the increased danger of erosion. Communal work, which we often find in shifting systems and still sometimes
S e cj son
9pi
4)
Fia. 6.3. Annual agricultural labour cycle ina systcm of' permanent rain-red farming inthe West Nimar District, central India (Crom Leshnik 1966, p.61).
in semi-permanent systems, scarcely plays any part at all, and this is a further feature of permanent cultivation systems. Similarly, the tendency disappears for the land belonging to a household to be divided up into smaller units cultivated by members of the family for themselves. In plough culture, wve usually see centrally organized holdings, where the farmer directs the family labour and employs hired labour if need be. (a) Seasonality oflabour demand. With the transition to permanent cultivation, the seasonal concentration of labour tends to become even more acute
SYSTEMS WITH PERMANENT CULTIVATION not than in fallow systems. The work is concentrated in the rainy season, the least because draught animals and hoe cultivators are unable to cultivate only found is force labour te of use regular fairly soil in the dry season. A in a few favourable regions where double cropping can be practised. 114 6.3). The situation in the West Nimar district in India illustrates this (see Fig. in grown usually is wheat and crops, chief the are Cotton, wheat, and sorghum winter (rabi season) on a plot which is then left fallow. The main summer crops two (kharif season) are sorghum and cotton. The distribution of cropping over is there when time The labour. of use balanced relatively a about seasons brings little labour demand is limited to February-May. In most cases with permanent rain-fed cropping, however, we find only one pronounced rainy season, with little scope for multiple cropping. Consequently, labour demand is decidedly seasonal. High population densities and a high degree of underemployment on the one hand, and insufficient labour at the beginning of the rainy season on the other are probably not as marked in any other land-use system as in permanent rain-fed cropping. Figure 6.4 illustrates the situation in an intensive arable farming area in northern Cameroon. The total available time is recorded under several headings. The distrithe bution of field work is highly seasonal for both the man and his wife. Most of of kinds various with filled is work field for time between the high labour demands work around the hut. Sickness takes only a small percentage of the total time. It should not be overlooked, however, that on two occasions both were sick together, It and once this took place during a time of high labour demand (October 1957). full the to lead together may appear from Fig. 6.4 that field and homestead work in employment of m.-n and wife. It is, however, difficult to present working time in measured was here (work differs day working the of length The days. of terms labour of peaks during higher much is work half-daily units), and the intensity of demand than in slack seasons. It is not only the workers who are underemployed. Wher plough cultivathe tion exists, the traction animals too are only seasonally employed. Yet since animals, draufght of number seasonal nature of work requires a large on every cultivator wants to plough his land at the right timc and sets value for reason important one having his own animals. Lack of draught power is late and inefficient cultivation of fields. Coj-.'*,,:ntly, a considerable part of the arable land is used to provide draught animals with fodder, while they work at full capacity for short periods only. This is particularly true in India. (b) Low productivity,of labour. Table 6.2 provides some information about typical labour input in permanently cropped areas. This is, as a rule, lower than in fallow systems because of the reduced input in land preparation. per Particularly where plough cultivation is practised, labour requirements to insufficient is information available hectare are significantly lower. The variation about conclusions general relate labour input to yields and to derive of labour productivity in terms of grain-equivalents per hour of work with increasing permanency of cropping. However, several case studies indicate
ON RAIN-FED LAND
115
that, with increasing permanency of cropping, output per man-hour and per man-equivalent available is likely to decrease (Boserup 1965; Rotenhan 1966; Ludwig 1967). It is for this reason that the husbandry patterns of
Man 30.
20
10-
0 D
1956
1957
20-
AO
OF
1955
1956
1957
12
Field
work
fn
Work around
the hut
1T
Not
working
Sickness
FIG. 6.4. Seasonal distribution of work of a hoe-cultivation holding with permanent rainfed farming at Golonpoui, northern Cameroon (from Guillard 1965, p. 343).
permanent cultivation in Africa are described by Gourou (1966) as 'siege methods', by which he implies that they came into existence because the tribes were surrounded by hostile neighbours, which prevented the extension of farming in fallow systems into expansion areas. The African peasants-the Wakara (see 6.4.2) being a case in point-are apparently quite aware of the
TAnLE 6.2
farm-management surveysL
Manioc Madagascar (Tananarive) 1963 Hoe/spade 152 70 76 148 446
256 62 356
8 86 64 224
16 32 56 136
'Without threshing or preparation for sale. b Summer crop without additional cultivation. Sandy soils; no cultivation before sowing. Sources: Millet, Bureau pour le d6veloppement de Ia production agricole (1963, p. 236); sorghum (India), Boserup (1965, p. 40); sorghum (Cimeroon), Guillard (1965), see also Bureau pour lc d6veloppenment de la production agricole Agriculture (1965, p. 38); manioc (Tananarive), Bureau pour Ic d6veloppement de (1965, p. 318); sweet potatoes, Taiwan Department of ]a production agricole (1965, p. 228).
ON RAIN-FED LAND
117
ict that under their system of permanent cultivation they produce yields ant are higher per hectare than those obtained in more extensive systems, iut are lower per man-hour. Wherever conditions permit them to leave their :rowded home area, because new land is opened up for settlement, they mmediately turn to fallow systems. Four case studies The following examples give an idea of the various forms taken by permanent rain-fed cropping. The first example from Kenya shows permanent cropping on fertile soils. The remarkable efforts by hoe cultivators in Ukara are an indication of how African farmers in a siege situation have developed the same methods of soil conservation as have Chinese farmers, but are nevertheless only delaying and not stopping the decline in soil fertility. Land use in Senegal is a typical case of soil mining, except in the mixedfarming system of the Serere. The farming system in the Bombay Deccan shows the final stage in the Malthusian process: stagnation on a very low level of yield and productivity. 6.4. 6.4.1. Maize-millet holdings in Kakamega District, Kenyva The Kakamega District of Kenya is very densely populated: there are 500-600 people supported by agriculture per square kilometre. The stage has been reached where the customary fallow period of several years has been almost totally displaced. The basis of permanent rain-fed farming, in some cases with two crops annually, is a deep fertile soil and a good rainfall pattern. One family normally owns 1-3 hectares. For generations; maize, sorghum, finger" millet, and beans have been the main crops. Both hoe cultivation and ox ploughing are practised. A rotation typical in the 1950s runs as follows: Maize mixed with beans (I) Spring (long rainy season) Autumn (short rainy season) Partly maize, partly beans, grown separately Maize mixed with beans (2) Spring Fallow used as pasture Autumn Sorghum (3) Spring Sorghum (self-sown) Autumn Finger millet (4) Spring Fallow used as pasture Autumn new hybrid varieties, has come to occupy a the In recent years maize, especially more important position in the rotation. Afallow period of several years is retained only in the border areas of the densely populated region, or in the few holdings with larger areas of land. Behind the huts are banana groves, consisting of early-maturing and late-maturing varieties for both cooking and eating. The women tend a kitchen garden. In the spring, peas and beans are planted among the bananas, while in the autumn, a plot is sown with vegetables. Two or three Zebu cattle and a few sheep and hens are kept. The fodder
SYSTEMS WITH PERMANENT CULTIVATION basis consists of grass in the spacious homestead, roadside grazing, harvest residues, fallow, and communal pasture. The farmers of the Kakamega District have been cultivating for some decades without fallow or fertilizer. Until a few years ago, the available dung was used for fuel, together with straw and weeds, and the concentration of nutrients in the kraals was at best utilized for a banana grove. The intensive campaign on the part of the Agricultural Advisory Service has in the meantime caused a number of farmers to collect and use animal manure. Where maize-recently also hybrid maize--is grown as a cash crop, mineral fertilizer has been adopted. 6.4.2. Millet holdings in Ukara, Lake Victoria The Wakara, who live tightly packed (500 persons per square kilometre) on a little island in Lake Victoria, 1000 metres above sea level, similarly practise permanent rain-fed farming. The soils are derived from weathered granite, with little natural fertility. The need to guarantee the food for a dense population on these soils has caused the Wakara to develop and apply the highly refined practices of a soil-conserving agriculture, although no influence from immigrant people is recognizable. The average Wakara family has only I hectare of arable land at its disposal, and this is cultivated by hoe. The land is cultivated chiefly according to a three-year rotation: () Summer season Bulrush millet, with manure and interplanted legumes for green manure (Crotalaria striata) Winter season Growth of the green manure (2) Summer season Bulrush millet, with green manure dug in; groundnuts planted among the ripening millet Winter season Growth of groundnuts (3) Summer season Bulrush millet, with manure Winter season Sorghum or manioc (in the case of manioc, a fourth year of manioc follows) In addition, small patches of irrigated rice are cultivated in the valley bottoms, and hill land in the interior of the island is grazed. The crucial problem in the Wakara holdings is that of acquiring manure, which can be done only by the production of fodder. The two or three cattle and the one or two goats and sheep that are kept on the average holding are fed with harvest residues, leaves of growing sweet potatoes and manioc, weeds collected while hoeing, fodder grasses planted and watered on the island shores, and ratoon rice, which grows on rice plots at the beginnings of the dry season. There are also some grazing areas. Sometimes the animals are tethered, in order to economize on the available grazing. The animals occupy one half of their hut and stand in deep litter. Bedding consists of the remains from harvesting and fodder, leaves and branches of trees and bushes, old roofing grass, and any other suitable material that becomes available. The household ash is applied to the nursery beds for rice. As well as using green manure, the Wakara use leaf manure; leaves from trees that are not suitable as fodder are carried to the field and dug into the soil.
118
ON RAIN-FED LAND
119
be remarkable, it is Even though the agriculture of the Wakara may is not motivated cultivation .rgely a relic of a poor,.r past. Their careful the livelihood guarantee to much by the desire to increase their income as than the harder work f a growing number of people. The Wakara obviously migrate they When ieighbouring tribes, but they are nevertheless poorer. of the agriculture o the mainland, they change to the semi-permanent disproportiona Wasukuma, which, considering the labour input, provides isolated example. The ately better income. The Wakara's land use is an on the island, where population in the surrounding areas regard staying merely to cover their people are obliged to work from morning till evening to use to manure their needs, as stupid, and the methods that they are obliged land as dirty. on moderately The Wakara's example shows that, in traditional conditions before situation emergency fertile tropical soils, there must be a special essential are that techniques smallholders will adopt the labour-demanding features of permanent rain-fed farming with soil conservation. 6.4.3. Groundnut-millet holdings in Senegal permanent rain-fNd expanding population has entailed a gradual extension of more or less China, in scale large a on farming, as for example in Ukara, or guarantees this and spread, have intensive methods of traditional fertilizing labour high a however, require, permanent cropping. These methods do Where done. work of input, and the yield is low in relation to the amount in increase an of basis permanent rain-fed farming has developed on the different.
is usually quite market production by the local people, the situation the less do the farmers The more rapidly production is commercialized, worry usually about conserving the soil. supply of a slowly Where the problem of ensuring the subsistence-food
holdings of Senegal (see An example of this occurs in the groundnut-millet 1850, and it has accelerabout began Table 6.3). The commercialization of the area crops groundnuts and millet ated in the last forty years. Rotations with the two three to five years of available, predominate. Where there is still enough land regions, however, many In duration. same the of cropping are followed by a fallow fallow period of a is then only and for seven to twelve years, cropping is practised is no fallow period at all. three to four years introduced, and in most cases there semi-permanent to perIncreasing numbers of farmers are thus changing fromabsent. Groundnuts and or short manent land use, in which fallows are either cultivation with either hoe millets are mostly grown on the flat, without previous isby
clearing before planting or plough. The traditional method of superficial hand with a seeder, giving row planted increasingly are far the least costly. Groundnuts of mineral fertilizer application cultivation. The use of improved varieties and the The quality of practices. standard become to groundnuts as well as to millets have has been fertilizer when the first two weedings conditions the yield, especiallyused. widely applied. Horse-drawn or ox-drawn weeders are
TABLE 6.3 Farm-management datafrom maize-nillet holdings, Kakamega District, Kenya; millet holdings on Ukara Island, Lake Victoria; and groundnutmillet holdings in Senegal
Location
Year Rainfall (mm) Type of farming Method Number of holdings People per household Labour force (ME) Size of holding (ha)a Garden, homestead grazing (iJa) Pasture (ha) Available for cropping (ha) Millet and sorghum (ha) Maize Groundnuts (ha) Manioc and sweet potatoes (ha) Rice (ha) Other crops (ha) Total cropping (ha) Cropping index* Livestock per holding cattle (no.) sheep/goats (no.) hens (no.) Yields Millet (q ha- 1) Maize (q ha-)l Groundnuts (q ha- 1) Millet (S ha- 1) Maize ($ ha- 1)' Groundnuts ($ ha- 1) Economic return (per holding) Gross return Expenditure for purchased material inputs Income
Less wages
Nyanza
1965 1600 Maize-millet Case study 1 7 3.0 2.2 0.3 Communal 1.9 0.9 b 13 c 0.2 -0 0.2 2.6 137% 2-3 2 7 10 20 55 130 280 13h 267
-
Ukara 1965 1600 Millet Survey 25 7 2.3 2.3 0.1 0.9 13 0.8 0ld 0.3 0.1 1.4 107% 3 2 7 10 44 182 3b 179
-
Senegal
1954 600 Groundnut-millet Case study 7 7 3.0 8.0 0.1 Communal 7.9 3.0 3'0 -
Family income Degree of commercialization (sales as % of total return) Productivity Gross return ($ ha-1)" Gross return ($ ME- 1 ) Income ($ ha" ) Income ($ ME-')
ON RAIN-FED LAND
121
nuts are grown as of continual impoverishis acquired at the expense by some decades high family income, but it exhausted been has fertility where formerly areas ment of the soils. Areas whose Large ones are cleared. farmers, who The of cultivation are abandoned, and new today. anything yield scarcely not conobviously widespread cropping was practised are from groundnuts, research the of strive to earn as much cash as possible report the point, Referring to this land-use cerned about loss of soil fertility. 'further traditional farming within the station at Bambey stated that 1954). (Tourte Senegal' of system must lead to the ruin general in Senegal is an exception to the The land-use system of the Screre groundnuts biannual rotation of millet and practice. This tribe practises a on the soil. appears to have no harmful effect with almost no fallow, but this of cropping and animal husbandry. The explanation lies in the integration not more than 600 millimetres of rain The land of the Serere. which receives per 100 hectares of total area, half of cattle grazed. annually, supports 60-70 head half more or less permanently other proof which is cropped and the also is land bare systematic folding. The tree The arable land is fertilized by leguminous a by the next sowing protected between harvest time and thus season, into leaf during the dry rainy (Faidherbia albida), which comes is scarce. It loses its leaves in the The viding valuable fodder when grazing interfere with farming operations. a season, and does not therefore resembles and larger Faidherbia albida, thus is countryside, dotted with smaller 1966, p. 121). Permanent cropping from widely spaced orchard (Dumont and a permanent fodder supply grazing permanent with associated trees in the dry season. Bombay Deccan 6.4.4. Millet-wheat holdings in the fertilizing Wakara try, through various Whereas, on the one hand, the the other on and in permanent farming, methods, to maintain soil fertility
Ukara including individually owned pastures. In Nyanza and Senegal without pasture: in
b
the groundworker, and because Because of the relatively large area of land perin attain a compar a major cash crop, farmers Senegal
hectares are planted in the autumn. in spring, and 0.3 hectares pure hectare maize-bean mixture is planted 1[0 Of which stand of maize in autumn. for cropping. d Bambara groundnuts. a percentage of the area available Cropping index: area cropped as Plus a bean yield of 3quintals per hectare. together. ' Maize and beans hUnconfirmed estimate. seeds. Ukara and in Principally purchase of groundnut from work outside the holding. In Most families in Ny?.nza get an income income. households have an outside Senegal only a small number of forcropping. available area In relation to the 206); Senegal. (1965); Ukara, Ludwig (1967, p. Data from: Nyanza, personal surveys Of which
O2
Tourte (1954).
SYSTEMS WITH PERMANENT CULTIVATION can observe in Senegal an obvious decline in soil fertility, in large India the yields in the course of several centuries of permanent have stabilized at a very low level, especially where there is no irrigathe soils are naturally poor. This is illustrated by the millet-wheat of the Bombay Deccan, which is known as a famine area.
A typical holding of the area comprises 6-7 hectares of arable land. 10 per cent of the land can be irrigated from time to time with the nelp of wells or tanks. In summer (kharif season), millet and sorghum are gro,.n, and in winter (rabi) wheat. Supplementing this, we find small irrigated plots of sugar-cane and vegetables. The cropping area amounts in all to 107 per cent of the arable land (see Table 6.4). Cultivation is carried out with a wooden plough with an iron spike. Seed-harrows and hand drill machines are found here and there. Low yields per hectare lead to the low gross return per farm of only $113, or $17 per hectare, at price levels of 1957. There remains a family income of only $82, or $30 per available worke", which is far less than in the African Systems illustrated in Table 6.3. Only about half of the labour capacity available is employed in work on the holding. The main object of production is to ensure food for the family and draught animals. Admittedly, the millet-wheat holdings reveal some featurcs that indicate an advanced level of agricultural technique. Animal manure and compost are taken by ox-cart to the fields. Grass and bedding are collected from waste land and become part of the animal manure, so that nutrients from land outside the holding are imported onto the arable land, particularly the irrigated plots. Grazing is stipplemented by arable fodder, and some hay is produced. But all this is not enough: the soils are poor and the rainfall uncertain; the yields of 2-4 quintals per hectare of millet and 3-4 quintals per hectare of wheat are lower than the yields usually found in shifting cultivation. The holdings are burdened by having to keep work animals. Draught oxen consume about one-third of the gross production, and they are only used to half their capacity. It is, however, essential to keep work animals to guarantee a livelihood, because large areas must be cultivated and this can be done only with the plough. The need to keep draught animals increases the need to keep cows, the main point of which is not so much to supply milk and manure-meat is in any case not eaten-but to produce bull calves. Slaughter of cattle, even of old and useless animals, is barred in the area both by traditional value systems and by legal prohibition.' The holding suffers, therefore, not only from the immediate burden of draught animals, but also from the burden of keeping several cows whose principal function is to continue the supply of draught animals, without being used as draught animals themselves.
6.5.
The scope for increasing production within traditional agricultural methods diminishes as land reserves are exhausted and yields per hectare stagnate at a low level. Permanent farming carried out on impoverished soils may well be I Section 4, Article 48 of the Indian Constitution asserts that the states should 'take steps for preserving and improving the breeds, and prohibiting the slaughter of cows, calves and other milk and draught cattle'.
7 Persons per household 2.8 Labour force (ME) Size of holding 6.63 Arabic land area (ha) Summer cropping (ha) 2.63 Millet 0.93 Sorghum 0.35 Chick-peas 0.28 (with some irrigation) Groundnuts 0.99 Other grain/legumes 0.35 Other crops Winter cropping (ha) 1"35 Wheat/safflower Both seasons (ha) 0.07 (irrigated) Sugar-cane 0.14 (irrigated) Vegetables 7.09 Toal 107% Cropping index Livestock per holding 3 Draught animals (no.) 2-3 Milk cows (no.) 1-2 Goats/sheep (no.) 867 Farm capital($ per holding) 74% Land 4% Implements 17 % Livestock 4% Farm buildings Yields 3"9 Wheat (q ha-*) 2.6 Millet (q ha ') 2'8 Chick-peas (q ha- ') I0 Milk (kg per cow and year) 30 Wheat (S ha- 1) ) 17 Millet ($ ha-' 15 Chick-peas ($ ha- 1) holding) per (S return Economic 113 Gross return 18 Expenditure for purchased material inputs 13 Tax, etc. 82 Family income Productivity 17 Gross return ($ ha-') 40 Gross return ($ ME-') 12 Value added (S ha-') 30 Value added ($ MF-') 49% Employment of available labour capacity At price levels of 1957. The all-India index number of wholesale prices of agricultural commodities stood at 109 in January 1957, and at 211 in January 1967 (base: 1952-3 = 100). Indian agriculturein brief. p. 154. 9th edn, Delhi (1968). Datafrom: Driver and Deasi (1957).
124
considered as a final stage in the land development that begins with shifting cultivation, leads to semi-permanent farming yielding somewhat higher incomes per hour of work, and ends with a low-level equilibrium in permanent rain-fed farming. The growing population's increasing subsistence demands and the extension of cash cropping, in order to cover increasing cash demands, are the driving forces behind changes in farming systems. Three overlapping phases may be distinguished in respect of these processes: (1) The first step consists usually of extending a given farming system by establishing new smallholdings in expansion areas, wherever they are available. (2) When there are no more expansion areas available, or when the distance between the expansion areas and the original settlement large, intra-farm land reclamation acquires importance, area becomes too and cultivation is extended at the expense of the fallow. The transition from shifting cultivation to semi-permanent and permanent farming occurs slowly but surely. (3) Cropping patterns change at the same time. As land use becomes more permanent, the relative suitability of some crops diminishes, while that of others increases. Thus a change in the type of crop ensues. The cultivation of such crops as manioc and sweet potatoes gains ground, since they yield high and certain calorie returns per hectare and per hour of work, and therefore release land and labour for the cultivation of cash crops. Millets and maize diminish in relative importance.' As long as these three possibilities still exist we can count tion growth that are at least as high as those of populationon rates of producgrowth. However, as soon as production reserves of this type are exhausted, there is a distinct danger that, to follow the nomenclature of Geertz (1963), agricultural evolution will turn into involution. Increasing subsistence demand leads to the reduction of the cash-crop area, and low-yielding though high-quality foodstuffs, such as millet and maize, are replaced by high-yielding but lowquality food crops, like manioc and sweet potatoes. The greater the population density and the smaller the farms, the higher is the percentage of crops like manioc and sweet potatoes and the lower the percentage of maize in the subsistence food. The final stage of this type of involution consists of very small holdings, without room for cash cropping, which grow root crops
2 Yields in traditional farming are frequently about 6 quintals per hectare for millet, 12 quintals per h.ctare for maize, 80 quintals per hectare for sweet potatoes, and 100 quintals per hectare for manioc. These amount, in kilocalories per hectare, to 2060 (millet), 4270 (maize), 7750 (sweet potatoes), and 33 800 (manioc). If we estimate the growing period of millet, maize, and sweet potatoes to be 4 months and that of manioc to be 18 months, the monthly returns in kilocalories per hectare 1070, with sweet potatoes 1930 and with manioc, which are with millet 515, with maize occupies the field for an especially long period, 1880.
ON RAIN-FED LAND
125
almost exclusively. The families living by this type of farming are particularly poorly nourished, diseases are more widespread than elsewhere, and the extent of underemployment is particularly great. The system of land use on Ukara in Lake Victoria is a remarkable attempt to overcome this process of pauperization. But even when all imaginable measures open to pre-machine men are applied to retain the fertility of the soil, as they actually are on Ukara, it is apparently not possible to avoid decreasing returns per hour of work when land use is intensified on tropical soils within the traditional state of agriculture. The average returns per hour of work are particularly low on Ukara (see Table 6.3). A similrr involution can be observed in the animal economy. In unregulated ley systems with sufficient grazing, the herds per family are large and the animals well fed. A spatial increase in arable farming reduces the grazing area until almost all of the land is annually cropped. The production of fodder crops is neither customary nor competitive. This usually gives rise to a situation with high numbers of poorly fed cattle. In some cases, cattle are replaced by goats and sheep. Only in exceptional cases is the cattle e'onomy integrated into some type of mixed farming. The cattle numbers per family decrease, therefore, with increasing permanency of cultivation, although livestock densities per 100 hectares may be high. The final phase of this involution is reached when the livestock economy is reduced to small herds of animals per family-mainly goats-which live on crop residues and on areas unfit for cultivation. Neither of the methods to increase output that are practised so skilfully by smallholders--extending the cropping area and changing the cropping pattern-is sufficient to secure a rising income for a growing population. There is a great danger of stagnation at a somewhat higher level of income, or of a decline in incomes, once these possibilities are exhausted. Increasing land shortage reduces not only the marginal returns of labour but often also those of land. The smaller the holding, the lower are, as a rule, the average gross return and thc marginal gross return per hectare, simply because the smaller the holding the smaller is the scope for cash production with crops yielding high returns per hectare. A situation like this is revealed, ;or example, in three neighbouring districts in Sukumaland. In the thinly populated Shinyanga area the family income rises rapidly and constantly as the labour capacity increases. In the more densely populated Kwimba area labour productivity is alrcady considerably lower. Finally, on the island of Ukerewe, Lake Victoria, where permanent cropping is practised, there is scarcely any connection between family income and the family's labour capacity (see Fig. 6.5). Because of the priority attached to subsistence-food production, the smallholdings with permanent rain-fed farming are caught in a situation that might well be called a 'low-level equilibrium trap'. The only way out, except
126
by migration, is to attain high returns per hectare and per animal by introducing yield-increasing innovations.
Shinyanga Kwimba Ukerewe I :
4000
0741
2-2,-10091 oo2 x
y2
=
'34-68x
y- 521.5182 .
2 0.660::.
-4000
3500L
.3001 2500F /
s-13000
-13500
.
'
-12500
~2000~1500k ~1000
500
0 1
*
-.2000
01500
-400a .
2 3 4 5 6 0 1
.
2 3 4 5 6 2 3
- 500
4 5 6 Man-equivalent per holdinG Man-equivalent per holding Man-equivalent per holding FIG. 6.5. The relation between land availability and labour productivity in three areas of Sukumaland, Tanzania (from Rotenhan 1968. p. 78), Twelve families out of the total of seventy-five studied have been omitted because they receive non-agricultural! incomes.
6.6.
6.6.1. Improved permanent cultivation systems (a) Technical innovations. The possibilities for intensification and improvement of permanent arable farming on rain-fed land have been drastically changed by technical progress. Intensive cropping with decreasing soil fertility and low yields per hectare used to be the final stage, a blind alley in the development of tropical farming systems for those who had not enough land for fallowing and not enough water for irrigation. Now, we have an almost unlimited number of useful innovations, which are of limited applicability in shifting and semi-permanent systems, but which, from a technical standpoint, are fully applicable in permanent farming. These include: (1) new crops, new or improved varieties, and improved animals; (2) new types of material inputs, such as mineral fertilizer, pesticides, equipment, and tractors; and (3) new and improved husbandry practices, which have become worth while because of new activities, better varieties, and new types of input. The work of many research stations indicates that permanent arable cropping with high yields can be practised in a tropical climate, with the
ON RAIN-FED LAND
127
liberal use of organic and mineral fertilizer, terracing, and soil-conserving crops, mineral rotations. The interaction between better varieties of new imporespecial of is fertilizer, improved husbandry, and the use of tractors operations, of timing tance for permanent cultivation systems. The proper of yields per which in rain-fed farming tends to be a major determinant hoe or oxhectare, can be achieved much better with tractors than with that the plough cultivation. On the other hand, we must not overlook the fact permanent realization of all that is technically and economically possible in types appropriate changes: institutional rain-fed farming requires significant r oth of range wide a and services, efficient of land rights, land consolidation, cannot be associated structural elements, without which full development achieved. It is important to realize that new farming inputs have to be purchased. permanent The way out of the low-level production trap of traditional must be Prices cropping is dependent on cash production and on markets. preliminary These sufficiently attractive to induce changes in input patterns. rain-fed requirements do not yet exist. in most cvses, so that most permanent farming stili takes place at a low level of productivity. introduc(b) Introduction of new rotations. Even more demanding than the farming, tion of new cash crops and new inputs is the introduction of mixed permanent putting in step important most the which might be considered starting points cropping systems on a sound, self..-sustaining basis. Two main on the arable production fodder of may be distinguished: the combination of greeninclusion the and land with intensive types of animal husbandry, manure crops in the rotation. ConspicuThe first way is the more intensive one, especially with dairying. leguminof inclusion ous increases in output may be expected, because of the application the ous fodder crops in the rotation, liberal application of manure, and conserving of fertilizer, better soil cultivation, and other soil-improving connected practices.' On the other hand, the high input of labour and capital farmwith intensive animal production must not be forgotten. I permanent preferred to ing, stabling with no grazing (cut and carry systems) is often required keeping the cattle on pastures. Consequently, to the high investment must be systems, Icy regulated in for dairy stock, which is also incurred and manure, spreading water, of added the cost of stables, the provision intensity, of level low a on transport for fodder and manure. Mixed farming
of animal manure Ac,-ording to Gtinard (1957), in the West African savannas 10 tons or sorghum. It has been per hecate is required annually for permanent cropping of millet of ,Crmyard manure per suggested that, in East Africa, dressing of the order of 2 tons live years. is p-obably every hectare par year, or 7.5 tons every three years, or 15 tons (Webster and Wilson 1967, necessary to maintain yields in the abs-nce of mineral fertilizers per hectare pe' year, were tons p. 188). Very small quantities of manure, amounting to 2A according to Alkali (1967, sufficient to maintain yields in most areas of northern Nigeria,
p. 8).
128
with a poor fertilizing system and traditional patterns of animal husbandry, is widely practised in India and yields low returns. But intensive mixed cropping, including heavy manuring and high-technology animal husbandry, is so far seldom found in the tropics because it requires so much in capital, knowledge, and markets.' Intensive green manuring is technically easier, substantially cheaper, and probably conserves the soil equally well. 2 The capital costs of livestock and the costs of transporting fodder and manure are not incurred: the greenmanure crop is p!our,:h;d i'tc, the soil where it grows. The main problems of a mixed-farming system based on green manure are: (1) Insufficient land supply-farmers cannot forgo the income derived from cropping the land that is expected to carry the green-manure crop. (2) The cost of mineral fertilizer for the green-manure crop. (3) The amount of traction power required to plough in the green manure -animal or hoe cultivation is often inadequate and tractors are essential. (4) The retention of nitrogen as the green-manure crop rots, because it is often difficult to plough in the green-manure crop in time, especially in the drier savannas. (5) The difficulty of arranging for green-manure crops to cover the soil in the dry season and protect it from the sun. Thus, ploughing in is best done at the end of the dry season. However, a good seedbed, resulting from the careful ploughing in of the green manure, caunot be created very easily at this time of the year. Intensive green manuring is tied to effective traction power and to favourable price conditions, but not to the same extent as i, intersive animal husbandry. The developments so far in Africa, particularly in West Africa, indicate that permanent cropping systems with green manuring are more economical, and therefore are more likely to spread rapidly, than systems with intensive animal husbandry. In the intermediate forest zone of Ivory Coast, for instance, where there are two rainy seasons, rain-fed rice and cotton are grown in one year for two to three years, followed by two to three years of a green-manure crop (Stylosanthes). (c) Introductionof intensivepermanentpastures. Where permanent cultivators in the tropics have permanent pastures, these pastures are, except in
I A special form of mixed farming is the combination of mixed rain-fed cropping with fodder cropping from bushes and trees. In West Africa, we already find an instance of this which is economically important: the seeding of Faidherbia albida (see 6.4.3), which, with rowplanting at the edges of fields, can be used in combination with plough culture. The disadvantages of this kind of combination are (1)the slow growth of the Faidherbia albida, (2)the difficulties of seeding (the seeds do not germinate until they have passed through the stomachs of the cattle), and (3) the thorns on the branches, which are tolerated by only the undemanding Zebu cattle. 2 In the savannas of West Africa, 30 tons of green manure are said to have the same fertilizing effect as 10 tons of animal manure (Guinard 1967).
ON RAIN-l:ED LAND
129
prosome highlands, almost always on land that is unfit for cultivation and techrecent more vides only poor grazing. It is conceivable, on the basis of nical progress, that future arable systems in the tropics will have intensively used, permanent pastures on potentially acable land. In the humid tropics may be the growth of herbage is rapid, and heavy yields of green material mineral obtained from sown or planted stands with a heavy application of areas is tropical humid in grass managed fertilizer. One hectare ol" properly (Whyte cows six, even or four, for fodder capable of supplying enough green 1967, p. 126). The difficulties of using tropical grasses are summed up by Whyte (1967, p. 56) as follows: However, quality is usually below that of good herbage fro'n the temperate zone, particularly in respect of dry matter. As one of the direct effects of a tropical the climate is to depress the appetite, a low dry matter content further reduces ability of the animal to obtain enough dry matter for maintenance and production. The dry matter intake of dairy cattle in Trinidad during the wet season is significantly less than that of the same animals during the dry season. The problem is further complicated, because the crude fibre content of humid tropical herbage appears to be consistently higher than that of temperate herbage are at the same stage of growth, an additional disadvantage to animals that forage. watery of bulk sufficient eating in finding difficulty He continues (p. 81): so There are only very limited areas in the humid tropics where the rainfall is available of supplementation that year the throughout uniformly distributed grazing resources is unnecessary when farms are stocked to full capacity. There is also a fairly long period in each year when the growth of pastures and fodder crops is greater than the requirzments of the farm animals. Unfortunately, even best when the best harvesting practices are adopted, the tropical grasses that are limited only of and coarse relatively are environments humid suited to hot nutritive value. Even when nutrient losses in conservation aire reduccd to a minimum, the final product will be inferior in quality to conserved fodders in the temperate zone. The fundamental problem with which agronomists in the humid tropics have to contend. is thac- fact that the surplus herbage which may be harvested for conservation usually grows at a time when weather conditions are quite unfavourable for the making of good hay and/or silage. In many respects the first steps have been taken to overcome these problems. New fodder plants, which are locally unknown, are being made available. Silage made from grass or fodder crops grown on the arable land can help to provide fodder in the dry season. As improved permanent pastures are created, it is debatable whether p. 69) grazing systems should be preferred to zero-grazing. Whyte (1967, writes: The arguments gen.rally produced relate to the extra costs and labour involved in cutting the feed and car'ying it to the animals, and to the fact that this system breaks the fertility cycle--the movement of plant nutrients from the soil to the
10
130
SYSTEMS WITH PERMANENT CULTIVATION plant, to the grazing animal and back to the soil in the form of dung and urine, The grazing of highly bred, highly productive dairy cows in the tropics or in the hot summers of the subtropics raises a number of problems relating to heat load. Under tropical conditions with a long growing season, extended, if necessary, by seasonal irrigation with clean water, cowshed wash or sewage, very high yields of palatable tropical grass may be obtained. Two factors then arise: (a) There is on average temperate pastures a loss of 40 per cent of the herbage, due to trampling and urine poisoning; on tall, Iiigh-yieldiig tropical pastures this percentage may even be higher, when grazed, and the loss of precious feed represented by that percentage may be too great to contemplate; and (b) There must be a threshold value for yield, above which too much of the herbage is rendered unpalatable through the dropping of very large amounts of dung and urine by grazing animals.
Thus, the higher the yield from a tropical grass crop, the less can one justify preferable to adopt the cut-and-carry system or zero-grazing and obtain 100 per cent of the yield, provided arrangements are made for full return of animal fertility back from the cowsheds to the land. It would be argued that the 40 per cent loss due to grazing and trampling is worth more than the costs involved in the cut-and-carry system. Others would say that the more expensive the labour, the more economic will the cut-and-carry system be compared with grazing, as it can be highly mechanized, can assure the effective utilization of all the fodder produced on intensively cultivated land, and will make it possible to obtain a higher return per labour hour than in any other system.
Where intensive dairy production is found in the tropics, for example in the
grazing it. Some specialists say that when labour is still available and cheap, it is
West Indies and in various Latin American countries, grazing systems on permanent pastures are usually preferred. In smallholdings with tree crops (see 8.4-8.6), and in irrigation farming (see 7.4), cut and carry systems are, however, establishing themselves. 6.6.2. Change from permanent rain-fed cultivation to other types offarming (a) From rain-fed cultivation to irrigation farming. Generally speaking, we there is the opportunity-to develop irrigation at the expense of rain-fed farming, especially where holdings are small. Even small irrigated plots can be important in reducing the risk of harvest failure, in using labour and traction power regularly and to their full capacity, and in producing a regular supply of fodder. Over the centuries, in south-east Asia particularly, the change from shifting and semi-permanent systems to wet-rice systems has taken place (Pelzer 1957, p. 75). Wet-rice developments begin, as a rule, by using valley bottoms, and as the land becomes more scarce terracing is developed on the slopes. In Madagascar this process is in full swing, and in East Africa it is in the initial stages. Relatively large areas could be made suitable for wet rice in Africa and Latin America,
find among permanent cultivators in the tropics the tendency-wherever
ON RAIN-FED LAND
131
Whereas we may expect an accelerated change from shifting, semipermanent, and permanent rain-fed farming to irrigation farming, and particularly to wet rice in Africa and Latin America, it seems doubtful whether this tendency will continue in the densely populated regions of south-east Asia. The conversion of marlginal hill land to irrigated terraces requires a high input of capital and manual labotr. The rice terraces on the slope are mostly narrow, and it is difficult to mechanize the labour operations. With the new means of increasing the yield on the existing rice land, by using new varieties, mineral fertilizer, pesticides, etc., it is in general more profitable to use the existing areas more intensively than to create new rice terraces. Technical progress has, moreover, considerably improved the production of rain-fed farming, so that the transition to ,vet rice no longer represents the only way of increasing the productivity of land and laboiu. Under the conditions of traditional agriculture, however, it was m,.ore or less the only way.
(b) From rain-fed cultivation to perennial crops. The change-over to
perennial crops, particularly banana:;, for food and to various tree crops for cash, is another obvious tendency as population density rises and commercialization increases. Bananas produce almost as many kilocalories per hectare as tropical root crops, with less labour per unit of starch. Banana groves offer an almost complete cover for the soil and iluprove its fertility. They are clearly an attractive alternative to permanent arable cropping wherever soil and climate are suitable. Over and above this, there are many indications that tree or bush crops are gaining in importance as development takes place. On the slopes of suitable areas small plantations are frequently spreading at the cost of permanent arable farming. Examples are the coffee and the tea plantations of the Kikuyu in Kenya, and the tea and fruit plantations of the smallholders in Taiwan. It can usually be observed that the displacement of permanent arable farming by tree or bush crops depends on the cash-cropping opportunities, i.e. on the availability of attractive markets.
7.2.
133
supplies, are characterized by a 'control of nature'. Both water supply and soil conditions become increasingly subject to human control. Therefore, we classify the forms that irrigation farming takes according to the water supply system, the cropping system, and the exploitation system. 7.2.1. Water supply systems Table 7.1 gives a synopsis of the various possible methods of water supply, water distribution, and water application. Water for irrigation can be obtained from natural flows by diversion, damming, or pumping. The various
TABLE
7.1
Irrigationmethods
(I) Methods of obtaining water (a) Diversion of water from natural flows (river, stream or spring) (b) Damming: (i) Retention of tidal water (ii) Impounding of rain-water in paddy fields (iii) Retention of the water of seasonal streams in tanks and small dams (iv) Retention of river water in large dams (v) Exploitation of natural stores of water, lakes, etc. (c) Pumping: (i) Manpower-pumping by foot. using a lever arm, etc. (ii) Wind and water power-wind pumps, water wheels, etc. (iii) Animal power-winches, inclined plane, etc. (iv) Motor pumps (2) Methods of water distribution (a) Ditches (lined and unlined) (b) Flumes (c Pipelines (3) Methods of water application (a) Flooding (controlled and uncontrolled)
ways of damming water are particularly prolific. They range from relatively simple methods like retention of tidal water (rice irrigation on the coast of Java, date irrigation in Schatt al Arab, Iraq), growing of swamp rice in naturally flooded areas of river valleys, and impounding of rainfall in paddy
134
FARMING
Secondary gitch
Ok
,,
.I
(a)
. ...
-
"<
--Supply
canal
Ch c
.S
h ".. Siphons
/
Border
,,... 7 ......
(b)
FI., 71 Types of suracc irrigation, (a) Flood irrigation; (b) basin irrigation; (c) furrow irrigation (from Cantor 1967, p. 31).
especially the damming of fields, to more sophisticated and artificial methods, enormous size and cost.' of water courses, which can involve capital works fir ditches, above ground canals Water is conveyed either on the surface in methods of applying inain The through flumes, or underground in pipelines. sprinkler irrigation and furrow, water in the fields are by flooding, basin, a larger plot, the onto water (see Fig. 7.1). Flooding is the practice of letting furrows. Basin irrigationis or by water being distributed either into basins smallholdings: wet rice is cropped intensively in the practice mostly applied stands in the basin during all or almost always basin irrigated. Here water with the help of small basins part of the vegetation cycle. Water distribution Furrow irrigation is usually is also usual for irrigated vegetable production. comparatively large, well-levellec associated with the use of a tractor and in Gezira, Sudan. There are, fields, as, for instance, fields for cotton-growing basin and furrow irrigatioa. however, numerous intermediate forms between dams and ettensi,e canal Large-scale irrigation schemes, with sizeable In the tropics, irrigation is systems, are found principally in the subtropics. traditional pump methods that mainly basin irrigation. Supply of witer by in India, but in recent employ people or animals is common, particularly in the tropics. Sprinkler irrigation times the motor pump has spread rapidly mostly applied to perennial crops in the tropics is still in its initial stages and like sugar-cane or coffee. problems of an irrigated Cropping patterns and the farm-managernent degree of water availability and holding are decisively influenced by the water reliability. availability on the holding can (a) Water availabilityand land use. Water points of view: be considered from a number of different to area. Holdings on which it is (1) The availability of water in relation afe found principally in the flood possible for all the land to be irrigated (Punjab). plains of large rivers (China) or neat canals or springs, usually only Where irrigation is supplied by small streams these cases the holding In a part of the holding area car. be irrigated. irrigated production and supplecontains two distinct types of farming: mentary rain-fed cropping. to time. Holdings whose land is (2) The availability of water in relation to water supplies that flow all the irrigated the whole year round, thanks of being able to crop intenyear, are in the especially privileged position example, of oasis agriculture, and sively and continuously. This is true, for
135
but obtaining water are used ir. highly populated on Rainfall I Particularly ingenious methods for of Yemen. the in water procurement instance dry areas. An example is the method areas-for some In fields. earth-bound between rocky surfaces is conveyed into adjoining of Sanaa-I found a systematic alternation of hectares 3 at Sunnatayn, 150 kilometres north surfaces i.e. 3, to I about in a ratio of fields and completely eroded rocky of rainfall they millimetres 300 the of cent pcr 50 about rocks with a high run-off convey receive to I hectare of arable land.
136
of rice-growing in Taiwan, Java, southern India, etc., where there are two or more harvests annually. Water discharges from rivers generally have peak periods, and irrigation farming has to be adapted to the availability of water during the year. In holdings whose water supply is sufficient for irrigation in only one growing season, the question arises as to whether rain-fed farming is possible for the rest of the year. In the semi-dry or dry zones, the irrigation area lies fallow in the dry season, or serves as pasture. In south-east Asia, the production of monsoon rice in paddies is typically confined to the wet season. Where the rainfall is high enough, however-for example in Ceylon and the highlands of Madagascar-irrigation farming and rain-fed farming alternate on the same land: in the summer it supports wet rice; in the drier, winter season non-irrigated grain or legumes. The availability of irrigation water in the dry season is of particular value in areas with pronounced and heavy rainy seasons, because during the rains crop development is often stunted owing to waterlogging, cloudiness, and plant disease. Crop yields are usually much higher during the dry season. (3) The availability of water in relation to quantity. Over and above this, we distinguish between: (i) crops that rely fully on irrigation, and could not be grown at all if additional water were not available; (ii) the supplementary irrigation of crops that could also be grown without additional water, but yield higher returns on irrigation; and (iii) semi-irrigated fields, which receive surplus water from fully irrigated fields whenever there is surplus water available. In the first two cases a further distinction may be made according to whether irrigation holdings have plenty of water; or water is scarce in relation to plant needs, or must be bought at high prices. Where there is no lack of water-as for example in the Burmese rice holdings-the objectives of farm management lie less in maximizing the value of output per cubic metre of water than in maximizing the value of output per hectare or per worker. In other cases-for example in the cotton holdings of Gezira, in the Sudai,-a high productivity per hectare is of less importance than a high productivity per unit of water. Limited availability of water leads necessarily to systems of water distribution, which then influence land use. In most irrigated areas, a common water supply is shared by a number of users. Division of the water in such cases may be based upon established priorities for use, or the users may share equally in the available supply, with division based upon the area irrigated by each farmer. The total flow is sometimes divided among the users so that, according to this entitlement, each obtains his share as a continuous flow throughout the season. A more common practice, however, is to rotate the use of the full stream among users. The length of time during which each
137 SYSTEMS WITH ARABLE IRRIGATION FARMING the area irriuser is entitled to make use of the water is usually based upon to share gated by him, and is limited so that all users will have an opportunity distributing of method Another intervals. in the use of the water at reasonable irrigation irrigation water, which is used on the more highly developed the requires This request. upon projects, is delivery of water to the users the throughout system canal maintenance of a flow of water in the entire irrigation season. irrigable The cost of the water, which is met from water charges, taxes on the determining in land, or betterment levies, is also an important factor irrigation irrigation method used. Where water costs are high, water-saving comparatively methods will be applied, and crops will be grown that yield high returns per unit of water (Boucher 1Q67, p. 3). of irrigation (b) Water reliability and land use. It is the ultimate objective under his water has he that position a such works to put the irrigator in attained. control at all times. This ideal arrangement is, however, not usually The drought. or flood of threat the Most irrigators in the tropics work under irrigation of organization the in reliability of water supplies is a decisive factor systems, which can be classified according to this. depends on (1) The water supply is particularly unreliable where it on the west areas seasonal flooding. Thus, for example, in river-flooded plains of flood coast of Madagascar, in the Niger Delta, in Mali, on the on land the Amazon Basin, or in the Rufiji Delta, Tanzania, rice is planted is water flood where water is sure to collect during the flood season. The normally sufficient to produce a reasonable rice crop. water. For (2) In swamp-rice systems use is made of naturally dammed hollows are example, in Ceylon and West and East Africa, damp, swampy planted with rice in the rainy season. is impounded in (3) Water supplies fluctu,.. widely where natural rainfall supply for paddies. The farmer has, as i rule, ao influence on the water the requirements his rice, but he tries to adapt the run-off from the basin to of the rice crop. streams, or by (4) Water supplies obtained by diverting the natural flow of in quantity storing water in tanks and small dams, will normally fluctuate The same year. to year from vary during the irrigation season, and will with years after dry become may often applies in the case of wells, which reckon to have will poor rainfall. Consequently, some of the irrigators with reduced water supplies in some years. associated with a (5) The most productive forms of irrigation farming are case where the the usually controlled and reliable water delivery. This is from an pumping by or water is obtained from a large surface reservoir underground reservoir (Boucher 1967, p. 2).
138
With a decrease in the control and reliability of the water supply, production risks increase, and these become greater as the intensity of farming increases. Irrigation systems with uncontrolled and unreliable water supplies are therefore in general more extensively managed and offer reduced scope for introducing agricultural innovations. As the water supply becomes less reliable, crops like sorghum, with a lower productivity but a low susceptibility to partial drought, are given priority over other, better but more vulnerable crop types. (c) Water quality and land use. Another important factor influencing land use is the quality of water, in particular itz content of salt, which may determine irrigation methods and cropping patterns as well as investments in drainage. Where the salt content of the water is high, we find a preference for salt-tolerant crops like date-palms, cotton, rice, barley, lucerne, and cabbage. As far as possible, farmers will try to irrigate in those seasons when the salt content of the water is relatively low, and they will prefer irrigation methods like sprinkler irrigation, which reduce the import of water and thus of salt. Under extensive farming conditions, fallowing, shifting of the crop area, and rotational irrigation are used to prevent the loss of land due to excessive salt accumulation in the soil. 7.2.2. Croppingsystems The close relationship between water supply, availability and reliability on the one side and socio-economic conditions on the other has caused various cropping systems to develop. The principal difference is between wetrice systems and other arable irrigation systems. In wet-rice systems, the land is flooded and the crop is grown in wet soils from the *.me of transjlanting or seeding until harvest approaches. The mthods used are not common to other crops, because the physical, chemical, and biological conditions in flooded rice fields differ substantially both from those met in rain-fed farming and from those that occur in other types of irrigation systems (see 7.4.2). Wet-rice systems have for centuries been the most important form of land use in south-east Asia, in terms of both area and the economy. In Africa and Latin America, wet-rice cultivation is spr-ading quickly. Other major tropical irrigation crops are cotton, sugar-cane, groundnuts, wheat, and vegetables. Large areas of irrigated cotton are grown in India and the Sudan. In India, we find also important farming systems with sugar-cane and wheat as the dominant irrigation crops. Numerous minor irrigation systems, frequently based on various arable crops and on vegetable production, are found in practically all parts of the tropics. 7.2.3. Exploitationsystems While we can say that shifting and semi-permanent systems are the province of smallholders, that regulated ley systems tend to be best suited to
139
units that are sufficiently large for animal or tractor ploughing, and that tropical permanert rain-fed iarming is again the province of smallholders, we cannot generalize about farm size in tropical irrigation farming. Tradtionally, irrigation farming and high population densities are two char'acter.. istics of the. same settimg. But wh~ile high population densities tend tc favour small farm sizes. there is the tendency to manage all the area irrigated from one source as a singe, large unit, and the more expensive the irrigation system was to develop. thie more pronounced is this trend. Where smallholders ar:. independent in their water supply, either because they have springs and wells of their own, or because impounded rain-water is sufficient for wet-rice production, they are equally independent in organizing their land use. In minor irrigation schemes of up to about 300 hectares, numerous smallholders usually operate either as tenants, getting their water from the landlord, or as owner-farmers, sharing the water according to traditional, purchased, or cooperatively agreed entitlements. Under such conditions, land use depends almost entirely on water righ s, in other words on the timing and quantity of water distribution. Water use i medium-sized (300-5000 hectares) and large (more than 5000 hectares) irrigation schemes is usually organized by some kind of scheme management or irrigation administration, and sometimes the scheme management is in a position to enforce definite patterns of land use. General characteristics of"irrigation farming 7.3. 1. A comparison with rain-fedfarming Irrigation farming has distinct advantages compared with rain-fed farming. Because of irrigation, the natural, ample, and regular supply of solar energy that is characteristic of the tropics is allied firstly with the ample and regular supply of water created by investments, and secondly with the nutrients brought by the water.' Moreover, irrigation, particularly in the form of impounding, avoids the stress to plant growth that is connected with very high temperatures on the soil surface. At the same time, the conditions for agricultural production are improved, particularly when the water supply is ample, regular, and reliable. (1) Irrigation farming compared with rain-fed farming under similar conditions produces higher gross yields per hectare. The basic possibilities (the realization depends naturally on the local conditions) are that: (i) higher yields per hectare of a particular crop may be achieved: (ii) several harvests a year may be produced;
Mainly by silting. in some ras, ground water shows a 11 i~h content of nitrate. At the Jumeisha farm near Hodeida, North Yemen, the amount of nitrogt brought into the soil by irrigation from wells was found to be more than 100 kilogramms per hectare.
7.3.
140
SYSTEMS WITH ARABLE IRRIGATION FARMING (iii) crops may be grown that produce comparatively high yields per hectare, some of which may demand a water supply all the year round; (iv) it becomes more profitable to use additional labour and other inputs; and (v) continuous rice cultivation becomes possible, thereby extending the area of this relatively high-yielding crop.
(2) Irrigation farming allows permanent land use, provided that salt damage can be avoided. In the case of rice cropping, it means that permanent agriculture can be carried on fo; centuries with high yields per unit area and without impairing soil fertility. The control of water reduces erosion. (3) Yield fluctuations from year to year are reduced. With reliable water supplies, production is almost independent of the weather. Annual fluctuations of yield are less than those in rain-fed farming. (4) Irrigation farming leads to relatively more continuous production. The chance of a regular household supply of food and cash income improves. The same applies also both to the provision of cattle with fodder and to the employment of the labour force, draught animals, implements, and buildings. (5) Arable irrigation farming is relatively adaptable with regard to both type and intensity of production. Holdings with perennial crops are obliged to adhere to one type of production for a long time. In holdings with rainfed farming, the rainfall determines the range of possible crops, and the less it rains the narrower is this range. In irrigation farming, such limiting factors are much less compelling. By eliminating the problem of drought, irrigation makes it possible to diversify production. (6) Irrigation farming allows the productive employment of a relatively large number of workers per hectare. It therefore enables a relatively high income to be earned without the use of expensive equipment, or it reduces the cost of such equipment when measured against the return. In hoe systems, one family can scarcely cultivate more than 2-3 hectares, which in rain-fed farming permits only a minimal standard of living to be obtained, but in irrigation farming the same area will allow a high standard of living. The increase in production on a small, irrigated plot means that the part of the proceeds spent on traction power is relatively small. (7) Irrigation development gives value to neighbouring land that could not otherwise be utilized, or would be under-utilized. Irrigation may permit speculative use of rain-fed land that would not otherwise be used. Rain-fed farming on soils with marginal rainfall, or stock-keeping on dry pastures, is in many cases only possible because irrigation production guarantees some yield reliability.
by saying that, in comparison with These arguments may be summarized manushows some of the advantages of rain-fed farming, irrigation farming cortinuous much to chance; there is a more facturing. Matters are not left so and factors; production the of control production process; there is a better production. there is a spatial concentration of of irrigation farming must be set advantages many However, against the definite costs and requirements: investment in water supply, delivery, (I) It necessitates a high level of the case of large land preparation. In - (or annual costs of and distribution works, and in per hectar schemes, development costs of $500-2000typical. High levels of investment are 2 SO.003-0.0 per cubic metre of water) that water has to be bought for a so costs, fixed of burden heavy a entail to practise rain-fed farming and high price. It is still often more profitable areas, invest in irrigation. Often, in arid a.cept its disadvantages than to or engineering resource, and therefore water is considered to be a unique investjustify to sufficient conditions agronomic feasibility are considered 6). of resources (Carruthers 1968, p. ment. This may lead to a misallocation farming the initial investment, irrigation (2) in addition to the high level of high level of labour or machine in its common form demands a continuous pumping, maintenance of the system), input to supply the water (drawing, fields and water-distribution or drainage and to prepare and maintain the channels. requires the cooperation of several (3) Irrigation farming, as a rule, or accept instructions from a private farmers, or their willingness to communal requires irrigation holding public water lord. Moreover, an water-delivery system. In the larger the work to maintain and improve schedule and scale of water distribution irrigation systems, a disciplined for the most important pre-conditions among the beneficiaries is one of economic success (Wittfogel 1931). production possibilities, but a good (4) Irrigation farming may offer great producthey are to be fully exploited, for deal of knowledge is necessary if is the One problem of particular gravity tion is technically complicated. of diseases like malaria and bilharzia, spread among the irrigation farmers so that medical costs have to be conwhich are encouraged by the water, irrigation development. sidered in addition to the costs of tenure production techniques and land It must also be borne in mind that and semi-permanent systems communal are interrelated. Whereas in shifting find in rights of usage prevails, we generally established with ownership land tenants. ownership with owner-farmers or have permanent farming individual land they that provided relationships, In irrigation farming, landlord-tenant are even more typical than in rain-fed not yet been abolished by land reforms, relationships is perhaps due to farming. The prevalence of landlord-tenant
141
SYSTEMS WITH ARABLE IRRIGATION FARMING the fact that, for water distribution to several irrigators, institutional arrangements on a larger scale are needed. 142 7.3.2. Spatial organization of cropping Having outlined some of the generally important farm-management characteristics of irrigation farming, we can now consider the more specific organizational features. Again, we find that the various plots of a holding are, as a rule, not uniformly cropped, but that definite spatial cropping patterns prevail. There is, however, a basic difference in this respect between rain-fed and irrigation farming. In rain-fed farming, the spatial organization of production is based primarily on the varying intensity of manure application, in other words on the distance from the farmstead and the consequent transport problems. In irrigation farming, it is based primarily on the availability of water. Three different forms may be noted in this respect, based on whether only a section of the farm area is irrigable land; or all of the land is irrigable, but water supplies are insufficient to provide all of it with the optimum amount of water; or water reliability varies from plot to plot. (a) Spatial organization of partly irrigable holdings. Where only a part of the holding area can be irrigated, land use is diversified, more often than not in the form of a catena. The cultivator usually concentrates his work on the lowest topographical level, which is generally more fertile and easier to irrigate. The seasonally or continuously irrigated fields which are the centre of the holding, and which are usually close to the water supply-a spring, river, or canal-may be supplemented in different ways: (1) There may be gardens close to the farmstead. The combination of irrigated rice fields and comparatively large gardens, containing various vegetables, t-ee crops, and ponds, is the basic pattern of rice holdings in large areas of south-east Asia. (2) Permanent or semi-permanent arable culti,ation may be practised on rain-fed land. This is the basic pattern of most of the holdings with minor irrigation. Irrigation farming close to a spring or small river in the valley bottoms, with rain-fed farming on the surrounding slopes, is typical of many Indian farms and of rice holdings in south-east Asia, Madagascar, and other places. (3) Perennial crops may be grown on rain-fed land. Commonly, we find only small plots for crops for home use, but sometimes there are substantial plantations of cash crops like rubber and coconut-palms. These activities are found in different combinations and proportions, and provide a variety of farming systems, differing widely in complexity and intensity. Two examples are given in Figs. 7.2 and 7.3. The situation depicted in Fig. 7.2 is typical of most holdings with wet-rice production in Africa south of the Sahara. The rice paddies are situated close to the
143 FARMING SYSTEMS WITH ARABLE IRRIGATION farming. One of the major river, and the neighbouring slopes are used for rain-fed isthe better distribuadvantages of the combination of wet rice with upland farming tion of labour requircments.
~~ Salt water
, 7"':.
.
I
"
',IF
Dam"
Bank of earth
ice
Rice nurseries
River (from Mohr 1968, Firo. 7.2. Cropping pattern in the region of the Lower Gambia p. 67). developed in south-east The spatial diversification of cropping is even more Singapore. from Asia, as shown in Fig. 7.3, which is taken (I) The valley bottoms are used for rice fields. produce fish and pig (2) In addition, on the low land there are ponds, which feed (green fodder). the waste food from (3) In the immediate vicinity of the ponds are the pigsties, fish. the which is put into the ponds to feed varieties like (4) Then come the main vegetable plots with water-demanding ponds. the from taken is cabbage, lettuce, etc. The water for irrigation little require which cucumbers, and tomatoes, beans, (5) Vegetable varieties like water, are grown on the higher land. manioc and sweet After this come non-irrigated plots with root crops like (6) maize. potatoes, which are sometimes supplemented by a little but there is also some (7) On the upper slope, fruit trees are principally grown, tobacco. like rubber and coconuts. (8) Finally, the high plateau is utilized for tree crops
00
Leafy vegetables
port
FIG. 7.3. Profile of a catenary cropping sequence in north Singapore tfrom !io 1964, Fig. i).
A -
J~'AA
A,'A
AAi~fAA A AA
~
if A AA .'
,A. . ..
AAA1A
A
Ar
_:_=_.
A'o'f
culivto
. ." , :
- .7 :.
- : :
.: -: - .
.'
. '.',.
'A
'
":
if
A, A:
it
.4g
O P.
.. .. ...
V-
TA,
;i
3. Valley bottoms in the savanna zone of the Ivory Coast being transformed into fields with irrigated rice
4. Dairy cattle on smallholdings with regulated ley farming in Nyeri D ,trict, Kenya
in the African savannas. 5. Tobacco isa cash crop well-suited to shifting cultivtion A small entrepreneur at Urambo. Tanzania
being transformed into 6. Valley bottoms in the rainforest zone of the Ivory Coast rice irrigated %%ith lields
8. Transporting manure from the stable to the field, Ukara Island. Tanzania
v~~
ip 4
lsland. Tanzania. The and green manure on Ukara 9. Applying stable manure manutre is placed at the ba,;e of the ridge
II. Swiss cows grazing on leys established on land traditionally used by shifting cultivators. Bush with tsetse-flies in the background. The picture was taken on a Swiss demonstration farm, close to Cotonou,. Dahomey
wil
11
12ImI,o
ric
i. d
f-4-
:: W
'
by seasonal 13. Land preparation for rice in Ceyl1on. The "vork is being performed paid labour
5 ' Relay-planting of sugar-cane in growking'rice in Tai%%an. Note the seed cane in 1 the mud between th rows of rice
,, .
W-
,.
... '..
j-
._,-, .
- .,
Mo..
ia
........
(Vtr
4A
J.i
..
*..
-. K.
.,
~-
India. Te field to te left shows' inproed 18. T orice fields close to Madras, nolThe field to the right sho\%S IRS. at traditional varieties. wich tend to lodge.
of.nitrogen (see Fig. 7.10) lodging variety which can productively absorb high doses
"
- "" -.
,.'-
. .
..,,- ,.
.""
"
- a
.4
...
..
- . . ...
, ".." , ., -.
o, - . --' SR . . . .
--. 9
' "" . . ,. . .
...
.* .
0*
7 -
*'-".,9 .- . : _.
. ..
..
.
.
.. -- '.-.
-
% ..
-. . . The work is performed after hoe or plough " -4'.. with buffaloes. in Ceylon. trampling 19. Preparation of a rice field for planting by 'pidtinage" called and applied is cultivation. In Madagascar the same method
-i4
4144
. --
-2"
_.
.4
4 J ' 4 4
'
14
....
444 4
_, 4
---
.-
: <
* ;,
-. :i . :, .' ! " i . :. "' 9
*Amp
,!
.
*1
,
z:
': , . > ,;
4/
':,
-'::,,
IL,
ral i m. m rFI !
...
...
....
. ...
I '
m n
' ,I
22. A field in the (jezira Sdcnie. in the Sudan. prepared for planting by the scheme nmanag~enent
23. A tenant in the Nvanagil Extension of the Gezira Scheme. in the Sudalnl. irrioating his CottnO
27. Young tea plaritation established %ith new clones, planted at high density. surrounded hbtea Mhi,'h is Ubout 50 %earsold. Ceylon. high-altitude leel
29. Smallholder plot ready' for the planting of oil-palms. The woman is planting a c, er crop I C'i'nzrcma/flhe.w/l. I
'Y.
.N
31. Root system of two perennial grasses in a dry savanna climate. Kontm~a. Tanzania
145 SYSTEMS WITH ARABLE IRRIGATION FARMING The various activities are not organized independently of each other; in particular, arable cropping on irrigated and on rain-fed land is closely related. Where the irrigated plot is small, and contributes only 10-30 per cent of the hrciding return, it is used essentially as a supplement to rain-fed farming, ,rincipally producing fodder in the dry season, and regularly supplying the household and neighbouring markets with perishable food like vegetables. Where staple food crops are grown, the primary intention is to guarantee the household's food supply. Since in most cases the irrigated land is cropped with particular intensity, any organic fertilizer from the holding is applied to the irrigated land. Where there is a relatively large irrigated area, rain-fed farming becomes a supplement to irrigation farming. The marginal return of additional labour on the irrigated land is usually higher than on the rain-fed land, with the result that the holding has on the one hand carefully and intensively cropped irrigated fields, and on the other hand extensively used rain-fed areas supplying seasonal grazing, fuel, litter, or undemanding food crops like manioc. (b) Spatial organi-ation co' totally irrigable holdings with limited water supplies (rotational irrigation). Where the total holding area is irrigable, but supply is insufficient to provide the optimum amount of water for a given crop, it is advisable only in exceptional cases to distribute the available water uniformly. Instead, systems with rotational irrigation are usually practised, i.e. the water delivery to the field follows a definite pattern from season to season or from year to year. We find primitive forms of rotational irrigation in the oasis agriculture of the Sahara. Nicolaisen (1963, p. 203) reports oasis farming where the irrigated plot is cropped with grain twice a year for six or seven years. after which it is left fallow and a different plot is irrigated and cultivated. In this shifting system of irrigation, the problems of soil fertility and of salting are solved without effort, because of the fallow period. In intensive agriculture, systems with rotational irrigation are incomparably more complicated, but also more productive. They are especially highly developed in Taiwan. In the example shown in Fig. 7.4 only a third of the area can be cropped with wet rice in summer. The other two-thirds supports sugar-cane and various short-term arable crops, which can grow p'operly with the natural rainfall. For the third supporting sugar-cane, which is not irrigated in summer, the little water available for irrigation in winter and spring is enough for initial growth, and as autumn approaches and the crop begins to ripen water is no longer necessary. Since the rice on the first third also needs no more water, sugar-cane can then be planted on the last third and irrigated until May of the second year. The crop rotation for each of the three plots is arranged in such a sequence that the full rotation on the holding's three plots is: sugar-cane (first year until January of the second year)-green nanure (February to May of the second year) -irrigated rice (July to October of the second year)--sveet potatoes (November of the second year to April of the third year)non-irrigated rice, soya beans, or groundnuts (May to September or October of the third year).
II
Year
Mon
-___
Ist year T 67 3
1?T
89
313
101 j T
]Jt~
-
3 12
45
Decode
1I~i 2 3 23 1 23 1] 2[31
St rl dt
11
Order
4re
-
3 1 73 13iJ
tiF 3 23 t
e,
12123 .-
4 -
3rdyear 67
!
-
J...-4..-.., " _ a rS
-f
IIIGQatao -hn
reen
i31 lI
1112
-a
Cro o
eo
ean e m
anur
he
-S ptanto
sa
w e
t s ae -
rice upland rice Fm. 7.4. Standard diagram of three-year rotation irrigation (from the Chia-Nan Irrigation Association, Taiwan).
Sweet potato.- Greent manure -- Sugar Green __ [ice __Sweet potato.- Green manure._ Repeat inorder wheat peanut .soya cane -- manurewheat peanut soya bean. bean, upland
spatial emphasis in systems We can see from the example in Taiwan that the to year. The type and intensity with rotational irrigation changes from year included in rotational irrigation. of land use is, however, the same on all plots water supplies. Other uncontrolled (c) Spatial organi:ation of holdings iwith may be found on holdings with types of spatial pattern in irrigation farming of uncertainty is different on uncertain water supplies and where the degree tend to adapt themselves to the different plots. Here again, the cultivators of crops, varieties, and conditions of the various fields by the selection supply. planting dates that lit the expectations of water inthe vicinity of Tananarive, MadagasAn example of this isthe wet-rice croppinghave a relatively ample and sure supply car. The plots in the valley bottoms, which with slow-maturing varieties in Aprilfirst of water for a long period, are planted The terraces on the slopes are May, and these produce relatively high yields. to December. Plots on the upper October planted with the standard varieties from are planted up to March, slopes, where the yield is particularly uncertain, availability. water according to the volume of rainfall, and resulting different varieties of rice on Similarly, the Pakistani rice-jute farmer plants of flooding and the ability of fields at various levels according to the danger the variety to withstand it (see 7.4.2 (b)). 7.3.3. Cropping principles are applied, in addition In irrigation farming distinct cropping principles in this respect is to definite spatial patterns of organization. Diversification ample and constant, the climate the more common where the water supply is purchasing power are high. is warm, and the population density and local diversified farming necessarily This does not mean that irrigation farming and cotton, and in particugo together. Irrigated monocropping with sugar-cane, is, however, a tendency There lar with rice, is very important in some areas. in some cases also hectare-and per yields increase to in irrigation holdings as ratooning, rotations, relayper man-hour-by following such practices planting, interplaiiting, and intercropping. in which two or more (a) Ratooning. In rain-fed farming. ratooning, is a practice generally consecutive c-ops are taken from a single planting, although it is of scarcely applied to sugar-cane and sometimes to sorghum, In irrigation farming, any significance as far as other crops are concerned. found in low-intensity is rice the possibilities for ratooning increase. Ratoon grazing and only in rice production, where it serves mostly as dry-season forms of ratooning intensive exceptional cases is the grain harvested. Highly The sequence: rice Taiwan. and Ph~lippines th.; in have been developed days)-sorghum, (80 ratoon (110 days)-sorghumi (85 days)-sorghum, first a production as (1968) second ratoon (80 days) is mentioned by Bradfield one year. possibility of tropical irrigation farming within
147
148
SYSTEMS
(b) Crop rotations. The crop rotations in irrigation farming may be classified according to (1) whether seasonal or continuous cropping is practised, (2) whether the rotation is based on one, two, or more crops a year, (3) whether short-term, long-term, annual, or biannual crops are grown, (4) which crops there are, and (5) what input demands the crops make (see 6.3.2 (b)). Among the many factors determining the crop rotation, the availability and reliability of water supplies play a decisive role in irrigation farming. In addition, interactions between rainfall and irrigation have to be considered. Multiple cropping is much easier to organize in areas with only one rainy season and irrigation during the rest of the year than in areas with two rainy seasons, where planting and harvesting have to be adapted to the rainfall. Consequently, the most intensive and diversified types of irrigation farming are found in areas with a single rainy season, as is the case in Taiwan. We can see from the rotations practised in Taiwan that are illustrated in Table 7.2 how the crop rotation, depending on the availability of water, varies under otherwise similar production conditions. TABLE 7.2 Crop rotations on irrigatedland in Taiwan
Fields irrigated for the whole year 1. Rice (February/March-July), rice (July/August-November), fallow (NovemberFebruary) 2. Rice (February-June), rice (July-November), green manure (November-January) 3. Rice (March-July), rice (August-November), wheat (December-February) 4. Rice (February-July), rice (July-November), vegetables (October-January) 5. Rice (February-June), vegetables (June-July), rice (July-November), sweet potatoes (November-February) 6. Rice (February-June), rice (July-October), tobacco (October-February) Fields irrigable for, 7/f a year 7. Rice (March/Ap il-July), sweet potatoes (August/September-January/February) 8. Rice (July-Nowember), vegetables (October-March), green manure (March-June) 9. Rice (July-November), sweet potatoes (October-May), green manure (May-July) 10. Rice (May-September), sweet potatoes (September-April), vegetables (May-Decem.
May)
Source: Chen (1963, p. 239).
on permanently irrigable land con(1) The most frequent crop rotation most non-irrigated winter crop. which is sists of two crops of rice and one the plots, or wheat. On poorly drained frequently sweet potatoes, maize. of size the As rice-rice-green manure. usual rotation is rice-.rice-aliow or becomes more important. holding decreases, vegetable cropping only, and prior to the summer rice, (2) On plots with summer irrigation potatoes, and the final crop is usually we find as a rule vegetables or swr a green-manure plant. the limited availability of water (3) In areas with rotational irrigation, of the principal crops rice and sugarnecessitates alternating combinations are three years with five or six crops cane. so that rotations of two to which crops five to rotations with two combined with the usual one-year irrigated (see 7.3.2 (b)). permanently predominate where the plots are extraordinarily high levels of land-use Plate 21 ives an example of the pracirrigation farming. The rotation year, intensity that are possible in tropical one in crops live includes four or tised in Taichung District, Taiwan, single a on year one Cropping within not all of which are short-term crops. piece of land is as follows. and it is laid out. In March the rice is planted, the sointerim In February the rice nurseries are planted until August. In the potatoes. harvested in June. Summer rice is not sweet beans, such as jute. melons, soya can be planted called summer crops can be grown. vegetables. Some of these crops the and plants second rice and various green-manure Between stand (relay-planting). season cool the as early as May in the growing rice of months four the rice. in tobacco. harvest and the planting of the spring including winter crops are cultivated, when growth is slow, the so-called sweet potatoes, wheat, flax. 'arious greenrape, seed in the maize, soya beans, peas, of these crops can similarly be planted the of use the manure plants, and vegetables. Some method this by adopting
sometimes growing summer-rice stand, and two winter crops are possible. that so field can be concentrated
149
like these, there is a strong tendency Generally, in production conditions of as high a yield as possible per month to grow those crops that produce crops occupy the field can be shortened field use. The t me for which various nurseries. Indeed, cultivators try to in considerably by raising the plants in the nurseries, and crops that lend possible leave the plants for as long as Moreover. it is necessary to prepare themselves to this method are preferred. the for the plants: a plot which in the field immediately after the harvest with evening rice should be green in the morning is yellow with ripening newly planted rice. systems. as depicted in Plate 21, (c) Rela'V-planting. Multiple-cropping techniques-the rotation crop is depend to a high degree on relay-planting crop. In the case of rice, relaypre~ious the planted in the ripening stand of the before the harvest and, so that weeks seven to planting takes place up above the the panicles of the ripening rice light, enough get can crop relay
IRRIGATION FARMING SYSTEMS WITH ARABLE is combined With some crops, reiay-planting row of plants are pulled apart. plants in nurseries. with the practice of raising the 150 The growing in Taichung District, Taiwan. An example of this is jute-growing but oetween in the summer, harvesting is 120 days 40 days. The problem is from sowing to only for period of jute available is plot they are then the spring and summer rice the for 40 days in the nursery, and of jute is thus plants jute the keeping by stand solved before the rice harvest. The pure and the planting of planted in the rice field 40 days the harvesting of spring rice between jute is planted in restricted to the 40 days rice is kept to a minirnum as the a pure stand of with summer rice. Disturbance to the compared losses in yield of rice, holes filled with compost. The these losses are more and of jute to 20 per cent, but cent, per rice, amount to 5-10 yield of jute. than outweighed by the additional force and on depends on the available labour The extent of relay-planting after the be heavy, the land must be prepared the soils. On soils that tend to for relay-planting again. In other cases, crops harvest and before it is used not be sufficient. may facilities and drainage can need to be planted in dry land. that the additional water requirements mind in borne be also must It adequate. Relayor the irrigation capacity is be met only where the rainfall but relaysummer, early particularly in that the planting in rice holdings is practised disadvantage rice suffers from the summer growing the in planting Thus, not in summer. tend to bend over. indica varieties, which are planted has to be rice the also to be separated, but only have the ripening rice panicles tied up. of planting of cropping possibility consists (d) Interplanting. Another form dlan a more ring the same time as a crop rec, does one short-term crop at about crops or manioc. The mixtiire of the year to ripen, for example sugar-cane while but only ripens. as in the case of rice, the in not last until the main crop is planted young. Sugar-cane. which still is plant space longer-growing months. The the somewhat cooler winter crops, like autumn, develops slovly in short-term for therefore be used can cane of rows the between making any appreciflax. or vegetables, without sweet potatoes, groundnuts. is a sufficient of the sugar-cane. When there has able difference to the growth interplanting of relay-planting and the in supply of water, the combination rice is interplanted in the ripening is crop proved its worth. The sugar-cane After the rice harvest, an intermediate should autumn (muddy-in method). shown that planting of sugar-cane. Experience has planted in the r. to weed and manure possible is it only, so that take place in every second row each crop individually. The combined relay-planting and interplanting. Figure 7.5 gives an example of in the ripening rice, so that it can benefit from row between the sugar-cane is planted in September the rice harvest, every second After days. warm the relatively potatoes. flax, or vegetables. growing cane is planted with sweet
(1) Crop rotation of the rice and the winter and summer cash crops
J
May
June July /
Dec.
,Jan.
Feb.
Mar.
m
1st crop
Rice
S _ ./' Winter cash crops /ti 12) Crop rotation of the rice and the sugar-cane. the several crops being interplanted at the beginning of growth of the sugar- cane. Summer rash [lops Feb
.---
Buckweat
Maizej eas Peas Vegetobles June / --"
. 7' , _ .," - _
f _/-__=
April Rice 1st crop
Mar.
Z._=1 Grinmue
Aplil May
July Aug. Sept. Oct. Nov. Dec. Jan. Feb. Mar. April Rce
.-2rl crop
.-
-- Z
__
___
y_-
ar-ca~e
F-'
."[ -1
r'kWIeat.. . .
Fax
1--Single-planting crop
_
t le ..
L .
FI;. 7.5. Rotations with relay-planting and intercrupping in Taiwan (from iso 1964).
152
(e) Intercropping. Relay-planting and interplanting lead necessarily to intercropping. i.e. different crops are planted in alternate rows, but with the
advantage that the mixed crop grows during those stages of the vegetation
cycle when the crops impair each other's growth relatively little. It should be remembered that intercropping also occurs in irrigation farming where two or more plants with approximately the same growing pet )I are planted patible.' An example is the intercropping of sweet potatoes with seed rape, in which the seed rape is planted about a week after the sweet potatoes and harvested a week before them. A general feature of intercropping in intensive irrigation holdings is the practice of row cultivation, which allows each crop to be given its own specific fertilizer and weeding.
7.3.4. Characteristics of the fertili-er economr on approximately the same date. These are preferably crops that are com-
The discussion of the various cropping principles indicates the high degree of intensity that irrigation farming can attain. It is obvious that, in order to exploit the prod-tion potential of irrigation farming, high yields, multiple cropping. and leac.-ing have to be accompanied by a highly intensive fertilizer economy, and by efforts to prevent salting. Most of the irrigated land in the tropics. particularly' most of the wet-rice areas, is still cultivated year after year without much manuring. Rice-growing over long periods without fertilizer application does not appear to lower soil fertility, for the following reasons: (I) In several large river basins, soil fertility is maintained by the annual deposition of silt. (2) In addition to silt. nutrients are brought in solution with the water. The paddies with wet rice are often on different levels and are slightly inclined, so that the water runs from the high forest land to the sea, draining slowly from one rice field to the next. and thus conveying nutrients from the higher levels to the lower plots. (3) In the rice terraces, the usual poor quality of tropical soils is improved because of the fixation of nitrogen by the blue-green algae which proliferate in the warm water, the slower rate of decomposition of organic matter, and the greater availability of many of the nutrient elements while the soil is waterlogged (Webster and Wilson 1967, p. 209). (4) Angladette (1966, p. 171) remarks on the observation that soils improve their quality for wet-rice production with time, because of the from A particularly interesting case of intercropping in irrigation farning is reported planted South Arcot District, Madras State, India. Cotton. finger millet, and groundnuts are of March.
vigorously, and is harvested in August.
in December-January at about the same time. Millet is harvested at the end Groundnut harvest takes place at the end of May. Thereafter the cotton receives an addigrows tional manuring and irrigation. The cotton, which was previously stunted, now
153
impounding of water and its influence on chemical processes in the soil. This process takes fifty to one hundred years. Intensive irrigation farming, however, requires manuring and fertilizing as well as the benelicial influences connected with water application and
water impounding. Traditional types of intensive manuring have developed in those areas where water supplies are sufficient and reliable and where population densities are high. Nowhere else in the world has the fertilizer economy of traditional farming systems reached such a level of intensity in the effort to reduce the loss in nutrients as in Chinese irrigL,.icn farming. The father of organic chemistry, von Liebig, in comparing nineteenth-century German agriculture with contemporary Chinese farming (1878, p. 453), viewed the "ormer as the procedure of 'a child compared to that of a mature and experienced man'. Webster and Wilson (1967, pp. 210-11) summarize the intensive manuritg practices used in south-east Asian irrigation farming as follows: All animal manure, household and crop wastes, and plant ashes derived from fitel, are carefully collected and applied to tie land, usually in the form of composts. Night soil is widely used. commonly L'ing stored in earthenware receptacles, diluted with water and thriftily applied to individual plants at appropriate intervals.' Mud, rich in organic matter. is periodically dug out from canals, reservoirs and fish ponds and spread on the land, or used in composts. Green manures may be included in the rotation, sometimes for direct ploughing-in but more often for composting. Hillsides bearing trees and shrubs are regularly cut over to provide material for trampling into the mud of the rice fields, or for the preparation of composts. Great use is made of composts, care be!ing taken to see that they are well rotted, with much of their organic matter broken down, so that they rapidly releas.. nutrients to a growing crop. It should be mentioned, howkever, that although the fertility of the densely populated alluvial plains of China has been maintained at a relatively high level by deposits of silt from irrigation of floodwkater, by application of night soil, and, to a lesser extent, by green manuring with vegetation imported from hilly land, this has been done partly at the expense of other land. Much hilly land in China has been depleted of fertility, and the general level of fertility of lands other than the alluvial plains is only
moderate.
It is interesting to observe that the concentric rings of land with decreasing fertility at increasing distance from the farmstead that are frequent in rain-fed systems are rarely found in irrigation systems. With the exception of the gardens near the farmyard. which receive more fertilizing, manures and composts are usually distributed more or less equally over the irrigated land. I Night soil is a traditional fertilizer of high productivity. It is, however, a carrier of various intestinal diseases. Buck (1930, p. 265) writes this about traditional China: 'From an economic standpoint it would probably be cheaper to throw away night soil than to incur the losses concurrent with ill-health which result from its use.' The transition to mineral fertilizer results in reduced losses through disease and reduced costs of health care.
SYSTEMS WITH ARABLE IRRIGATION FARMING 7.3.5. Characteristicsof animal husbandri' The intensification of irrigation farming is closely tied to an intensive fertilizer economy. Cattle activities, howe'er, are not in general intensively organized. In the irrigated farming systems of the subtropics, stock-keeping is of growing significance. In Israel. Japan, and Maghreb th iroduction of milk and meat is increasingly based on irrigated fodder croi.t;ing. However, conditions in the tropics for the moment are different. Arable cropping receives priority, as it always has, although it is usually supplemented by various livestock activities. In a few cases, irrigation farmers practise a division of labour with seminomadic herdsmen. The fallows of Gezira, in the Sudan. for instance. are grazed in the dry season by migrating herds of cattle, sheep, and goats. As a rule, however. we find stationary animal-keeping with low livestock numbers per holding but rather high livestock densities, rising to 2 livestock units per hectare in intensive irrigation areas of India and China (Buck 19K ', p. 255). The main objective in keeping cattle and buffaloes is the provision of traction power for ploughing and pumping. The provision of manur,- is a secondary function, and the production of meat and milk is usually of less importance. The need of the irrigation farmer's family for protein is not usually met from cattle and buffalo products. but by small animals, sheep, goats, pigs, poultry. and fish. Pigs are of particular importance in south-east Asia, being fed partly on crop resduts and partly on household waste. Ducks are typical of the Chinese type of holding. and chickens are found almost everywhere in the tropics. Part of the animal excreta may be used to fertilize fish-ponds, and mud from the pond may be applied to the gardens. Attention should also be drawn to forms of combined irrigation farming and fish production. Fish are found almost everywhere in the water distribution system. in many places, the fish-pond by the house is a typical feature of irrigated holdings. Hickling (1961) notes the integration of various forms o1" fish production into crop rotations: (1)In Travancore, southern India. after the rice harvest the plot is filled with brackish water, and in this way prawns are brought into the field. When the tide recedes, the water passes through nets. ana the prawns are collected. The yield amounts to 700-1600 kilograms of prawns per hectare in the dry season (p. 263). (2) In Java, Madagascar, and other places, fish are sometimes found in the growing rice. For this to be possible there must be a sufficient and continuous depth of water. Fish are brought to the plot in the irrigation water, or the, can be put in 5 days after planting. Hickling (1961, p. 265) quotes a yield of 28-50 kilograms of fish per hectare in 100 days. (3) In some cases, land is used alternately for rice cropping and as a fishpond. The water is let in after the rice harvest. The yield of fish that swim
154
SYSTEMS WITH ARABLE IRRIGATION FARMING 155 in with the water amounts to about 3 kilograms per hectare in five to six months. By introducing tilapia, up to 300 kilograms per hectare can be produced in 3 months. (4) In some plantations in Guyana, the sugar-cane land is covered with brackish water for nine months before the next planting, and the plot can be used for fish (tilapia). 7.3.6. Characteristics of the labour econom ' As we might expect, the labour input of irrigated holdings shows great variation, depending on the system of water distribution, the intensity of land use, and the degree of mechanization. Where labour is particularly plentiful, hoe and spad systems occur, relying entirely on manual labour. Ir most irrigation systems, farmers employ animal power for ploughing, transport, and pumping. The use c,f the tractor is rapidly spreading, for irrigation farming can support higher monetary inputs in cultivation than can rain-fed farming. Despite the varied forms of organization, however, some features and tendencies are found that distinguish the labour economy of irrigation farms from that of rain-fed farming. (a) High labour input. Irrigation farming in the tropics almost always involves a high :nput per hectare of manual labour. This is because of the extra operations in irrigation, such as maintenance of channels, control of water distribution, and land preparation. The manual labour input is particularly high where water has to be pumped by hand or with the aid of animals. In the example of rice production in Madras (Table 7.3), these tasks amounted to 48 per cent of the total. According to Buck (1930, p. 306), farmers in Hopeh Province, China, in 1923 spent on irrigation 58 per cent of the total labour input in wheat and 27 per cent of the total labour input in vegetables. In regions that depend mainly on inundation, the labour demand for irrigation iscomparatively low. However, the provision of controlled water supplies and of careful water delivery is Lusually connected with a high labour input. Of even greater importance is the labour input resulting from the choice of intensive crops, and the greater amount of labour involved in cultivation, weeding, and harvesting. The methods of producing and applying manure, compost, green manure. etc., require a great deal of labour, especially where multiple cropping, relay-planting, and intercropping play a major part. This kind of production method is not only expensive in terms of field labour: a considerable input is also needed to maintain the irrigation system and nursery beds. Thus the basic difference from rain-fed farming is not simply that more work is involved in cultivating each hectare. but also that irrigation often facilitates a higher frequency in cropping. Table 7.3 indicates the level of labour input for one crop of wet rice. With the exception of the example in the United States, it fluctuates between 456 man-hours per hectare on Madagascar and 1663 man-hours per hectare in Japan. and thus is
TABLE
7.3
United States of America 1960 5-7 0-5-" 05 n.a. 5-9
--
33
147
1
105 128 656 163 40 n.a. --
126 40 0-3 21
38d
* Sowing. * Including animal hours for water procurement. Cultivation by pitinage. d Two-wheel tractors. Thailand Minitry or Agri(1963, p. 17): Madras, Zacharias (1957. p. 9); Thailand, 2 Sources: Japan, Ozaki (1960, p. 13); Taiwan, Ch. Ch. Chang agricole 1965, p. 78); United States, Mullins(1960). production la de daveloppement le pour Bureau culture (1959, p. 19); Madagascar,
157 SYSTEMS WITH ARABLE IRRIGATION FARMING irrigation much higher than the labour requirement in rain-fed cropping. Within of reliability and availability increasing with rise to tends input farming, the labour labour in rise disproportionate a water. As multiple cropping increases, there is 4000input, and for example in intensive wet-rice systems on Taiwan it reaches 0 0005000 man-hours per hectare. We find the highest recorded labour input-1 type 15 000 man-hours per hectare-in irrigated vegetable holdings of the Chinese (see 7.4.1()). labour (b) High marginal returns on labour. The production functions of farming. in irrigation systems are, as a rule, different from those in rain-fed and lower usually are farming rain-fed in labour on The marginal returns irrigation decrease more rapidly with greater employment of labour than in irrigation in returns marginal inputs, farming. For a wide range of labour constant. farming decrease slowly or remain almost India, the This is shown in Fig. 7.6. In the irrigation areas of Uttar Pradesh in considerable for high remain (and higher much are labour marginal returns on marginal labour inputs) than in rain-fed farming in Andrha Pradesh, which shows in the growing wet-rice of functions production The returns on labour close to zero. holdings irrigated artificially in labour on returns marginal the show Philippines in those with a reliable supply of water in comparison with the marginal return that show clearly functions The rainfall. of impounding the on rely holdings that artificial reliable of help the with much more labour can be absorbed productively irrigation. More often than not, the improvement of water supplies and of farming labour than techniques on irrigated land shows higher marginal returns on facilities irrigation managed the constructior of new, but less adequately of *Thecapacity writes: 35) p. and irrigation plots. As to rice, Geertz (1963, most terraces to respond to loving care is amazing.' seasonal (c) Less pronounced peaks in labour demand. Farming systems with of irrigation suffer, like rain-fed farming, from seasonal underemployment irrigation large the available capacities of labour and traction power. In demand and the schemes with one major cash crop, the peak in labour (the provision of resulting problems of the employment of seasonal labour and so credit for their payment, the necessity of using expensive machinery, systems. rain-fcd diversified more in than on) may be even more pronounced of the In wet-rice systems with but one crop a year, the underemployment The proverbial. is animals draught labour force and the great number of of rest the During days. 40-70 animals work several hours a day for about but places, such other the time they graze on embankments, waste land, and are not required for work. of Thailand This is demonstrated in Fig. 7.7. Rize holdings in the central plain and planting for labour in demand with but one crop a year show a highly peaked the in crops irrigation other various and rice with cropping harvesting. Diversified a find still we but work, of northern zone of Thailand shows a better distribution Multiple March. aaid November between demand pronounced peak in labour almost even cropping in the middle rice region of Taiwan is connected with an
SYSTEMS WITH ARABLE IRRIGATION FARMING employment of the family labour. The peaks in labour demand are met by seasonal labour. Seasonal underemployment in irrigation farming may, however, be reduced or even abolished by an improvement in the water supplies, and wherever continuous cropping is practised employment is found throughout the year. Relay-planting helps to reduce labour peaks at harvest time and when the
158
E 1
--
c E 0.8 0'8"6
MZ.6
---
Marginal costs-
" '
-0
0'4
4 (2)
I Io
20
40
120
FIG. 7.6. Marginal returns on labour in irrigated and rain-fed farms. (a) (1) Marginal returns per labour month in irrigation areas in West Uttar Pradesh 1955-6. 63 farms; average farm size 4.7 ha; average labour input 9.3 months. (2) Marginal rmurrs _,:r labour month in unirrigated farming. Andrha Pradesh 1953-4. 72 farms; average fain. ,i-e 3.7 ha; average labour input 19.6 months. (b) (I) Marginal returns per man-day in 3: ificially irrigated wet-rice farming ip the Philippines 1964-5. 24 farms; average farm size 4 ha; average labour input 75 man-days. (2) Marginal returns per man-day in wet-rice farming with impounded rainfall in the Philippines 1964-5. 12 farms; average farm size 4 ha; average labour input 64 man-days. The coefficiants for labour in the production functions are significant. It should not be overlooked, however, that estimates of the marginal return of labour are not free from serious shert-comings (specification bias, multicollinearity, etc.) The functions in the figure are expected to show general tendencies only. From Agrawal and Foreman (1959), cited by Heady and Dillon (1961, p. 621); Barker and Soothipan (1968).
159 FARMING SYSTEMS W!TH ARABLE IRRIGATION can be another way of next planting takes place. Commercial stock-keeping labour. It is noticeable, morecreating regular employment for the available very well in this respect. over, that irrigated and rain-fed farming combine frequently do not coincide, The labour requirements of both kinds of holding with seasonal irrigaholdings in and the cultivation of rain-fed land, especially tion, can help to regulate the use of labour.
24
0 NJFMAMJJAS (C)
rice holdings in Thailand and on Taiwan. FiG. 7.7. Monthly distribution of work-days on rice. Total labour input: 1332 work-days Th".. nd central plain; monocropping with (a) with rice Thailand northern zone; div-rsified cropping at 6 hours per day on 47 ha. (b) multiple region; rice middle Taiwan (c) na. input: and other arable crops. Total labour work-days 501 input: labour Total crops. cropping with two rice crops and other arable and Ong (1964, p. 69); Taiwan at 10 hours per day nn 0.96 ha. From Kuithongkham Ministry of Agriculture (1959, Thailand 174-7); pp. Department of Agriculture (1966. tables 35, 39).
farming is bc'i :r labour planning. and in timing. The irrigator, depend on rainfall, which fluctuates in quantity In systems with conwater. however, usually knows when he will acquire be adjusted to the can trolled irrigation, irrigation dates and water quantities apparent in especially is This labour economy of the holding and vice versa. affects the essentially rainfall intensive wet-rice systems. Neither drought nor undertaken be can control production process, so that planting and weed
Another advantage of irrigation (d) Better planning of the labour econonv. The labour operations in rain-fed farming
160
even on rainy days. The timing of the labour input is dependent on dry weather only during the harvest. By regulating the water, levelling the land, maintaining a high degree of fertilization, and controlling diseases by the application of pesticides. the farmer has obtained almost complete control of all growth factors except natural catastrophes. Thus the irrigation farmer is not as dependent as the rain-fed farmer on his daily assessment of tile soil and plants. and on choosing the type of labour operations and when to terminate them. Instead, the production process is similar to that in manufacturing, which permits relatixely exact planning. 7.4. Eight case studies Tile features of irrigation farming that have been discussed vary in importance in the different irrigated land-use systems, and the following examples are chosen to illustrate this point. Examples of holdings with various arable crops in India, Pakistan, and Singapore illustrate how the irrigators adapt land use to the varying availability of water. The example of the Chinese holding concerns one of the most intensive forms of land type of vegc .',e use of all. Farm-management data taken from Pakistan, Thailand, and Taiwan supply information about the characteristics of wet-rice systems. In the example of the Gezira. attention is drawn to the special factors that are important for the organization of land use in large-scale irrigation projects. 7.4.1. Water a'ailabilit.1, and irrigation farming: fire case studies from India, West Pakistan, and Singapore
(a) A millet-roundnut holding in Bombal' State. The millet-groundnut holding in Bombay State (Table 7.4) consists of 4.5 hectares, of which Ii hectares are irrigated all the year round. The non-irrigated field isplanted in summer with pearl millet, after which it serves as fallow grazing. Permanent rain-fed farming over the centuries on poor and scarcely fertilized soils has led to the yields stagnating at a very low level. Only 2.8 quintals of millet per hectare are harvested. The farmer concentrates on the irrigated area, which supports two harvests a year and is sown with crops that produce relativeiy high yields, thus ensuring both saleable produce and food for personal consumption, including groundnuts, wheat, and chick-peas. Intensive vegetable cropping means that the irrigated plot has the character of a garden. The area cropped annually amounts in all to 120 per cent of the area available.
111 in January 19b7 (base: 1952-3 = 100). From Directorate of economics and statistics: Indian Agriculture in Brief. 9th edition, Delhi, 1968. p. 154. r Pearl millet. - Finger millet. i Traditlonal varieties. 6 Moot households receive some additional non-farm Income. and Madras, Daniel (1960); Uttar Pradesli, Agrawal (1957); Puniab, Budde (private communicaSources: Bombas tion).
'
of agricultural commoditie stood at 95 In January 1956 and at The all-India index number of wholesale Prices
TABLE 7.4 Farm-managementdata of irrigationfarms with various arablecrops in India and Pakistana
Location Year Rainfall (mm) Type of farming Method Number of farms Workers per firm, (ME) ME per 100 ha Size of holding (ha) irrigated in summer irrigated In winter Cropped area (ha) In summer (kharif) (ha)
Pearl millet Groundnuts Rice Cotton Tobacco Finger millet Fodder (maize. millet)
Punjab Lahore Multan 1963 150-250 Cotton-wheat Case study 4.0 80 5.0 60% 750 6'7
--
Uttar Pradesh 1955 750-850 Sugar-cane-wheat Case study 3'3 85 3.89 90% 1001!. 5.1
-
Madras 1955 900-1000 Tobacco-millet Case study I 4.2 344 1"22 100*%2'97 0-24
-
1"60
-
0'46 --
0.95 0'71
-
1'20
0'21 0'42
1.60 0"40
-
1.22 0.46
-
085
Fodder
013
0.33
0-40
120
-
112
185
-
0-12
-
0'10
Sugar-cane
Fodder Bananas
0'05 120% 11"0 4'0 2,0 2353 524 78% 1 17". I'l, 3.
3b
Cropping index Livestock per holding Animal units Draught animals Dairy cows or buffalo Farm capita ($per holding) F~arm capital: hai Land Buildings Livestock Tools Other capital items
Yields Millet IQ ha- ')
0'10
243% 4"5 2.0 2.0 2149 1761 47% 4". 7% 13% 29%
24
Groundnuts (Qha -1
Cotton (qha-l
18
-
-8
13 -
12 390
320
-
133'
41
b
16
146
-
120
54
98
-
115
310
-
578
-
325
321)
274
ELioialit rtliri (s per holding) Gross return 248 flame consumption 71". Expenses for purchased material inputs 40 Home consumption Income 208 Wages II Family income 197 Pruductliity Gross return ($ha" t) 1 Gross return ($ME ) Value added ($ha" t1 t Value added (0ME ) 55 125 41 104
SYSTEMS WITH ARABLE IRRIGATION FARMING Rain-fed farming and grazing contribute to irrigation farming in that they supply fodder for cattle and goats, while an) stable manure is applied to the irrigated field. four Thus stock-keeping helps to accumulate nutrients on the irrigated area. The higher the for tial essen is labour their and water, 1 drawing for work animals are used yields per unit area of the irrigated field. Milk cows, on the other hand, are kept animals, mainly because their bull calves are used as the next generation of draught the and thus cattle-keeping as a whole serves primarily to supply animal power. At season dry tile in that in animal-keeping, same time, the irrigated area also assists an even supply of fodder is guaranteed by planting small plots of fodder sorghum and lucerne. to Only a quarter of the gross yield is sold, and the primary aim of production is cover subsistence requirements. The family income amounts to S197, which corresponds approximately to the income of an Indian farm labourer in 1955. Irrigation in this case affords little more than a bare existence. even when measured against the generally poor conditions in Indian agriculture. (b) .4 ,itheat-cotton holding, in the Punjab. The wheat-cotton holding in the the Pakistani Punjab demonstrates that. by so\ ing twice a year, an increase in both sufficient, not is water 'he obtained. is cropping of extent the and frequency cent lies however, for cultivation of the total area twice a year. In summer, 40 per point of the from land, of area total the although cent. per 25 winter fallo%% and in is no view of channel creation and levelling, is irrigable. In these conditions, there menholding millet-groundnut the in is there as plot, emphasis on one particular the tioned above. Rotational farming, including half-yearly fallows, is applied on all as return high a produce that crops on much so land. The farmer concentrates not for. paid be to has which water, the of use best the make can that on crops Cotton, sorghum, and fodder maize (kharif crops) are cropped in summer. chickWinter cropping (rabi crops) takes the form mainly of wheat, oil-seed, and peas. Growing berseem (clover), and in January-February fodder maize, guarantees fodder rich in protein during the winter months. Double rotations are usual: wheat by in winter is followed by cotton or, if a half-year fallow period is introduced, 162 maize cr rape. Cattle are kept for two main purposes, namely work and milk production. Almost with the 7 hectares have to be cropped annually. and to prepare the wheat seedbed consequence. in kept are oxen draught Two labour. days' 5-10 desi-plough requires cotton, Milk is produced by buffalo cows. The stable manure is applied to the size of the of account On fertilizer. mineral with whereas wheat is usually manured per yields high comparatively the and cropping dottble under area irrigated the in than higher much are productivity hectare, the gross yields, income, and labour the Bombay example. Since 1963, when the data were collected, new wheat varieties have been introper duced, which, supported by some additional fertilizer, yield 20-30 quintals together hectare, even under average farming conditions. Higher yields of wheat, to a with an increase in the area of wheat at the expense of other rabi crops, led 1963. with compared situation farm-economic the of improvement remarkable (c) A sugar-cane-iheat holding in Uttar Pradesh. In the sugar-cane-wheat sugarholding in Uttar Pradesh, the ample and regular availability of water allows are cane, an intensive cash crop, to be cultivated. Household food requirements met by wheat and rice. Holdings of this type already have considerable latitude in forming crop rotations. Two-year and three-year crop rotations are practised, encompassing three, four, or five crops. The following are examples:
FARMING
163
sugar-cane-cotton-arable fodder; sugar-cane-arable fodder-chick-peas Three-year rotations: maize-wheat/chick-peas-cotton-arable fodder-sugar-cane; sugar-cane-ratoon-pearl millet-fallow. which is In our example, almost a third of the arable land is planted to sugar-cane, hectare. per manure stable of tons 20-30 to up with fertilized intensively Extensive cropping of fodder legumes, particularly berseem, helps to guarantee indirectly the yield of sugar-cane, directly through the effect of the rotation and labour through animal feed and the resulting stable manure. The wages for seasonal income family a holding the from 1955 In expenditure. the of constitute a large part introof $355 remaincd. Since 1955 new varieties and mineral fertilizer have been sixties. the in prices duced, supported by rising in (d) .4tobacco-millet hohtin', in Madhras State. In the tobacco-millet holding The land. arable of hectares 1.22 on Madras, 4-2 man-equivalents are employed a high degree. possibility of intensilication that irrigation offers is being exploited to of the land cent per 243 is which The annual cropped area covers almost 3 hectares. crops that are they and annually, crops thrce or two available. The land supports demand a high lexel of labour and fertilization. The stress is on the cash crop, finger millet, tobacco. HLouehold food requirements are met by the cropping of are used for plots Smaller hectare. per yields high remarkably produces which two dairy and animals draught two keeps arable fodder and bananas. The holding is based kept cows of number The pastures. communal on gra/.e animals, w.hich is still area this in cox,-keeping only partly on economic considerations, because tradition. religious of partly a question High cropping intensity involkes a high degree of fertilization, and in holdings tnat is 22 of this nature. 6 tons of stable manure per animal unit can be relied on, the acdition, In hectares. 1-22 and cattle of head 4-5 with annually hectare tons per mineral and bone-meal, oil-cake. manure, farmers make compost and buy animal two draught fertilizer. Another dominating itern of expenditure is the keeping of are higher animals draught oxen on 1-22 hectares. The estimated costs of keeping levels. wage local of terms in measured labour, manual than the expenditure on 3 hectares but the farmer cannot manage withot draught animals, since practically have to be cultivated and prosided ,,ith water. two or three On this holding, the high income from cropping each area of land advantage full that fact The land. of times a year largely compensates for the scarcity that means irrigation through intensilication for possibilities is being taken of the of area small a having of the farmer succeeds in making up for the disadvantage land. hIohlin, in Singapore. There is hardly a more Chinese tipe of Ue47et(, "le (e) -4 world than the Chinese vegetable producer, the intensive farmer or gardener in system of market gardening to the humid Chinese subtropical the adapted who has a holding tropics of south-east Asia. Table 7.5 shows the farm-management data of occupy gardens vegetable The brothers. three by run Singapore. of 1-21 hectares in land is the flat land of the holding, covering not more than 0.6 hectares. The garden These ponds. man-made live are beds the among divided into 127 beds. Dispersed growing not only provide water for the vegetables, but also function as sites for stocked '.ater hyacinth, an important source of pig fodder. In addition, they are dwellings by occupied is slope hill the and gardens the between ground The with fish. and farm buildings. The hillside of the holding is left unutilized, except for some fruit trees. mainly The bulk of the ,egetables produced are of the quick-growing, leafy type,
164
Vegetable
1965 2500 Case study 30 10 5 826 1.21 0.60 4.80 800%
300
Poultry (no.) Economic return (S per holding) Gross return from cropping 3 559
440
Gross return from animal husbandry Total gross return Less purchased material inputs Concentrates Fertilizer Other means of production
Income
Productirii),
Source: Compiled from Ng Kay Fong, Tan Chee Lian, and R. Wikkramatileke (1966).
This publication does not contain a farm-management balance sheet. The figures in this table in some instances had to be derived from various data, and they give only a rough idea of the economic state of the holding.
Brassicaceae. Most of them are grown in stages-first in the nursery beds, and later in the main production beds. The vegetables usually occupy the production beds for 4-6 weeks. In the afternoon or the morning following the harvest, the beds are planted again. The fallow beds amount normally to not more than 5-10 per cent of the total. The average number of crops per year and per bed hovers around eight. Fig. 7.8, taken from farms in Hong Kong, shows the sequence of cropping that is typical for this type of farming. The rainfall, amounting to 2500 millimetres annually, is too irregular and inadequate to rely on. The watering of the vegetables is roughly equivalent to an additional rainfall of 1400 millimetres per year. The land provides hardly more than the physical medium for plant roots. The gardens are fertilized with both farm-produced and purchased fertilizers, including diluted pig manure, chicken droppings, mud from the ponds, compost, prawn dust (a product of the shrimp industry), and mineral fertilizer. Night soil is traditionally used in the farming system, but is no longer allowed for vegetable production,
165 FARMING SYSTEMS WITH ARABLE IRRIGATION which lend themselves to a The rearing of pigs and poultry, livestock activitiessystem. The pigs are fed on high turnover, is an important part of the farming waste. Apart from supplying concentrates and on a mash prepared from vegetable important source of manure. Most much of the income, livestock also provides an fertilizers, but as feeding stuff. purchased as of the nutrients are not imported gardener to cultivate Intensive techniques thus enable the Chinese market fcrtility is low, rainfall soil natural where vegetables successfully in an cnvironr.i.)nt often too intense, radiation solar inadequate, often irregular, natural soil aeration and harmful organisms numerous.
Pattern Apr. May June July Aug
_. '
Sept,
.
Oct
.
7:7,
Nov.
. . .
Oec.
.=
Jan.
>7
?eb.
_.
Mar
L -
..
:3 4
--------
U ]Chinese radish
Leaf iniustard ], ' Lettuce Chinese spinach Flowering cabbage W1 Maitrirnoiy vine
9 : ::: Splircdl
Cj
Kong farms Itfrrn Wong 1968). F)G. 7.8. Multiple-cropping system on some long
show the infltence (f) Conclusions. The examples f'rom India and Pakistan organization of holdings. that differing levels of water availability have on the from the management In the millet-groundnut holding. the operative factor and non-irrigated tields. point of view is the interrelation between irrigated production through double In the second example it is the expansion of cash a highly priced cash crop. cropping, and in the third it is the growing of example illustrates the which takes a lot of water and fertilizer. The fourth climate, water. good production possibilities of small plots when a tropical fifth example, the the in soil, and plentiful fertilizing are combined. Finally, supply of unlimited an Chinese type of vegetable holding, ample rainfall, hands of the in cultivation irrigation water, tropical heat, close markets. and decisive the time, same particularly careful farmers all coincide. At the
166
FARMING
principle of intensive, tropical, irrigation farming becomes apparent: the concentration of production in a small area, which is worked with a high intensity. 7.4.2. Wet-rice systems: three case studiesfrom East Pakistan, Thailand, and Taiwan (a) Characteristics of wet-rice cultivation. In climates with heavy rainfall, rice is one of the usual crops in rain-fed farming, particularly among shifting cultivators. Much more important than this upland or dry rice is wet rice, or rice cropped on plots that are covered with water permanently, or at least for a lengthy period, during growth. The main types of wet-rice cultivation are as follows: (1) Real deep-water rice is grown, for example in Bengal, where seeding or planting takes place before the rainy season, and the growth of the rice keeps pace with the rising water. Along some coasts of south-east Asia there are harvests of floating rice, which is sown from a boat and develops rhizomes 6-7 metres long. (2) In rain-fed farming, rice is normally sown broadcast. This practice is still sometimes retained in irrigation farming, for example in Ceylon and some parts of India, when water is impounded in the paddy fields after the sowing. This is a procedure that we term the traditionalsoi-ing method. (3) In south-east Asia, rice is traditionally started in a nursery bed, and 6-8 weeks after sowing it is transplanted into the puddled mud of the field. The production is carried out without many purchased inputs. This method is called the traditionalsouth-east Asianimethod, but it is common also in Madagascar and is spreading in Africa. (4) The modernized transplanting method, which is largely influenced by the Japanesemethod of rice cultivation, is common in Taiwan, and is being introduced into African and other south-east Asian countries, and this is the most intensive method. It differs from the traditional method in its high input of purchased materials and in the special care taken with the nurseries and with cultivation in general. (5) Whereas the methods named above are based on the use of much manual and some animal labcur, the American method of sowing is characterized by a low manual labour input, extensive use of machinery, and high labour productivity. The soil is prepared with a tractor, and the rice is sown with drills or from aeroplanes and reaped with combine harvesters. Northern Australia and Surinam are tropical areas where this type of cultivation is practised. Weed killer.,, which reduce the need for nursery techniques, are beginning to make modern sowing techniques attractive in some low-income countries also. No other agricultural product in the tropics and subtropics has anything like the same economic importance as wet rice. It constitutes the basis of the
167
economy in most Asian countries, and it is expanding in Africa. The expansion of wet-rice cultivation in the tropics and subtropics is based principally on the following facts: (1) It permits constant land use, even for centuries, without decline in yields. (2) It produces hig,, certain yields per hectare and consequently allows the concentration of population. (3) It enjoys a relatively favoured position within traditional agriculture: mannfactured inputs like mineral fertilizer, herbicides, and implements make wet-rice cropping more productive, but even the rational use of the traditional production factors of soil, work, and water can produce a relatively high yield. The really important phenomenon is the permanence of rice cropping. Increased population densities in traditional rain-fed systems result sooner or later in a deterioration of the habitat. In a wet-rice region the habitat remains undamaged. Some areas of Java. for example, support 2000 persons per square kilometre, without any significant decline in rice production per hectare (Geertz 1963, p. 33). Many of the famous terraced slopes in the Philippines have been planted to rice annually for more than 2000 years. Wet rice fosters, therefore, economic stability, a dense population, and an advanced civilization. It forms the technical basis for the formation of permanent Kulturlandschafteu' in the wet valley bottoms, which contrast noticeably with the shifting cultivation of the adjoining forest landscapes. Experience in tropical Asia shows that these two types of laad use represent the two forms of arable farming that are most suited to the productioV conditions of a warm, damp climate. On the one hand there is irrigated rice culture, with high population densities, permanent housing, high labour inputs, and high yields per hectare, conquering the difficulties nature presents by conscious manipulation of the countryside. On the other hand there is shifting cultivation, with low population densities, semni-nomadic populations, low labour input, a low yield per hectare of the total area. and adaptation to nature. In order to pract;-e balanced cultivation where there is a shortage of land it is often essential to make the transition from the most extensive system of land use to one of the most intensive systems. and, in addition, to concentrate production in the valleys and plains. The advantage of this transition is not really that there is a higher average return on the labour input. The point is rather that, whereas intensified shifting cultivation is accompanied by rapidly sinking marginal returns from extra labour input, intensified irrigated rice, at least up to a high level of yield, seems to have almost constant marginal returns from extra labour. Low average but comparatively high marginal
Landscapes strongly influenced by human cfforts to make them useful Ibr man.
168
returns characterize the farm-management situation in rice holdings, wherever the water su nly is sufficient and controlled. Irrigated rice culture is thus in a position to absorb productively much more labour. At the same time it permits food prou:iction to be concentrated in narrowly circumscribed areas, and, vith comparatively short transport routes and the concentration of population, the formation of towns can occur. In this N\ay, irrigated rice has laid the foundation for Asian culture. Rice can be adapted to suit most types of agronomic conditions. It thrives in the mountains and at sea level, in the tropics and in the warmer regions of the temperate zones, on good and on poor soils (with differences in yield of course), in the desert and where there is a heavy rainfall, provided that heat and water are available to a sufficient degree. Irrigated rice is not as strongly affected by wateilogging. floods, and stlty soils as other crops. By the cropping of rice, areas that would otherwise be waste land can be opened up for agricul'ural production. In fact, as long as there is a suitable supply of water, some rice can be produced on soils that are too infertile to produce any other grain crop (Jordan 1967, p. 3). Rice grows well on its own-and monocropping is tnerefore possible-but it can also be included in crop rotations without any difficulty. The various labour operations, like preparation of the soil. planting. and weeding. are largely independent of the weather, provided there is enough water. There are many varieties, with different growing times (100-200 days). differing demands on soil and climate, differing abilities to benefit from fertilizer, and a range of other characteristics. In handling and transporting, rice shares the advantage of all cereals over the root crops, while in storage it suffers smaller losses than practically any other tropical food crop. Raw paddy is resistant to insect attack in storage. and parboiled milled rice is similarly resistant. Another advantage of rice as a food is its ease of preparation, and its low requirement of fuel in comparison with bread. Rice also has a high nutritive value, and its protein is of an exceptionally high quality and digestibility for a vegetable protein, while it is also exceedingly palatable (Jordan 1967, p. 5). Furthermore, wet-rice production can be fitted into most social and economic structures. Rice cropping is practised in large and small holdings. in countries with high and low wage levels, almost without implements or with a very high investment in machinery. In short, irrigated rice growing is more adaplable than practically any other type of agricultural production. Rice is also an ideal crop for a rapidly developing country, as the technique for its cultivation can be made to keep in line with the development of the country as a whole, and its extent is unlikely to be restricted due to rising economic levels, as may happen v.'ith other crops (Jordan 1967, pp. 4-5). It is this adaptability, not only to agronomic conditions but also to social and economic conditions, that makes wet rice so extremely attractive as a crop.
SYSTEMS WITH ARABLE IRRIGATION FARMING 169 The basic features in the organization of various types of rice holding can be demonstrated with a few examples taken from rice-jute holdings with an unregulated water supply in Bengal, monocrop rice holdings in Thailand, and a holding with multiple cropping in Taiwan. Table 7.6 contains farmmanagement data of rice holdings in Japan and the United States, so that comparisons may be drawn.
(b) Rice-jute holdings in East and West Bengal. In the Ganges Delta in
East Pakistan, 1500-1700 millimetres of rain fall insummer on relatively fertile soils. At the beginning of the rainy season the water accumulates in the fields, supplemented by flood water from the rivers. At the end of the rainy season the land is covered with water up to one metre in depth. Thus nature supplies water free, and with it silt and nutrients are introduced. But nature does not supply the water regularly and in optimum amounts: sometimes there is too much water and sometimes too little. Consequently production conditions have been developed that in many respects combine the features of irrigation farming and rain-fed farming, and can therefore be regarded as an intermediate stage between these two forms of land use. Rice and jute are the main crops. The typical East Bengal holding that is illusminimize risk. In the rainy period, the whole plot is cultivated. 15 hectares are used for various formns of rice cropping, and 0.3 hectares for jute. Directly after the rainy season, a part of the arable land iscropped with iegumes, wheat, and oil-seed. Total cropping amounts to 3.0 hectares. or 150 per cent of the available area. A rice-jute holding normally includes four head of cattle, of which two are draught oxen. The fodder basis zonsists of rice straw, fallow grazing, and roadside grazing. There are scarcely any pastures. Stable manure is used as fuel, and if any is left over it is applied to the jute. In the case of the rice farmers of Bengal-and similar observations can be made elsewhere in wet-rice farming-various measures are taken in order to adapt production to an unreliable water supply:
(1)The farm land consists preferably not of one block but of numerous
trated in Table 7.6 comprises 2.3 hectares of arable land, divided into 22 plots to
plots on different soil types and at different altitudes. Some of them are expected to yield reliable returns. In the case study, the average farm has 22 plots. (2) Various varieties are grown. In the area of the case study there are 25 different rice varieties. The higher the risk of drought on a given plot the greater the prefere-nce for quickly maturing varieties (aus). The higher the risk of flooding, the greater the preference for varieties with long stalks (aman). There are also distinct varieties for each season. (3) The higher tFhe risk, the greater the preference for seeding rice instead of transplanting. Most of the rice in the area of the case study is seeded. (4) Husbandry practices in nursery establishments are tuned to the reliability of water supplies, i.e. they are the more intensi,,e the higher the
TABLE
7.6
Farm-naitqqementdata of ricefarms in East Bengqal (Paki.tan). JI'est Benir'l (India). Thailand, Tai-an, Japan, and the U"nited State's
Location Year Method Number of holdings Workers per holding (ME) ME per 100 ha Size of holding (ha) Area cultivated (ha) ist crop Rice Jute 2nd crop
Rice
East Bengal 1960 Sample n.a. n.a. n.a. 2-3 2-0 1-5 0-3
--
West Bengal 1964-5 Case study I 3.0 150 2.0 2-0 i-i 0-9
United States
1959 Sample 20 6 1 (A) 481)
1-0
-0 0-I 0-5 04 3-0 273% 3-4 170I
-
I.S
480
03 0-6 -
0-3 07
--
--
i-I 05
-
Vegetables Biennial crop Sugar-cane Area cropped (ha) Cropping index Livestock per holding Draught cattle Dairy cows
3-0 150%
. 4-5 100./ 2
-
480 80%
Pigs
Chickens
2
n.a.
n.a.
2 n.a. 2 18
2
-
n.a. n.a.
30
66 % 3% 17 %
n.a.
7 n.a.
265
-
79
n.a.
350
25 -
1034
38 3232
286
-
Total gross return Rice Household consumption Less purchased material inputs Fertilizers Fodder Equipment Income Less wages Less taxes and rental payments
Fziuiiy *,ncome Productivily
139 n.a. 57* n.a. n.a. n.a. n.a. n.a. n.a. n.a.
n.a. 60 n.a. ,1.a. n.a.
Cross return ($ ha') Gross return (S ME ') Value added ($ ha ') Value added ($ ME- )
"Tax value, below market value. Sources: Kung (1962); West Bengal Government (1968, p. 93); Ozbileni (1962-3); Thailand Ministry of Agriculture (1959); Japan Ministry of Agriculture (1960); Andreae (1964). United States Thailand Taiwan Japan East Bengal West Bengal 1960 1959 1964-5 1955-6 1960 1960
172
FARMING
chances are that sullicient water will be available when seedlings of an optimum ,ize are available lbr transplanting. (5) Transplanting occurs only after heavy rainfall indicates that the monsoon has definitely started. This may mean a loss of time and yield in average \'ears, but risks to production are lower. (6) Mineral fertilizer is often not applied at the times that would be optimum with reliable \wttcr supplies. Many farmers wait until rainfall has been ample enough to indicate that sufficient water will be available. (This can be obser\ed particularly in areas with unreliable monsoons, as for instance in Madras Stat,.) (7) Rice farmers with unreliab: x\ater supplies tend to supplement food production by growing crops that yield less per hectare but show a high yield reliability, such as various millets. (8) Rice ser\es principally to supply food, whereas jute is the principal commercial crop. It is sensitive to waterloging in its early stages, and it competes with rice less for the lower fields than for the higher ones, where water builds up only when the rainy season is well under way. Jute requires more manual labour (approximately 150 work-days of 8 hours per hectare) than sown rice (approximatey 75 work-days of 8 hours per hectare). Harvesting is particularly laborious, for in the months of peak rainfall the jute. which is standing in deep water, is cut. tied up in bundles. and left to ret. The rice-jute holdings of East Bengal have considerable production potential. owing to the fertility of the soil and the ample water suppl). but development of this potential meets with a number of obstacles: (I The irregularity of the water supply is of overwhelming importance, and damage due to floodin,- as well as drought is frequent. Bengal has a long history of famine. The amount of water required to guarantee yields in irrigation farning is often not available. (2) As well as the uncertain availability of water, there is the high cost of draught power. According to Ozbilen's investigation (1962-3), there are 133 draught oxen per 100 hectares of arable land, and they burden the holdings disproportionately. because of the fodder that has to be purchased or the amount of fallow grazing required. In other words, provision of feed for the cattle and the expansion of double cropping compete for land. On the other hand. this type of land-use system must have effective traction power, because it is important to make full use of the best planting dates as determined by the rainfall. (3) The second annual planting is difficult, because the water supply is limited to a few months and in the dry season there is no irrigating at all. The second crops must manage with the moisture remaining in the soil after the lirsi harvest. These catch crops are predominantly legumes, which
173
have a short vegetative period and low yields After a short monsoon there is scarcely any point in a second crop; it is only a practicable proposition in years with a long monsoon and on soils that are able to retain moisture. The jute-rice holding in West Bengal, which contiasts with that in East Bengal (see Table 7.6), produces under almost identical conditions of soil and climate. It has, however, the advantage of a regular supply of water and is less exposed to flooding. The markets of Calcutta are close by and prices are relatively high. Land use is therefore much more intensive; in particular, yields re higher, thanks to the combined effects of new rice varieties, mineral fertilizer, and controlled irrigation. (c) Rice holdings in Thailand. In the fertile, comparatively thinly populated rice-growing areas of Burma and Thailand, rice is planted in small fields, the paddies, which are seldom larger than 0.5 hectares, and are irrigated by water drawn from the ri'er during the rainy season. They belong to relatively larger export-orientated holdings than in Pakistan. The conditions we find in Nakom-Phatom Province in Thailand illustrate this type of farmirg. Whereas in the densely populated regions of India or Java rarely more than 2 hectares per holding are used for rice, ve tind here 4.5 hectares of rice. This area supports rice once a year, year after year, and has done so for centuries. After the rice harvest, the soil quickly dries out, hardens, and lies fallow until the next planting, during which time it serves as grazing. This kind of holding has, apart from the rice fields, a fair-sized garden, work cattle, and ,ome poultry. The fact that a couple of pigs are kept indicates the similarity to the Chinese type of rice holding. %%herepig-keeping iscustomary. Since harvesting takes place only once a year. slow-maturing varieties. which produce a comparatively high yield. are preflerred. The method of cultivation shows the basic features of the traditional southern Asian rice culture, namely: (I) The raising of seedlings in nursery beds. (2) Levelling and preparation of the wvater-co ered soil with draught animals, plough, and harrow (8-12 ploughings with iron-shod ploughs. usually not deeper than 3-4 inches). (3) Transplanting about two months after germination. (4) A reasonably well-regulated water supply (in some holdings in the region of this example, water is already conveyed by pumps). (5) Harvesting of the cars with a knife or sickle, the straw being left for the most part on the lields. (6) Transportation of the harvest to the farm on dry land by carrying, by sledge. or by bicycle. (7) Threshing by hand or by the trampling of animals. (8) Drying of the rice in the sun.
174
SYSTEMS WITH
ARABLE IRRIGATION
FARMING
As the water supply is regulated effectively, the planting method is preferred to the sowing method, and there are indeed obvious advantages: g:owing the seedlings in nurseries saves seed, water, and time. since only relatively small areas (a tenth of the cropping area) need to be irrigated inl the first two months of growth. The farmers have these two months to prepare the soil and suppress the %%eeds in the paddies. Morco~cr. transplanting produces yields higher than those of sown rice. while the certainty of the yield isalso greater. and the crop ripens more eenlx. \lodern. intensi\e ricegrowing methods are. howe\er. not vet applied. Neither oruanic nor inoreanic fertilizer iscommon: people rely on"the nutrients brought by the water. Tihey de not plant in rows. neither is it customary to combat wkeed growth systematically. Between planting and har\esting, the control of the water stpply and the scaring labour input is limited to the of birds from the ripening field. The yields, at about 1[7 tons of paddy per hectare, are scarcely hal'of \\hat is normally obtained in Taisan.
The main economnic feature ofi his t'ype of rice production isthat. despite Il)%\ rice prices, a comparatively high iamilh income are found in countries that are dependent onis atiained. Rice holdines of this kind the export of rice. and %hose farmers consequently receive a lok price as producers. In Thailand, this is only S0.05 per kilogram of paddy, whereas in Japan it is SO18 per kilogram and in the United States SO 10. The holding input isin consequence limited to land and labour. In the holdings in the example sho\% in Table 7.6. only 3 per cent of the holding capital is allotted to equipment. but 86 per cent is allotted to the soil. The purchased inputs per holding are kept to .>36. and of this sum practically half isused for purchasing fodder, mainly for the iraction atinials. Because of the soil fertility. and despite the low prices and the small expenditure on production means, a gross return of S505 is achieved. A large proportion of the output, keted. The family income amounts to .'390. approximately 75 per cent. is marwhen non-agricultural earnings and
payment of rent are taken into accOunt (half the holding area isrented land). This income iscomparativel,, high. particularly
"her' %%econsider that the cost of li\ ing is relatively low on account of the loxk price of rice.
The economic problem in tl: type of holding iscaused by the fact that the labour and soil potential are insufficiently exploited. The family has at its disposal 3.19 man-equi\alents and. with 250 %%ork-davs a year. a total of 800 work-days. The holding. hoN\ever. uses only 362 "ork-days annually. Planting and harvesting constitute the work peaks, and in the intervenin, time the farmers have little to do. In addition to the need for improvement in the method of rice cropping, therefore, there is also the question of full employment of the aailable labour and land by double cropping. Because the soil hardens so quickly, the chances of sowing a second time after the rice harvest are slight. One solution, already adopted in some cases, lies in additional irrigation with the help of pumps during the dry season. (d) A rice farm with multiple crol)piiq in Taichung District. Taiwan. The relatively small rice holdings of Taiwan are on an island with a long-term
to new labourers and farmers, easy access land shortage, skilled and efficient in the are, power. In addition there technologies, and rising purchasing systems warm winters, and irrigation Taichung District, fertile soils, relatively methods of rice production have with effective water control. Consequently, proproductivity per hectare with high been developed that combine high ductivity per man-hour. in Tai.van, which benefited from The highly intensive rice-growing method distinguished by: been Japanese experience, has traditionally soil, animal manure, fish. oil-cake, (1) intensive organic fertilizing (night green manure, ash, etc.); the water supply is sufficient; (2) two rice harvests a year where and in more recent times by: applications of fertilizer; (3) varieties that are suited to heavy in the nursery, at transplanting, and (4) a high level of mineral fertilizing during growth: (5) treatment of the seed with fungicides: herbicides (6) the employment of pesticides and sowing; after weeks three (7) early transplanting and in preparing the field: techniques (8)1 particular care in nursery the to south to take full advantage of (9) planting of rows from north sunshine; rows and between plants: (10) optimum distances between (11) the use of improved implements: (four times in forty days): (12) several hoeing and ridging operationswater anl Ja high dose of water of (13) the alternation between drainage during the growth period, and by means of small, stationary threshing (14) the threshing of the harvest machines. depicted in type is practised in the holding the island. for Intensive rice production of this 1.08 average hectares is about of Taichung the for Table 7.6. The total farm area istypical which 273, figure of potatoes, sweet Multiple cropping reaches the index and a winter crop (wheat. cabbage, cropped area District. Spring rice. summer rice, total The 7.5). each plot (see Plate activities. seed rape) are usually grown on livestock In addition, we find poultry and bufLalo is kept for thus amounts to 3.00 hectares. of the total gross returns. One which contribute about a quarter and concentrates are the imtoortant purchased and per cultivation work. Miner.l fertilizer 200 kg K-O are applied per hectare inputs. 320 kg N, 200 kg P20 5 , and of liveof each crop, and a high level and per Multiple cropping, high yields per hectare hectare per return behind the high less much are stock production are the key factors low prices for the products, which holding, and this in spite of the to 5121 1. of which 17 per cent isconsumed amounts than inJapan. The gross return of the costs of purchased inputs and hired labour. inthe household. After deduction there remains a family income of S552.
FARMING
175
176
For comparison, Table 7.6 includes farm-management data for Japanese and American rice farms. Japanese rice farms differ increasingly from Taiwanese farms because of the introduction of tractor ploughing (single-axle tractors), mechanized harvesting, increased fodder growing, and dairy production. Where dairy animals are kept, the fertilizer economy acquires a different character. The supply of nutrients through mineral fertilizer is replaced more and more by fodder concentrates, whose nutrients reach the soil as stable manure. The return per hectare and per man-equivalent in the Japanese case study is even higher than in Taiwan. This, however, is mainly owing to the higher prices to the producer. The American rice holdings show that mechanized rice production combines very high labour productivity with a medium level of returns per hectare. 7.4.3. Farming systems in large irrigation schemes holder settlements have been increasingly established in areas that are being opened up for arable farming in large-scale irrigation projects. It is a characteristic of large-scale irrigation developments that there is a great measure of physical interdependence between different users of the water, and, therefore, to achieve maximum benefit, the resources must be planned and operated as part of an integrated system (Carruthers 1968, p. 7). In several countries, large-scale enterprises have been created by the government and are operated by government-backed agencies. The farm-management position of smallholders in these irrigation schemes is different from their position in holdings that obtain water from minor irrigation works. The minor schemes have been set up in the course of time not so much by the use of capital as by work on the part of the local population. For this reason, they rartly entail high fixed costs. Large irrigation schemes, on the other hand, necessarily involve high fixed costs, and this means high productivity and high water charges. The farmers, who are not as a rule highly experienced in modern irrigation farming, are not in a position to realize the production potential within a reasonably short period after the water has been made available. The answer to the problem is production under close supervision, that is, the obligatory application of modern techniques by all water users. The state, which has the right to distribute land to settlers and to grant water rights, has a 'power key' by which it can introduce better techniques. This is selective pressure in the sense that it is used only on those smallholders who are interested in becoming irrigatioia settlers. They must agree to abide by a series of principles that guarantee technically and economically efficient production. At the same time, participants may retain ownership of the means of production that they possessed previously, or which have come into their possession, with the proviso that they may not dispose of this property just as they please.
(a) Characteristicsof production under close supervision. Recently, small-
177 SYSTEMS WITH ARABLE IRRIGATION FARMING the scheme They are dependent in this respect on the rules laid down by management. usually not in In large irrigation schemes the management functions are agency and scheme the hands of one man, but partly in the hands of the partly in those of the cultivators. He sees that (1) The project manager runs the framework of the scheme. the acquisiproduction is carried on in an organized fashion, he arranges of purchased lion of water, he organizes tractor ploughing, the supply benevolent but strict a ef that is function His inputs, and the marketing. is hardly management Cooperative and technically progressive landlord. supervising and water distributing of task feasible in irrigation schemes. The are the repreproduction is too delicate to be left to a group of people who irrigation large of running sentatives of the irrigators. The successful (see authority centralized schemes obviously requires a high degree of Wittfogel 1931). the rules, (2) The smallholders do the work delegated to their plot. Within and organized be which are laid down by the project agLncy and can the and knowledge enforced on a broad or narrow basis according to the own discretion. drive of the irrigators, the smallholders operate at their of independent than sharecroppers of that to Their position is often closer tenants. farmers. Frequently they are in certain (3) The irrigators can. if necessary, join forces and cooperate The produce. theix selling or credit undertakings, as for example raising manageproject the by activities of the cooperative are again supervized ment. 20 2 ) give us an idea of the returns The details supplied by Giglioli (1965, p. successful irrigation of production under close supervision in one of the most in Kenya. He writes: schemes in Africa, the Mwea-Tebere irrigation scheme Only one crop of Tenants are allocated four one-acre plots as a rice holding. ... The rules practised is transplanted rice is grown per year... No rotation relating to matters on detailed particularly ... and are %,crycomprehensive ... the crop. of disposal and cropping practices, water control, absentee ownership, are. in These recalcitrants. with deal to tools Management has three disciplinary of termination and prosecution warnings, written severity: order of increasing the and disinherited, the to limited licence ... Selection (of tenants) has been the Settlement the Clan Committee have a very human tendency to unload onto in the three years less desirable members of their community ... (Nevertheless) have had their that the Irrigation Rules have been in force only 19 tenants unit ... is a licences terminated (Mwca section only). The basic organizational Assistant The ... families 600 supporting block of 2400 acres of irrigated rice, to day day The ... b!ock the on activities all of charge in is Agricultural Officer responsible is man Each Assistants. supervision ... is in the hands of the Field Settlement staff. A for 150 tenants ... All water control is in the hands of the in this field ... regime water tenant is not allowed to interfere in any way with the .. The execution. for responsible are Guards the Head Water Guard and his
13
178
last link in the chain between the Manager and the tenants is constituted by the
Head Cultivator . a tenant selected for his farming and leadership qualities...
The introduction of mechanical cultivation has made possible a degree of planning and discipline and extension unthought of in the past. The orderly progression of cultural operations made possible by mechanization of puddling has produced an atmosphere in which strict discipline can be enforced without
opposition
. .
beginning of each stage of the cultivation and to indoctrinate them on the measures. Repetitive indoctrination made possible by mechanization is showing its effectiveness by both the increase in yield and the even distribution of the increase over all the tenants on the Settlement. Irrigation farming under close supervision combines a number of advantages that characterize small-scale and large-scale farming. The stimulus for efficient production, which is generated by the direct relationship between the individual's efforts and the returns of the land, is maintained. Every irrigator receives an income according to the yield of his land; since the quality and amount of land and other inputs provided for each settler are the same, the yield is mainly a direct reflection of the amount and skill of the labour input. Large-scale operations, such as ploughing or spraying with pesticides, can be organized by the management on a large-field basis. The plots of the irrigators can be arranged in such a way that large machinery can be used efficiently. Innovations like new varieties or crops can easily be introduced by the scheme management. (b) A case study: the Gezira scheme, in the Sudan. The biggest example of large-scale irrigation farming under close supervision is the Gezira scheme in the Sudan. In the project area, which including extensions comprises 756 000 hectares and 73 000 tenants, the land was leased or bought from its owners, reorganized, and de eloped for cropping with cotton, sorghum, and various fodder legumes. The division of operations between the government, the scheme management and the tenants is well defined. The government has four main responsibilities: (1) It buys or leases the land from its owners and pays small rents. (2) It builds and maintains dams and the primary and secondary distribution and drainage channels. (3) It finances the acquisition of water and land development. (4) Finally, it establishes the scheme management as an autonomous body -the Gezira Board. The scheme management is responsible for the following functions: (I) It distributes the land among the leaseholders and enters into lease contracts with them. In the original Gezira area, the smallest irrigation unit is a rectangle of 1420 metres by 292 metres. Each of these blocks is divided up into nine plots, each of 4.22 hectares. Each leaseholder receives one plot of land in four blocks, giving 16.8 hectares altogether. Further
179
subdivisions and amalgamations are permissible. The holdings in the Manatgil extension are laid out on a smaller scale (6.3 hectares), because the original holding size was too great for the tenants' labour force to manage.
Block A
Plot 1 11,101 LdIll, ..l1Jd Plot 2 Plot 3 Plot 1
watel
I.
Block
Plot:
8
Plot 3
aSmall
s'I
A)
:"
6 t/A
Aia_
A2
~AA
B3A3
B2 B)
0.i35
OEI
@6~.S
Au
A5
B6
.7
A7
A,,
A s
.Suply
c, nlel
held track
Lubij
1Fjio'
100
250m1
of the 2.1 ha 2.1 ha per hoAlingl. AI and 3l ire the cultivators of the first strip, each farming 33). p. (1966, Worz From lubia. ha 1'05 and lih.'r,12. ,, cotton. I o; ha (2) Land Use is not open to the settler but prescribed by management. The
Managil extension FIG. 7.9. Or"gan1/iztion of water delivery, land use, and field layout in the (three plots of
Gezira scheme, Sudan. No blocks, each with nine holdings are shown
usual crop rotation. embracing eight years, is as follows: cotton-fallow'allow-cotton-lhallow-sorghum-lubia( Doliclios lablal)-fallow. In the Manag.il extension. a six-field crop rotation was introduced: cottonfallow-cotton-lubia-sor-LhLfm-fallow. Fig. 7.9 shows the plot division in this rotation and illustrates the connection of smallholder farming with large-field management. Each field is 2.1 hectares in size and forms. with seven others, a unit of 16.8 hectares (see Table 7.7). Each tenant uses his plots in obligatory rotation. Thus. for example, tenant Al uses a cotton field of 2.1 hectares. half a sorghum field, and half a lubia field. In the
180
IRRIGATION FARMING
following year his cropping isadjusted according to the rotation, but only within the strip that he has been allotted together with the tenant BI. (3) It organizes the water supply, maintains the canal systems, and levels the land. The date when the waier is supplied and the amount are determined by the Board. Furrow irrigation is supplied twelve to fifteen times
TABLE 7.7 Farm-management data of a tenant holding in the Gezira, the Sudan (1949) Size of farm (ha) Irrigable Cotton (ha) S'rghum (ha) Lubia (fodder) (ha) Fallow (ha) Cropping index Yields Cotton (kg ha-) Sorghum (kg ha-2 ) In holdings worked with wage workers Economic retur', ($ per holding) Gross return' 881 168 1600 42 2-1 2.1 84 50% 1400 2000 In holdings worked with family labour 998
Cotton Other cash crops Subsistence food Less expenses For cotton For sorghum For fodder crops Family income
Productivity
225
166
656
822
Gross return ($ per ha total land)' 52 59 Gross return ($ per ha cropped)b 104 118 40 per cent of the value of the cotton crop. 100 per cent of the value of the cotton crop, including the share of the government and the board. Source. Vcrsluys (1953-4).
S
in the growth period ofcotton, i.e. 1200 millimetres, which is supplemented by 200-400 millimetres of rain in a year. (4) The annual preparation of the soil with tractors is also carried out by the scheme management and presents the tenants with land ready to be sown-the arrangement of cropping in large blocks with one crop facilitates mechanization. (5) Seed, mineral fertilizer, and other material inputs are supplied. (6) Pest control is centrally undertaken by the management.
181 SYSTEMS WITH ARABLE IRRIGATION FARMING (7) It also ensures that operations delegated to the tenants are carried out according to plan. (8) The transport, preparation, and sale of cotton is centrally organized. (9) The scheme management organizes the distribution of small amounts of credit for the payment of seasonal labour for cotton picking. (10) It is responsible for the collection, ginning, sorting, storing, and sale of cotton. (11) It also finances research. (12) It supports the social institutions in the project area. The tenants are assigned special tasks: (1) They are resporsible for all manual work-seeding, weeding, harvesting, uprooting old cotton stalks, ditch clearance, etc. The work must be carried out at specific times. Each block isunder the supervision of a block inspector, who checks the timing of operations in cotton growing, but the tenants are left to do the work themselves, or to employ seasonal workers. (2) The cotton has to be surrendered by them to the scheme management. (3) Tenants can keep their cattle in the project area. The fallow is used as communal grazing. Cropping of fodder legumes bridges the fodder shortage in the dry season. (4) Farmers are members of a tenants' organization, which represents the interests of the tenants at the scheme management level. The harvested crops of sorghum, other crop products, and animal products belong to the tenant in their entirety (see Table 7.7). The dues of the project management are deducted from c, ton receipts. First of all, those cost items are deducted that have to be carried by the project and the cultivator jointly. These include: mineral fertilizer, seed, pest control, tractor ploughing for uprooting cotton stalks (which are burned after the harvest), the wages of field guards, the ,:osts of transport, processing, and marketing of the cotton, and applied research. The amount remaining after these deductions is distribtited as follows: tenants 48 per cent, government of the Sudan 36 per cent,' Gezira Board 10 per cent, reserve fund of the tenants, social fund, and local communities each 2 p-r cent. The payments are apportioned to the individual tenants, and thus each tenant receives an income that amounts to 48 per cent of the value of his cotton, in addition to the return from his other activities. 7.5. Weaknesses of thG syst.OM Agricultural innovations during the last decade have increased so rapidly that the gap between actual and optimum output is usually high, and particularly so in irrigation farming. The main weaknesses of irrigation systems
share 'The government's share declircd from 40 per cent in 1965 and the smallholders' increased from 44 per cent in 15 and 42 per cent in 190.
182
are not inherent in the system as such, as is the case in shifting systems and generally to a lesser extent in rain-fed farming. The major weaknesses lie rather in the inadequacy of water control, husbandry practices, and irrigation institutions. 7.5.1. Insufficient control of 'watersupplies The most important bottlenecks in irrigation farming are usually the insufficiency and unreliability of water supplies in the season when water is needed, and waterlogging at other times. This is particularly trae for terraces with wet rice, for minor irrigations with various arable trops, and for the large river plains, where drainage is often a great problem. Irrigation water should be under the ccntrol of the irrigator at all times, and this is usually not the case. Most of the wet rice in Asia and Africa is grown during the rainy season without a reliable source of water. Wet-rice production relying mainly oni impounded rainfall rarely exceeds 1.5 tons per hectare, while twice the yield can be expected with controlled irrigation, even without improved practices. The major reasons why wet-rice producers harvest low average yields are primarily outside the control of the individual farmer. The modifications in the environment necessary to achieve effective water controlirrigation and drainage, during both the wet and the dry seasons-will have to come primarily from investment in water-distribution structures. Pu'blic investment and communal work are required to narrow the gap between typical and poten~tial yields in wet-rice production (see Ruttan 1966, pp. 32-3). The failure to develop an effective water storage, transportation, and drainage system in large parts of the tropics is partly due to the physical geography of the areas concerned. Japan avd Taiwan, for instance, are characterized by short river valleys anu narrow coastal plains, which lend themselves to locally organized, small-scale, labour-intensive irrigation, and drainage works. Most other rice-growing countries in south-east Asia and Africa are characterized by broad river valleys and plains. Under these conditions, the physical geography dictates the organization of large systems of water distribution on a national basis. The construction of such systems requires high levels of capital investment. Accomplishment of this scale of investment is held up by the generally low profitability of large-scale irrigation schemes, which is to be attributed mainly to low product prices, and difficult organization and administration. The rational use of water in minor irrigations, which in terms of area and production potential are even more important than large schemes, suffers for various reasons, the most conspicuous being the following: inappropriate water and land rights, administj ative inadequacies, difficulties in organization of underemployed rural labour without payment in communal work, difficulties in organiztion of efficient cooperatives for water use, lack of small
prepared to incentives for farmers who are amounts of credit, and lack of undertake development. husbandry practices 7.5.2. Wasteful irrigation and which prevail in most of the minor irrigation practices, Wasteful irrigation is lost in field irrigation farming. Much water of drawback another surface are areas, to irrigate iand with an uneven Attempts seepage. of because Sufficient canals of water use and low crop yields. generally result in low efficiency of loss of land high yields and the prevention for prerequisite a is doses of drainage water, in particular too high salinity, irrigation of Misuse in soil due to salting. waterlogging, and a build up of water and water, can cause soil erosion, metre of output per cubic terms in efficiency lower of and can thus facilities. In particular, the sharing per unit of investment in irrigation water meters, as is customary in minor water on a rotational basis without uneconomic use irrigators, often leads to the irrespective irrigations that supply several turn, take their full share at each of water. The users will tend to tradiof the water needs of their crops. encourage the tendency to continue have Wasteful irrigation practices mineral fertilizers, and pesticides tional farming, although new varieties, possibilities. The numerous problems this concreated a new situation with attractive are dealt with elsewhere.' In innovations introducing in of its involved that irrigation farming, because out point to necessary is to change nection, it more on the willingness of farmers Multiple production potential, depends systems. drive than do other land-use and and on knowledge, skill, ann Preparation of the land, planting,cultithe if cropping makes special demands. only several times within a year harvesting can be accomplished observe the and to discipline themselves to vators are prepared to work hard necessary schedule of work. institutionalarrangements 7.5.3. Dependence on inadequate irrigation and inadequate husbandry Short-comings due to uncontrolled of lack of knowledge, skill, and optimum practices are, however, only partly a matter of the gap between actual and closing systems, irrigation In institutional drive. any other land-use system on farming depends more than in of land and true This is particularly production. promote that arrangements on water rights. few farmers are able to organize relatively that practice a is that Irrigation and drainage are complex processes their own. Water storage, supply, action that will affect a large number of joint organized can be carried out only through the way in which these things are Consequently, individual farmers.
I Ruthenberg (1964, 1966, 1967).
183
184
can be a crucial factor in the progress of irrigation farming. Often the functions of land ownership and water supply on the one hand, and their use in farming on the other, are divided between different people: the private or public land!ord and the smallholder. A practice detrimental to irrigation farming in such a situation is sharecropping, wlich reduces the incentives for intensification. On the other hand, a high fixed monetary charge per unit of land, in the form of either a land or a water charge, although highly effecLive in introducing intensive farming, is politically difficult to implement. In some regions, where the normal rainfall is adequate, the farmers tend to regard irrigation only as an insurance against years with insufficient rainfall, because water is free of charge or there is no fixed water charge sufficiently high to induce more intensive farming. A number of projects that have been operating for years-the Gezira scheme in the Sudan is the nost famous example-show that irrigation farming under close supervision allows modern farming techniques to be used by unskilled smallholders where required. Production under close supervision, however, is not a generally applicable princip!e. The approach shows several weaknesses, the most important of which seem to be the following: (1) High costs of supervision and mechanization have to be borne either by the tenants or the government. Projects must be large enough to work economically. Since the scheme managements only rarely have any competition to contend with, there is the danger that top-heavy, authoritarian, and arbitrary bureaucracies might develop. (2) The layout has to be simple and uniform. Concentration on one or two major cash crops is advisable. The patterns of land use are necessarily rigid. Smallholder farming under close supervision is not characterized by variety in cropping like other types of irrigated smallholder farming, though this variety helps to increase yields and reduce risk. (3) The tenants are inclined to employ labourers and to become miniature landlords. The more compulsive and complete the supervision is, the smaller the scope for individual initiative and the less willing farmers are to invest in their holdings. (4) Production under close supervision is usually unpopular, and the success of production depends, therefore, upon political backing. The participants very rapidly feel themselves to be capable of running the production properly without supervision, an opinion that is not as a rule confirmed in actual practice. Minor irrigation areas are too small to be organized along the lines of production under close supervision. The traditional landlord, who had the authority to distribute or sell water economically, has largely been abolished by land reforms, but most of them were neither benevolent nor technically progressive. Cooperatives may fill the gap, but experience shows that it is
185
very difficult indeed to organize irrigation properly wihout the backing of some outside authority, and this runs contrary to the principle of cooperative action, 7.6. Development paths of irrigation farming Further development of holdings with arable irrigation farming will lead to fundamentally different land-use systems only in exceptional cases. The change to irrigated tree crops or to irrigated pasture are two relevant possibilities. The return to extensive grazing is another possibility where land is lost due to salting. As a rule, the real issue for farm management is how to increase production within the framework of arable irrigation farming. The following basic possibilities present themselves: (1) Increase of the irrigated area, by increase of water supplies; or obtaining of a more efficient use of the available water. (2) Increase in the land productivity, by getting higher yields per crop cultivated; change to higher yielding crops; and increase in mildtiple cropping to get higher output per month and per hectare. Irrigation farming in Japan and Taiwan is particularly advanced in its use of these possibilities. Although the ecology of much of the tropics and the price relationships that usually prevail rule out a direct transfer of the production techniques of these two countries, some of the principles successfully applied in economically advanced countries of high population density may be taken as indicators to the development path of other irrigation systems in the tropics. 7.6.1. Innovations and improvedhusbandry There is much scope for economizing on water use in irrigation farming. Carruthers (1969, p. 34) summarizes the relevant approaches in the following way: (1) Adaptation of cropping pattern to the availability and the reliability of water supplies. (2) Allowing crops to draw upon stored soil moisture. (3) Allowing crops to suffer water stress for some time, if this can contribute to increasing overall production with the given resources of water, labour, and capital. (4) Fulfilment of leaching requirements at times of the year when there is no peak demand for water. These are economic rather than technical criteria for determining water application. Furthermore, there is a good chance that important watersaving innovations might be developed in the near future. A case in point is the replacement of the continuous flooding of rice by intermittent irrigation.
186
Even more important in the short run are innovations in plant development. A two-fold or three-fold increase in yields per hectare and per crop under field conditions can be realized in extended areas, thanks to the interaction between improved irrigation, improved varieties, increased application of mineral fertilizer, the use of pesticides in combination with proper nursery techniques, planting, and weeding. Particularly far-reaching is the impact on rice production. In fully irrigated areas with traditional farming it was common for average yields to hover around 2-3 tons per hectare in the wet season and 2.5-3.5 tons in the dry season, with some particularly careful farmers reaching yields of about 4-5 tons. The new rice varieties seem to have a yield potential under field conditions of 6 tons in the wet season and 8 tons in the dry season (Ruttan 1966, p. 32). The general principle is illustrated ,y Fig. 7.10, which shows the interaction between the application of nitrogen and use of rice varieties of the short, erect type. These varieties show several advantages compared with improved indigenous tall varieties: (1) They have superior yield-increasing characteristics like a high number of effective tillers per crown, a large number of grains per panicle, heavy grains giving a high 1000-grain weight, and a large number of fertile grains per panicle. (2) They are early maturing, thus making it possible to expand multiple cropping. (3) They are not sensitive to the length of the day, i.e. they can be grown in any season. These advantages are clearly more important than the disadvantages, which are higher susceptibility to various plant diseases, less straw, and slightly different cooking qualities. Innovations in irrigation and plant development thus offer much scope for economies of scale by increasing the output on a given area. 7.6.2. Irrigationfarming and mechanization Most tropical irrigation farming relies mainly on hand labour and animal power. Mechanization has been introduced to a significant degree only in large schemes with controlled production. There is,. however, some evidence from the subtropics that irrigation farming, mechanization, and employment are largely complementary to each other. Furrow irrigation and the use of machinery combine well, because levelling of the land fo: the furrows usually leads to even fields of sufficient -Azc for tractor ploughing and ridging. A principal disadvantage of basin irfigation is that the small dams which keep the water in numerous small basins interfere with the movement of animaldrawn or tractor-drawn equipment. In Japan, however, established and
187 FARMING SYSTEMS WITH ARABLE IRRTGATION mechanization of small farms with economically promising prototypes for rain-fed farming, farm machinery basin irrigation have been developed. In this is frequently not the case, often replaces farm labour. In irrigation. 10000IRO-288-3 9 000' 0977kg hd " 86-2kg hWday
6000
I.G
6000
S00Binafo
C.-18
6000
Tjoremas
3000.
Bengawan Peta
2000 0 30 60 90
")
120
of rice to nitrogen fertili.ation (from Bradficld Fio. 7.10. Responses of lifferent varieties 1968). in demand for traction power, and because tractor ploughing eases the peaks possible. The tractor may also this may make increased raultiple cropping the irrigated area and the be used for pumping, and thus may increase reliability of water supplies.
SYSTEMS WITH ARABLE IRRIGATION FARMING The scope for multiple cropping in systems with adequate irrigation depends largely on operations being carried out rapidly at the right time. The bottlenecks often arise in periods of only a few days, when there is insufficient labour available to carry out tasks whose timing is particularly critical. Intensive irrigation farming, even in smallholdings with ample labour, can therefore be stimulated by improved implements, in particular single-axle tractors. Where the ratio between input and return justifies the use of tractors, they are introdu.ed into irrigaiion systems relatively quickly, and bring about particularly conspicuous production increases. 188 7.6.3. Introduction of intensie I 'pes of animal husbandry As the tractor replaces the draught animal, the tendency in irrigation holdings is increasingly to introduce intensive types of animal husbandry, dairying in particular. In any case, harvest residues and grass growth on the edges of tracks and canals usually provide fodder that cannot be used in any other way. There are the additional possibilities of multiple cropping with fodder crops, and the introduction of fodder plants into the rotation. With various grasses like elephant grass, Serariasplendida, and Guatemala grass, yields of 50-100 tons fresh weight per hectare and per year can bi attained with irrigation. It will perhaps be easier to cover the rapidly growing need for animal protein in tropical countries by intensification of animal activities in irrigation farming (fish, poultry. pigs, and milk) rather than by improved grazing in hill country or dry areas (ranching). The production of milk and beef on intensively irrigated grassland, with high returns per hectare, might perhaps be a more economic and realistic approach to the development of a tropical cattle industry than any preoccupation with large grazing areas of low productivity.
8.1.1. Definition as A FORM of land us., in the humid savannas and the rainforest areas, and of typical as shifting cultivation and irrigation farming, is the cultivation several of perennial crops, primarily tree and shrub crops with growth cycles the land occupy which sisal, and pineapple, decades. Crops like sugar-cane, crops for several years but not for as long as shrubs or trees, are akin to arable perennial as regarded best are and view, from a farm-management point of the farmfield crops. Bananas are also included in this category, although management characteristics of bananas are distinctive. are grown Palms, trees, bushes, bananas, and various perennial field crops chapter, This systems. in gardens and fields tinder almost all tropical farming crops perennial which however, is concerned with those farming systems in return. gross are the main activity, in terms of their contribution to the terms 'estate' Perennial crops are grown in estates and smallholdings. The any land and 'plantation' usually designate large units. To avoid confusion, and the corthat is planted with perennial crops will be called a plantation by a responding holding an estate, provided it is large in area and operated large number of paid workers. 8.1.2. Genesis may Plantations of perniial crops, in particular of tree and shrub crops, plant, and come into existence in various ways. Large estates prefer to clear within and where land is ample estates tend to shift their plantations either land and the estate's boundary or by acquiring virgin land, i.e. to clear new was practised to abandon run-down sites. Until recently, shifting of this kind Paraguay. into spreading present at is it and with coffee in southern Brazil, and estates banana American Central for Intra-farm shifting was customary land as But reserve. in land East African sisal estates that hod sufficient the estate becomes scarce, plantations tend to become stationary within boundary. The genesis of smallholder plantations does not usually follow the sequence clear and plant. Two main types of development in smalliholder plantations sequence: are distinguishable. The development1a9
190
SYSTEMS WITH PERENNIAL CROPS (1) exploitation of fruit trees or palms that grow naturally in forests or exploitation of infillings (irregularly planted seedlings) in the forest;
(2) exploitation of systematically established plantations under the cover
of some remaining forest trees; and (3) exploitation of systematically established plantations with no forest trees left; is characteristic of the way coffee cropping has developed in Ethiopia, West Africa, and Madagascar, and oil-palms and cacao in many areas of West Africa have evolved in a similar manner. Plantations are developed directly from forest vegetation mainly where there are extensive forest areas. However, more frequently the evolution of smallholder plantations has followed a different sequence, stenming from shifting and semi-permanent arable farming. It usually intcludes four phases: (1) The planting of perennial crops around the house slows down the rotation cycle of shifting cultivators. (2) The area devoted to perennial crops is extended. Cleared plots with arable crops are interplanted with perennial plantc. (3) In time, the perennial crops predominate in the mixed-cropping system, and young stands of tree crops are usually interplanted with arable crops. (4) As the perennial crop creates more a:nd more shade, intercropping becomes less important. When this occurs, the cycle arable cultivation-fallow is replaced by the sequence arable cropping-perennial crops, and at the same time shifting systems become stationary. Perennial-crop cultivation does not, however, always lead to stationary farming. The abandonment of old plantations and the establishmnt of new ones, i.e. the slow shifting of tree-crop production, can also be found. The slow movement of cocoa holdings in forest areas of Ghana is a case in point. Generally, however, the introduction into smallholdings of commercial production with perennial crops is associated with stationary farming. Groeneveld (1968, p. 222) and Meillassoux (1964) illustrate how cultivation of shrub or tree crops is integrated into an arable system. Figure 8.1 shows the establishment of smallholder coconut-palm plantations. Manioc, maize, peas, or beans, and also coconut-palm seedlings are planted or sown on one plot simultaneously. The annual crops, maize, beans, and peas are harvested after three to five months. Their harvest also provides the final weeding for the manioc. The manioc harvest commences 13-15 months after planting and continues on the same plot up to the twenty-fourth month. The planting of manioc, peas, and beans is repeated on two to three consecutive occasions. Meanwhile, the young coconut-palms are growing. By weeding the arable crops and thereby preventing the bush from regenerating, the farmer also protects the palms. After four to six years, arable cultivation comes to an end. The plot is left to the coconut-palms, which at this time yield their first nuts. Afallow ensues after three to four manioc cycles and
191
can extend over several decades. In the meantime (eight to ten years after the first planting), the coconut-palms have grown so large that the regenerating bush can no longer endanger their development. The establishment of smaliholder coconut plantations is thus a by-product of arable subsistence farming. The stands of coconut-palms are set tip without expenditure for clearance and care, since the planting proceeds almost without cost and effort as a consequence of the manioc cultivation. On the other hand, palm development is slower, owing to interculture. It should be noted, however, that the area of coconut-palms is dependei.t upon the extent of
Plant growth
Gruwvng peiod HarveSting I period
Maize.pen$l,tonf
ntercurture
Pure stnds
Fio. 8.1. Interculture of coconut-palms and various arable crops, Tanga, Tanzania (from Groeneveld 1968, p. 223).
manioc cultivation, since young palms are planted only where manioc is cultivated on freshly cleared bushland with soils suitable for coconut-palms. Planting of additional coconut-palms would give rise to high expenditure for clearing and tending the palms solely for their own sake. Figure 8.2 provides art example of this type of cropping from the Ivory Coast. Whereas in the manioc-coconut system the permanent crop is planted right at the beginning of the cropping cycle, the smallholders of the Ivoiy Coast plant their coffee later. After clearing the land, yarns and manioc are grown, interplanted with short-term crops like grour,dnuts and beans. About a year after clearance, coffee is interplanted, and occupies the plot for fifteen to thirty years, together with spontaneous bush.
8.2.
The various ways in which plantations are established give rise to different
types of farming with perennial crops, which can be conveniently classified according to their cropping and exploitation systems.
192
8.2.1. Cropping systems amount of cultiOn the b tsis of the length of the vegetation cycle and the distinguished: vation involved, the following cropping systems can be pineapple, sisal. sugar-cane, include we Here, crops. (a) Perennial field characteristics of bananas, and other similar types, which do not have the they mostly require shrubs or trees. With the notable exception of bananas, considered from the a considerable degree of cultivation, and can thus be and tree crops. farm-management angle as something between arable farming
Yams
Manioc
-1o
10 months
2
J A S
15 years
J A S 0 N D J F M A M J
of shifting cultivators with stationary Fio. 8.2. Intarplanting of coffee into arable piot, housing, Ivory Coast (from Mcillassoux 1964).
the transportation Where these crops are grown for export, cultivation and of this kind Crops mechanized. largely of the harvested produce are often regions where they can are therefore grown principally on large estates. The soil conditions as by and climate by be grown are determined not so much knowledge. technical ar.d the availability of markets, transport, labour, The vegetation crops. shrub (b) Shrub crops. Coffee an,' tea. are typical above. mentioned crops cycle lasts longer than that of ti' perp-,aal field labour manual of input Shrub crops usually differ from tree crops in the high that product end an required fori weeding and pruning. They usually yield relaat subsequently require. early processing and can thus be transported are therefore deterareas coffee-cropping and tively low costs. Tea-cropping mined principally by soil and climatic conditions. coconut-palms, and (c) Tree crops. Finally, tree crops like cocoa, rubber, them and other oil-palms constitute another group. The difference between and the lower cycles, vegetation long perennial crops is their particularly in a tree Work weeding. and labour input required for their maintenance place takes and crops, shrub plantation is usually less strenuous than with largely in the shade.
193
8.2.2. Exploitation systems Whereas arable farming in the tropics has been from time immemorial a domain of the smallholder, until the twentieth century cultivation of perennial crops for cash was carried out mainly in large estates. However, this distinction no longer applies. Plantations vary in size from very large estates of over several thousand hectares down to only a few shrubs and trees. Almost all types of perennlial crop are grown in large estates as well as in smallholdings, although some are better suited to large-scale production than others, and are mainly to be found in estates. The distinction between cropping systems is, therefore, in many respects a distinction between large-scale and smallscale production. (a) Estates. Although virtually all perennial crops are grown by smallholders, whose production tends to increase steadily, there is still important large-scale production, of which the following are the most important examples: (1) sugar-cane estates in almost all tropical regions; (2) banana estates in Latin America and West Africa; (3) sisal estates in East Africa, Indonesia, and Madagascar; (4) tea estates in India, Ceylon, and East Africa; (5) rubber estates in Liberia, Malaysia, Ceylon, and Indonesia; (6) oil-palm estates in West Africa, Malaysia, and Indonesia; (7) coffee estates in El Salvador, Brazil, Kenya, Angola, etc.; (8) coconut estates in Malaysia, Ceylon, Indonesia, India, and the Philippines. In estate production, a distinction can be made between individual planters and firms, the former operating usually fairly small estates. In the normal way, the owner of the small plantation is also the estate manager, and he owns only one estate. This form of undertaking in the tropics is typified on the one hand by the shortage of working capital, exposure to risks, and often obsolete technology, and on the other by adaptability and low overheads. Firms, however, usually each own several estates. The tendency to cultivate very large areas spring from the necessity to use industrial processing and transport to their utmost capacity, in order to recoup the high costs of overheads through economies of scale. The motive underlying this kind of large-scale production is frequcntly to obtain a regular supply of goods of constant quality, whether for delivery to a factory (tea and sisal), or for delivery to an established market (export bananas). Almost always, production of this kind is associated with well-developed, reliable marketing outlets. The producing firms often regard their farming activities as a necessary evil, undertaken only because other farmers are not in a position to provide the raw materials in the desired way. On estates, emphasis is often laid more on
1V9i
producing industrial processing and the organizr, tion of transport than on the raw material. form of Large-scale cultivation of perennial crops is found usually in the is for This monoculture. pure stands, and there is an obvious trend towards a number of reasons: of a single (1) The producer is interested only in obtaining large quantities raw material. more easily (2) Where monzoculture is carried out, unskilled labour can be productivity. of degree high supervised and brought to a relatively wages that (3) A relatively large part of the cost consists of paying the while in worth not is production managers can command, and diversified required. be would many cases because additional skilled personnel number of (4) The production conditions favour monoculture. With a to necessary not is it perennial crops, labour is required so regularly that year the all diversify production in order to find productive employment lead to soil round for the labour fcrce. Monoculture does not inevitably even be can erosion. Indeed, with several perennial crops monoculture termed soil-conserving compared with arable farming. are settled, (5) Whereas smallholders usually have to operate where they diversify to compelled thus are and habitat, and adapt to the natural natural and economic favourable most the production, the firm can select monoculture. for location, which is chiefly on land suitable Jiversified In addition to estates with monoculture of perennial crops, supplement to large-scale estates occur that are engaged in other activities the main crop. The following forms can be distinguished: or several (1) The predorninrnt tree w-op may be supplemented by one croppingmixed by either other perennial crops, which can be integrated carrying for instance bananas under coconut-palms-or by part of the land another perennial crop. Managers (2) The perennial crop may be supplemented by arable farming. of tree rows the of estates choose only rarely to grow arable crops between example for as or shrub crops. Instead, they prefer to use separate plots, groundin the coffee holdings in Brazil, which grow maize, rice, cotton, nuts, and onions (see 8.5.1. (a)). particularly (3) Perennial crops may be supplemented by stock-keeping, production beef Brazil, of estates coffee the cattle, for example, dairying in coconut-palms the among dairying and in some sisal estates of East Africa, in Ceylon and East Africa. in kind from The motives behind diversification in estates are different part, since in those in smallholdings. Subsistence needs do not play a great
195
general the workers are allocated some land to cover their food requirements. Hardly any draught animals are kept, which means that there is no need to crop fodder. Similariy, the idea behind diversification is rarely to employ the available labour force more regularly and to its capacity, since many of the estates have seasonal labour at their disposal in the surrounding countryside. Reduction of risk by a more varied production is of more importance, but there are two factors that are usually vital to diversification: the need to make better use of soils with different fertility, and the wish to exploit favourable marketing prospects. These two motives are manifested particularly clearly where there is a shortage of land, and where urban markets have been developed with a demand for a diversified supply of farm produce. (b) Smallhokdings. A large and growing proportion of perennial-crop cultivation is in the hands of smallholders. Of particular importance for the world's economy are the semi-commercialized holdings, for example: (1) the cocoa holdings in Ghana, Nigeria, and Trinidad; (2) the rubber holdings in Malaysia, Nigeria, and Ceylon; (3) the coffee holdings in Colombia, the Ivory Coast, Kenya, and Madagascar; (4) the oil-palm holdin-s in West Africa; (5) the coconuL-paim holdings in the Philippines, Ceyion, Venezuela, and Samoa; (6) cashew holdings in India, Tanzania, and Mozambique; (7) vanilla holdings in Madagascar; (8) sisal holdings in north-east Brazil; and (9) banana holdings in Central America, South America, and Asia. The primary cash crop is usually supplemented by yams, manioc, rice, maize, beans, etc., for household use. A strikingly large ploportion of the cash production is in the hands of a numerically small but economically important group of smallholders, who employ seasonal labour, and who might be termed small entrepreneurs. The development of these hcldings from pure subsistence :arming brings about a situation where the household's food requirements compete with the desire to earn money with cash crops. The competition for land takes two main forms: (1) In some cases, perennial crops are grown on separate plots as supplementary activities. The husbandry techniques are usually similar to those on large estates, and operations are often carried out by seasonal, paid labour. However, as cash cropping increases, there is less land available for meeting food requirements. At the same time, arable farming is undergoing a change, which as a rule is a process of degradation, because the
196
are obliged to abandon fallow is shortened and fertility declines. Farmers and more to manioc. more millet and maize, for example, and to transfer a larger number of supply Manioc is eaten with less relish, but it does the stage is infrequently calories per hectare and per hour of work. Not small-scale a within i.e. reached where 'dualistic' peasant holdings evolve, is as production of holding the interrelation between the two branches subsistence the traditional weak as that between the large estate and is carried on by the family farming Arable society. holdings in a dualistic fallows and declining yields. in the traditional way, although with reduced undertaking, The plantations of the same holding are regarded as a special 1964). (Blanckenburg apply standards and new husbandry and economic mixed in crops are grown (2) In other cases, cash crops and subsistence by the composition of the stands. The degree of competition is reflected are grown in mixed stands, mixed stand. Thus, when coffee and plantains tends to be greater the the proportion of the subsistence crop (plantains) vice versa. Compared and fed, more people per hectare there are to be has the advantage that with the dualistic smallholdings, mixed cropping avoided. On the other is the progressive degradation of the arable branch technical advances such to hand, mixed cropping proves to be an obstacle as use of mineral fertilizer and plant protection.
have developed from subBesides the semi-commercialized holdings that that have come into being in sistence farming, smallholdings are found in relation to non-farm work. consequence of large-scale undertakings or of part-time farmers who freThis applies primarily to the great number few years in towns, quently cultivate perennial crops. People work for a on a regular wageor workers, large-scale holdings, and mines as seasonal a base and a measure of support, earning basis. Their own holding provides investment of savings in shrub and a home in old age, because, through the high degree of security. or tree crops, the holding can offer a relatively sometimes been evolved on Moreover, smallholder plantation fayning has on a large scale. The smallland where it is no longer profitable to operate for whom there is usually no holdings command the labour of their families, them to buy land from runalternative but to work their own land. It pays In some areas, like southern down estates and to establish small plantations. smallholder coffee farming on Brazil, this procedure has led to the spread of coffee estates, in which the land that has been purchased from run.down channels created by largenew producers take advantage of the marketing use is very similar to that in scale production. Their organization of land estates. establishment of plantations Finally, a trend of increasing importance is the reliable return on capital and a by middle-class entrepreneurs who aim at a secure farm as a base for their old age.
197
8.3.
Although perennial crops are a heterogeneous group, grown under various circumstances by various types of farmer, some common characteristics exist, which are mainly relevant to shrub and tree crops and which to a lesser degree apply also to several perennial field crops (see Table 8.1). As a
TABLE 8. 1
first crop
I-1.5 1-2 1.5 3 3 3 3-4 4-7
8-11
Years to
Years to
maturity
I-1.5 3 2 5 5-6 6 7-9 8-1
15-20
production
4-6 5-50 3-5 8 12-50 50 35 35
80-100
Years of
Harvested
portion
Urgency for
processing
High Hil9h Fair Fair High High High Fair
Low
Coconuts Cloves
4-6 8-9
8-15 20
60 100
Low Fair
first step, the farm-management characteristics of perennial cropping can be compared with those of arable farming, and then some general principles, applied by farmers to fit perennial crops into their farming systems, can be outlined. The various aspects of the maintenance of soil fertility, of animal husbandry, and of the labour economy are discussed within the case studies. 8.3.1. A comparison with arable farming Cultivation of perennial crops involves a number of farm-management characteristics that in many ways can be regarded as advantages compared with those of arable farming: (1) Perennial crops give rise to stationary living and to investment in permanent improvements like houses, tracks, irrigation systems, etc. Where perennial crops are grown, the individual ownership of land tends to become established. (2) Most perennial crops are characterized by a comparatively high productivity per hectare. This encourages the use of fertilizer and thus the conservation of soil fertility. On account of their high productivity per
198
SYSTEMS WITH PERENNIAL CROPS hectare, some perennial crops actually economize on land use. Thus in Bukoba, Tanzania, 0.3 heciares of bananas mixed with beans are sufficient to cover the basic food requirements of one family, whereas in arable farming the annual cultivation of about 1.2 hectares is necessary to meet the same need. Most perennial crops produce by-products that are valuable to the household, can be fed to livestock, or can be sold in local markets: fuel for the kitchen, timber for construction purposes, branches for covering roofs, tops and by-products for fodder, etc. (3) There is a good chance of conserving fertility when perennial crops are grown. The frequency and intensity of cultivation is disproportionately less than in arable farming. Perennial crops hade the soil and permit or require a permanent cover of grass or leguminous vegetation. Some tree and shrub crops influence the soil in the same way as the forest, and moreover permanent planting encourages the construction of terraces, the control of water courses, and other permanent land improvements. No matter how great the destruction of fertility is in cultivation of some perennial crops, like coffee in southern Brazil or tea in Ceylon, the fact remains that the damage is usually less than in arable farming under similar conditions. (4) Various perennial crops allow the farm-management advantages of monoculture without a reduction in soil fertility. Properly husbanded, pure stands of rubber, oil-palms, or tea may well be considered as soil-conserving types of land use. (5) For some perennial crops, land can be used that is either out of the questions for arable farming, or would be usable only if high costs could be borne. Examples are steep slopes (bananas, tea, rubber); rocky terrain (tea, rubber); land with unreliable rainfall (sisal); and land where there is a high salt content in the water (date-palms). (6) Important perennial crops like coffee, tea, cocoa, and sisal require a great deal of manual labour, but repay it with a relatively high income. They are therefore suited to the production conditions of developing countries, which are characterized by the availability of plenty of cheap labour. (7) The labour required for many perennial crops can be spread fairly evenly over the year. This facilitates the expansion of monoculture. Moreover, most operations required by perennial crops tend to be disproportionately easier and more agreeable than in arable farming, particularly hoe farming. (8) The fluctuations in the yield of crops like sisal, rubber, and coconutpalms are small-at least smaller than in arable farming under similar conditions. Thus, the sisal hedges in north-east Brazil and in the smallholdings oiu Lake Victoria are cut chiefly when arable crops suffer harvest
199
failure; in other words they can counterbalance such risks. Similarly, fodder bushes and trees are a source of fodder when required. The greater certainty of yields from perennial crops is important in small-scale tropical holdings, chiefly because the yields from arable rain-fed farming are often unreliable. (9) The products are chiefly goods that can be transported and stored, and have a high value per unit of weight, or they acquire these qualities through industrial processing. Perennial export crops make it possible to develop areas for commercial production where only subsistence farming could otherwise be practised. (10) Perennial crops nresent considerable scope for intensification according to variety, raising and planting method, density, fertilization, pruning, harvest, and processing. With some crops, irrigation is sensible and increases the yield appreciably. With perennial crops the economic scope for increased production is greatly enhanced by various technical innovations. (II) In the case of most perennial crops, cultivation can begin with only a few plants. The enterprising individual can progress without having to overcome any particular technological threshold, nor is he dependent on the cooperation of neighbours to the same extent as in irrigation farming, for instance. (12) Wealth in the form of plantations, and therefore security in old age, can be accumulated by planting shrubs and trees. This is particularly important where there is no private ownership of arable land, and where investment in cattle as security is not possible. (13) Perennial crops give rise to social differentiation-perhaps more than ariy other cash crop. A family's income is based on the number of trees and shrubs it has and their yield. This forms an outlet for the common human desire to excel, and at the same time provides an additional incentive towards economic activity. On the credit side we find that perennial crops have many agro-economic advantages, which naturally do not obtain in every case, but which characterize the group as a whole. But on the debit side there are certain important farm-management requirements that have to be met: (1) Creation of plantations usually requires a high initial investment per hectare. Yields do not reach their capacity until some years after planting (see Table 8.2). To cultivate perennial crops it is necessary to finance the establishment of the plantation; maintain plants during the years when there are no yields; and bear the burden of accruing interest or-which is more important in smallholdings-forgo for some years the return in food or cash that the plot would have yielded under arable farming. On the other hand, a plantation is a form of investment, which a family
200
of early treatment, which is designed to effect some change in the composition of the produce. chemical or biological deterioration would set in and
SYSTEMS WITH PERENNIAL CROPS can build up by supplementing small amounts of cash with much unpaid family labour. (2) For many of these crops, it is very important that processing should take place within a relatively short period after harvesting. In the absence
TABLE 8.2 The economics of plantation development in the West African forest zone (per hectare, fields only, without nurseries and overheads)
Age of plantation
lb
6 432 55 12 67
7 432 55 12 67
8 432 55 12 67
9-25 432 55 !2 67
Labour (hours) 816 328 126 280 272 352 Wages ($) 106 42 18 36 35 45 Material inputs (s) 354d 30 12 12 12 Total input ($) 460 72 18 48 47 57 Yield (tons bunch-weight) 1 70 Gross return* 16 112 Net return ($) -460 -72 -18 -48 -31 +55 Cocoa (Nigeria, 1961) Labour (hours) 1604 520 520 408 497 432
90 12.5 15.0 16,0 144 200 240 256 +77 +133 +173 +189 480 552 552 552
Wages ($)
Yield (kg)
161
54
158
-
97
54
42
54
96
-
114
60
41
57
50 138 87 -51
88 250
111
58
43 131
88 580
48 146
500 98
55 153
98 880
55 153
98 1160
99
130
-96
45 -54
21 31 70 17 57 17 57 37 286 103 66 68 103 50 177 137 177 157 . - .-.-400 650 900 1100 . - .-.-132 215 297 339 -286 -103 -66 -68 -103 -50 -45 +78 +120 +182
513 51 1l)l 52
328 1200 1200 1200 12030 1200 33 33 120 120 120 120
a Year of clearing.
b Year
of planting. The labour input is low because of mechanical clearing. From year 6 onwards, one-third
of labour is required for plantation upkeep and two-thirds for harvesting. Costs of mechanical clearing mainly. Mainly plant material, fertilizer, and pesticides.
About five-sinths of the labour from year 5 onwards is tapping. I From the 6th year onwards fertilizer is applied every second year. I Yields according to Ceylon Rubber Research Institute. Price of $0.33 per kg of dry rubber sheet RSS Grade I.
Sources: Oil-palm, S.O.D.E.?.A.L.M. (1965); cocoa and rubber, Food and Agriculture
Organization. (1961),
201 SYSTEMS WITH PERENNIAL CROPS either reduce the quantity of valuable constituents extractable from the harvested material-oil-palm and sugar-cane especially-or else spoil the quality of the final product, as with bananas, coffee, and tea (see Table 8.1). (3) Where perennial crops are the raw material for a local industry, in addition to the initial costs of the plantations it is necessary to invest in a local processing plant; to invest in transport for bringing the product to the factory; and to arrange delivery very carefully, so that the processing plant can work to capacity. (4) Consequently, holdings with perennial export crops, like irrigation holdings, operate with high fixed costs, but, unlike arable irrigation holdings, they are committed to one type of production for a long time. (5) Plantation husbandry demands different kinds of work and skills from those of arable farming. Cultivation is of less importance, and with several crops not even necessary. Weeding is labour dcmanding only in the early phases of plantation development. Instead, much work is required for raising plants in nurseries, planting, pruning in the case of shrubs, spraying, mulching, harvesting, and so on. The labour input is particularly high when crops are being established and with mature crops. The intervening period usually requires little attention. In comparisons of arable production with shrub-crop and tree-crop production, it is also important to realize that the influence of husbandry measures may extend ov.r years and decades. Poor plant development, unskilled pruning, and delay in pest control may damage the crops for years to come, while in arable cropping, last year's failure may be avoided next year (MacArthur 1969). (6) Another disadvantage of perennial crops is that the work is difficult to mechanize. Where cultivation, weed control, and transport in bulk strain the labour economy, as in sugar-cane and sisal estates, mechanization is more likely to follow. With tree and shrub crops, where the main emphasis is on tending of the crops and harvesting, operations cannot be mechanized at all, or only with rather poor rates of substitution. In assessing the various advantages and requirements of perennial crops, it is clear that, provided market outlets are available, they are relatively well suited to tropical agriculture, and several of the shrub and tree crops are particularly suited to sraallholder production. 8.3.2. Some farm-management characteristics ofperennial crops usually uniformly organized, but in smallholdings, planting patterns near the house or village are mostly different from those some distance away, so that more or less pronounced spatial patterns are found in the arrangement of the various perennial crops. In this respect they are not basically different from arable crops, where spatial arrangements of cropping in smallholder farming are common.
(a) The .patial and vertical organization of cropping. Large estates are
202
A specific feature of shrub and tree crops is their vertical arrangement, since plants differ in height and in light requirements. In estates we occasionally find a 'two-storey physiognomy', i.e. two perennial crops-for instance rubber and cocoa-are interplanted. In smallholdings we find multi-storey physiognomies, including not only perennial crops but also various annual
(a)~j~
1:
(b)
12
(C)
1'2
45
Fio. 8.3. Various types of land use on Polynesian islands. (a) High island, Rarotonga: (1) ocean; (2) coral calcimorphic soils, growing coconuts, sweet potatoes, tomatoes; (3)hydromorphic soils, taro gardens; (4)colluvium clay loam, citrus fruits and gardens; (5)lateritic soils, scrub and secondary forest. (b)High island, Atiu: (1)ocean; (2)uplifted coral plateau with clay loam in pockets, ;.few gardens and coconuts; (3) hydromorphic
soils, taro gardens; (4)colluvium clay loam, gardens; (5)lateritic soils, ferns and Casuarina. (c) Atoll, Tarawa: (1) ocean; (2) coral calcimorphic soils, coconut and Pandanus; (3) Cyrtosperma pits; (4) coral calcimorphic soils, coconuts, and Pandxnus; (5) lagoon. From Barrau (1961, p.26).
crops. A typical case in point is the land use on some Polynesian islands, as depicted by Barrau (1961). In Fig. 8.3 (a), (b), and (c) show distinct spatial arrangements in land use combined with several two-storey and three-storey physiognomies.
203
(b) Monoculture, interculture, and combination with grazing. The spatial and vertical organization of cropping leads naturally to various crop combinations within a holding and within a field. Large estates usually prefer to cultivate pure stands, in order to make optimum use of the husbandry techniques specifically required by one plant. Where production is concerned with only one crop, pure stands result in monoculture, of which two types have to be distinguished. From the standpoint of farm ranagement, we have monocuiture where only one crop is grown in the holding. This is usually the case in oil-palm or banana estates. From the standpoint of agronomy, we have monoculture where one field is planted with' the same crop for each succeeding vegetation cycle, which is usually the cas! with sugar-cane or sisal. Both types of monoculture are identical in some but certainly not in all cases. While estates tend to prefer pure stands and monoculture, smallholdings usually prefer intercropping, because it helps them, to obtain higher shortterm returns, higher returns per hectare (lower returns per unit of labour being accepted), and the benefits of a more diversified production. However, some types of mntercropping are also prevalent in estates. The following types of intercropping with perennial crops are important: (1) Prennial crops and cover crops. In large estates, the planting of leguminous creepers (e.g. Centrosen pubescens, PuerariaJavantica) as cover crops is a usual practice in protecting the soi! in the early stages of plant development (oil-palm, cacao, rubber) or to help in weed control (sisal). (2) Tree crops and shade plants. Some tree crops need shade in their early stages. The shade plants frequently produce a yield of their own, for instance bananas shading coffee. (3) Perennial crops of various sizes. A widespread practice, in estates as well as in smallholdings, is the combination of various perennial crops of differing heights (see 8.4.3 (b)). This multi-storey physiognomy is closer to the natural climax vegetation than are pure stands. In some cases, interculture of this kind is limited to the early stages of plantation development. (4) Perennial crops and annual cropping. In large estates, perennial and annual crops are not usually combined, whereas in smallholdings their combination predominates. With some crops, like rubber, oil-palms, and coffee, interculture is limited to the early stages of plant development, and arable crops are grown only until the main crop branches out and forms too dense a shade for annual crops. With other crops, like coconut-palms, arable farming continues to be practised under mature stands. (5) Perennial crops and grazing. In addition we find various combinations of perennial crops with grazing. in some cases (sisal, oil-palms), grazing is limited to the early stages of plant development, when ample grass grows between the rows. In other cases (as frequently with cocunut-palms) cattle may graze under mature stands.
204
The importance of interculture varies with the type of crop, the farm size, and price relations. lnterculture is generally of importance in smallholdings where the farmer, by practising arable farming in the early stages of shrubcrop or tree-crop development, may significantly reduce the cost-s of plantation establishment. This may be more important economically than the reduction in future returns which is likely to result friom poorer development of the dominant crop. (c) Vegetation andproductioncycle. Another variable peculiar to perennial crops is the length of the production cycle and the distribution of input requirements and returns over several years. MacArthur (1969) summarizes this as follows: During the early stages there is a slow but accelerating build-up towards maturity, when maximum yields are obtainable. A high level of production is thereafter obtained for several years, but ... age ... begins to affect the levels of production, and yields fall off. With tree crops a fairly distinct type of production pattern can be discerned, but with crops whose natural form is that of a bush or shr-ub, prcductivity can be renewed through pi uning back almost all of the growth above the ground and permitting new main stems to grow up from the capitated main stem. This generalized scheme is subject to many exceptions. Sisal follows a peculiar pattern wherein leaves grow and can be harvested for a given period until the plant puts up a stem with a flowering head, after which no new leaves are produced and the plant dies. Pineapple and sugar-cane grow in a series of 'ratoons' or successive waves of new shoots which grow up from the root stock, and the yield level falls steadily after the first harvest. ... this ... usually implies that there is a period during the early stages of growth (which can last for several years) when annual costs wil! exceed the value of harvested produce, but eventually a stage will be reached when new costs each year will be less than the product value. This phase may persist for a very long time, but there will come a stage... when production will fall to such a low level that its value is insufficient to cover annual production costs. At this stage, the crop's useful life will be over. With sisal, there is a firm limit fixed by the botanical behaviour of the plant, whilst with other crops like pineapples... and sugar-cane the point of unprofitable diminishing output is encountered after only a few years. Theorelically, it may be possible for naturally or artificially self-regenerating crops like banana, coffee and tea to remain in profitable production almost indefinitely, but most plants of these species, if they do not... fall victim to a disease... are eventually superceded by new varieties of the crop whose yields exceed those of established plants by so much that they merit uprooting and replacement, despite the interruption in producticn that this means. Consequently, in the case of shrub and tree crops we distinguish between the following stages in plantation development: (1) The period between clearing and planting and the beginning of the crop's useful life (in the calculations in Table 8.2 for cocoa, rubber and oil-palms, this period is four years).
205 SYSTEMS WITH PERENNIAL CROPS (2) The time between planting and the year after which the annual costs are covered by the annual yield (in the examples five to six years). (3) The period between planting and the year after which tile capacity yield is produced (in the examples eight years). (4) Tile period of capacity production, which begins when the first full yield is produced. In the following years, a regular yield, sometimes rising or falling slightly, is produced. This period comes to an end when the yield tends to decline more rapidly. (5) The length of land occupation. According to the level of running costs, the availability of land for planting, the cost of replanting, and the returns of new plantations in relation to old ones, it is sometimes advisable to continue using the land beyond the period of full production, i.e. in spite of rather low yields. The period of land use thus extends from the first to the last harvest. (6) The growth cycle, i.e. the time between creation of the plantation and clearance of it or letting it run wild. Plantation farmers may influence the economic length of the production cycle of their various perennial crops. Extension of the time in nurseries, special husbandry techniques, proper weeding, and manuring may reduce the initial period before a crop is obtained, while interculture may delay development. Pruning and harvesting techniques may hasten or delay returns. Husbandry practices in general influence the economic life of perennial crops. With sugar-cane the grower has to decide the number of ratoons. With shrub and tree crops in general the grower has to choose between production cycles of various lengths, and has to decide when to cease cropping and to plant afresh. Short production cycles are frequently, but not always, an indication of intensive farming. (a) Rotations. When a stand of perennial crops is created, the way in which the plot is used is fixed for a long time. An annual change in the use of the plot, such as we find in arable farming with crop rotations, is only possible to the extent that interculture can be practised. In perennial cropping longterm rotations do occur, and the following are the basic types: (1) Perennial crops may be followed by a lengthy forest or bush fallow. Thus old cacao plantations in West Africa revert to bush vegetation, whicn is then cleared after one or two decades, The rotation runs: fallow-rain-fed farming-cacao-fallow or rubber (Blanckenburg 1964, p. 38). Planned reforestation after coffee cultivation is practised occasionally in Latin America (Sick 1969, p. 164). (2) Perennial crops may be followed by grassland. Thus in southern Brazil old coffee plantations are chopped down and the ground becomes covered with grass and serves as rough pasture. in tropical high-rainfall areas, an
206
SYSTEMS WITH PERENNIAL CROPS even more negative sequence can be observed: exhausted coffee plots gradually turn into fields of useless hmperata grass. (3) Perennial crops may be followed by arable farming. The Kavirondo of Nyanza, Kenya, for example, cut down their banana groves after about ten years and sow the plots with millet (Wagner 1956, p. 29). (4) One perennial crop may be followed by another. The change may occur by uprooting and replanting. On some banana plantations in Central America, for instance, cacao or oil-palms are planted after bananas. It may also occur by under-planting the outgoing crop by a ntw one. Thus, in some estates in Ceylon, old rubber is underplanted with cacao. The rubber slowly gi..cs ".,, a pure stand of cocoa. With declining cocoa yields the area is cleared and planted with rubber again. (5) A perennial crop, or a mixture of perennial crops, may be renewed by continuous planting, as in the coffee plantatior. of El Salvador, or the coconut plantations in Ceylon.
The sequence is determined partly by the characteristics of the particular crop and partly by the availability of land and price relations. Where there is plenty of land, one of several growth cycles of perennial crops are followed by many years of forest or bush fallow. As land becomes more scarce, there is a tendency to replace the fallow by use of the land for pasture or arable crops, or perennial crops may follow each other. Where it is technically possible, continuous replanting (infilling) is a sign that land is particularly scarce.
8.4.
crns The basic farm-management characteristics of perennial-crop production are best depicted by reference to existing types of farming in both the estate and the smallholder types of economy. 8.4.1. Sugar-cane holdings Commercial sugar-cane production is carried out in three distinct exploitation systems (see Table 8.3). Firstly we have large-scale units combining cane production and a factory for processing. Secondly we find plantations of large and medium size producing mainly or exclusively cane for sale to factories. Thirdly we have smallholder production, where cane is produced in addition to subsistence food. Large estates are often surrounded by a belt of smallscale producers, who sell cane to the factory and provide extra labour for cane-cutting. (a) Sugar-cane estates in Afauritius. The examples from Mauritius in Table 8.4 illustrate the farm-management characteristics of sugar-cane production. Of the 91 000 hectares of land used for agriculture on this island,
207
84 000 carry sugar-cane, -ind 23 estates equipped with factories occupy onehalf of this area. The other half is cultivated by cane-selling farmers. The major part belongs to a few large estates .r, the rest is divided up into 24 000
TABLE
8.3
Country
Hawaii Peru United Arab Republic British Guiana Florida (United States) Barbados Taiwan Indonesia South Africa Trinidad Australia Puerto Rico Mauritius Jamaica Fiji Mexico Reunion Louisiana (United States) Cuba Brazil Argentina India Source: Frank (1967, p. 83).
Predominantly in estates; irrigation Mostly in large estates; irrigation Large estates; some cooperatives Very large estates; irrigation 80 per cent mill-owned estates; 20 per cent peasant farms One.third mill-owned estates; balance is peasant production 75 per cent large estates; 25 per cent peasant production 25 per cent factory estates; 75 per cent small and medium-sized farms 50 per cent estate production Mostly medium-sized independL.,' farms Mostly peasant production 50 per cent mill estates 55 per cent mill estaes Mc3tly small farms; cooperative har,/esting Small growers organized into large cooperatives 46 per cent large estates Medium-sized holdings As of 1961, 35 per cent cooperatives and 65 per cent state and individual farms No more than 50 per cent mill-estate cane (by law) One-third estate-grown cane Mostly smallholdings; larger-scale irrigated farms in the south
smallholdings each with an average area of about one hectare. Each estate crops an average of 1800 hectares of sugar-cane, and of this, 200 hectares are newly planted and 1600 hectares are yielding cane. A number of basic farmmanagement features of sugar-cane estates can be observed:
208
(1) A high degree of intensity. Sugar-cane estates constitute an intensive form of tropical land use. In the example of Mauritius, the gross return amounts to $500 per hectare, which is several times the return from annual
TABLE
8.4
Mauritius 1959 Sugar-cane Average of total 23 1826 70 77 (cane) n.a. n.a. n.a. 5 13c 425 8% 18% 34% 17% 8% 7% 8% n.i.r 88 59 513 733 380 545
Ivory Coast 19%52 Banana Case study 1 60 100 30 (bunches) n.a. n.a. n.a. n.a. 6080 10% 20% 19% 23% 7% 16% 5% n.a. n.a. n.a. n.a. n.a. n.a.
Tanzania 1964 Sisal Case study 3 1200 35 12 (fibre)h 550 45% 55% 151
151r
20% 7% 17%
-
Yield of sisal in cutting age. c Investment in 1964, without factory. d Not including interest payments. Including processing and transport to harbour. Including factory processing. 'n.i. = not included. Sources: Mauritius, Commission of Enquiry (1962): Ivory Coast, Champion (1963, p. 206); Tanzania, Pdssinger (1967, pp. 16-22).
arable crops under these conditions, and this in spite of the fact that in Mauritius a relatively extensive form of cane cropping is carried on, with
209 SYSTEMS WITH PERENNIAL CROPS (2) The preservation of soil fertility. Sugar-cane is mostly grown in monoculture, and for more than a century on some plantations. Cane production involves the removal of large amounts of plant nutrients from the field. It may nevertheless be assumed that cane growing aids soil fertility, for the following reasons: (i) Dense vegetation covers the soil for most of the year. (ii) Large quantities of organic matter are conveyed to the soil in the form of trash, and by an extended root system, supplemented by plentiful fertilizer (formerly manure produced by the many oxen required for transport, but nowadays mineral fertilizers). Burning, however, represents a possible danger to soil fertility. It has been introduced into high-wage economies as harvesting has been mechanized. (3) High labour input. Labour costs are the greatest expense on sugar-cane estates. In the examples from Mauritius, 59 per cent of the total costs in 1959 consisted of wages and salaries; 70 man-equivalents per 100 hectares were employed. In contrast with most other perennial crops, operations with sugar-cane, including harvesting, can be fully mechanized, and they actually are in high-wage economies like Hawaii, Queensland, and South Africa. However, the estates in Mauritius in 1959 still relied on seasonal labour to cope with the peak in iabour demand during cane-cutting. (4) High capital input. The economy of sugar-cane estates is highly sensitive, to transport costs, because cane is heavy, bulky, and perishable. Large sums must be invested in field tracks, roads, tractors, lorries, cranes, etc. (5) The development path. Sugar-cane cropping has shown itself able to benefit from yield-increasing innovations to which sugar-beet is far less responsive: improved varieties, improved disease control, sprinkler irrigation, increased application of mineral fertilizer, improved cultivation practices, etc. The subtropical cane growers are mostly ahead of the tropical producers in applying these techniques. Thus, if the trend continues, the yields per hectare are likely to carry on rising, and proven mechanical methods to be adopted where the wage level makes it necessary. Sugar-cane cropping tends towards monoculture, although yields from rotations are usually higher. The practice of incorporating a few years of field grass between the cane-cropping cycles, which was common when the estates depended on large numbers of oxen and therefore on a fodder and manuring system, has become rare now that tractors and mineral fertilizers have been introduced. Where the cropping cycles are interrupted by arable crops, we find either green-manure crops on large estates or intensive crops in smallholdings, like tobacco in Puerto Rico and rice in Taiwan. The trend towards monoculture is closely linked to the development of larger processing plants and the transport problems that they entail. Thus.
15
210
in Mauritius for example, the sugar factories in 1832 had an average processing capacity of 200 tons, in 1903 of 3000 tons, and in 1961 of 24 000 tons. As the capacity of factories has been increased, and when the plots are near the factory and transport is easy, other crops are even less able to compete with sugar-cane. (b) Sugar-cane sinallholdings. Sugar-cane is definitely a crop that is most suitable for estates, because it is bulky and perishable, and costs limit the distance over which it can be carried. The modern factory should therefore stand in the centre of a relatively compact cane area, where cane is grown in monoculture. Monoculture, however, rarely fits into the working pattern of smallholdings. Cane cutting e'ates peaks in labour demand that can hardly be tackled by the labour capacity of a smallholding. The organization of cooperative cutting is usually a very difficult and intricate undertaking. Also, cane production must be accompanied by thorough cultivation, which is best carried out with heavy tractors, and cane is very susceptible to poor husbandry and numerous diseases. It is therefore more important than with other crops that technical advances should be available to the estate manager. Moreover, the smallholder cannot establish a sugar plantation as cheaply as he can a tree or shrub plantation. The superiority of large-scale producers of cane was so great in, for example, former Dutch Java that the smallholders recognized the advantage of leasing land to estates. Smallholder production of cane has proved to be competitive only: (1) in satellite holdings near an estate-often on adjoining marginal land, as for example in Mauritius, where this type of holding produces cane and is also a source of seasonal labour; (2) in an area with a good transport system, where the smallholders are settled as tenants and are subject to closely supervised contract farming, including supervised cooperative harvesting, as in the Fiji Islands in the South Pacific; or (3) in places where the smallholders have technical training and farm very intensively, as in Taiwan. Thus if estate production is not feasible or desired, sugar-cane can be grown in medium-sized holdings (cultivated by tractor) more economically than in smallholdings. Such medium-sized holdings are typical of production in, for instance, Queensland, Australia. 8.4.2. Sisal holdings Sisal, like sugar-cane, is a typical estate crop, and for sisal estates are part of larger firms operating several similar reasons. Most undertakings. Small holder sisal, except in the north-east of Brazil, is economically hardly significant.
211
(a) Sisal estates in Tanzania. The growth cycle of the sisal plant and the processing capacity of the factory determine the organization of a sisal estate. A decorticator has a given annual capacity to process leaf into fibre (1500-2400 tons). The object of planting is to guarantee regular delivery of the right amounts of leaf during the year, and for year after year. The growth cycle of sisal, in a well-organized holding, lasts nine years. The plantation must therefore have nine fields, in which to have sisal at all stages of growth, to guarantee regular delivery (see Table 8.5). Under favourable conditions of production, one hectare produces 12 tons of fibre in nine
TABLE
8.5
Total tonnage of leaves per year
-
1
2
-
125
125
3
4
0"8
1"5
4500
6750
100
188
5 6 7
8 9
1500
Area of sisal (ha) Area fallow for replanting (ha) Area for nurseries, roads, etc. (ha) Total area of plantation (ha)
The calculat:on of the holding programme is based on a yield of 12 t ha- of fibre produced over 6 years. The calculation refers to an unusually well-managed holding. Source: Lock (1962, p. 317).
years, and, with this level of yield in the example, nine fields of 125 hectares each are necessary if a decorticator with the capacity of 1500 tons is to work to full capacity, i.e. the area of sisal must be 1125 hectares. There are also seedbeds, etc. Thus an efficiently organized sisal holding must crop 1200 hectares, and where yields per hectare are lower the necessary size of field and plantati i is correspondingly higher. Most sisal estates have fallow land in reserve as well as cropped areas. The following features are typical of sisal estates, (1) There is a high initial investment in the factory, the water supply to the factory, plantations, equipment, and buildings. According to the data of Guillebaud (1958) and Pssinger (1967) for the years 1958 and 1964, an investment of S550 was required to produce a ton of fibre per year.
212
(2) The transport costs are high. 40 000 tons of leaf have to be taken every year to the decorticator, which means a journey of 160 000 ton kilometres at an average of 4 kilometres from field to factory. Consequently fields have to be concentrated round the factory and transport routes. (3) Declining soil fertility is a major problem in this type of farming. Yields tend to fall seriously after several rotations of sisal, and, because of the price of sisal fibre, it would not at the moment pay to apply mineral fertilizer. However, recently developed hybrid sisal could transform this situation by changing sisal from an extensive to an intensive crop, producing two to three times the customary yields. (4) A main technicHl problem is the achievement of a high yield of fibre per ton of leaf. According to the quality of the leaf, but mainly according to how the decorticator is adjusted, the return varies between 3 per cent and 5 per cent of leaf-weight. Estate managers usually pay most attention to the processing of the leaf, and regard planting as a secondary activity. (5) The main labour tasks on a sisal estate-replanting, weed control, and harvesting--can be spread evenly over the year. According to Lock, in 1959 a 1500-ton estate needed 1200 workers, or 100 workers per 100 hectares in an estate with high efficiency per hectare. Sisal cropping is more amenable to mechanization than most other perennial crops (for controlling weeds, loading leaf, etc.), and in 1965 in Tanzania 35 man-equivalents per 100 hectares of sisal were used, compared with 70-10 man-equivalents ten years previously. Nevertheless the main operation, cutting of the leaf, must still be done by hand. (6) Harvest fluctuations from year to year are relatively small. However, price fluctuations do affect this export crop, perhaps even more than almost any other. Falling prices in recent years, due to the competition from artificial fibre, may force sisal producers with high costs-largely those with yields less than one ton per hectare-to change to other activities. They may turn to cattle and a variety of crops such as cotton and maize, sunflower and rain-fed rice, or abandon farming altogether and leave the land lying idle. (b) Sisal smallholdings. Sisal production in smallholdings is found on an important scale only in the north-east of Brazil, where sisal is intercropped with maize. Because the yields of maize are uncertain, sisal serves primarily to counterbalance the risk of a poor harvest. The sisal is processed with small, mobile decortication machines (raspadors), which produce unbrushed fibre. The further processing is done in brushing and baling factories. The produce is of poor quality, however, and the labour input is high. This applies even more to smallholder sisal in East Africa, where hedge sisal is processed by hand into unbrushed fibre. With falling world market prices, production is no longer competitive and has ceased almost altogether. Efforts to introduce
213
sisal as a settlement crop have been made in Kenya and Tanzania, but without a reasonable chance of success (Pssinger 1967). The reasons for the poor performance of smallholder sisal are even more obvious than with sugar-cane. Sisal production without a processing factory yields low-quality produce at very high labour inputs. Smallholder sisal in combination with factory processing is extremely difficult to organize because of: (1) the rotation cycle in cropping (see Table 8.5)-a smallholder would need ten separate fields; (2) the need for heavy equipment in establishing the plantation; (3) the peak in labour demand for cutting, because sufficient leaf for truck transport has to be cut in a given day; and (4) the bulky nature of the harvested produce, which is perishable and has to be delivered in definite quantities at definite times. 8.4.3. Banana holdings Bananas are not only one of the most important food crops in the tropics, but also an outstanding export crop. They combine many of the advantages of annual arable crops and perennial crops: (1) The gross return per hectare, measured in calories or money, is usually high. The input for clearing the land is low, establishment of the plantation is relatively simple, and the first yield is produced twelve months after planting. The banana does require a high degree of humidity, but apart from this it is adaptable, and will even grow on soils of poor fertility and in cooler mountain locations. (2) The adaptability of the banana stems from the great number of varieties, which not only are suitable for differ.nt purposes (for cooking, beer; and use as a fresh fruit), but also have varying location requirements, and grow to different heights (the tall-stemmed banana and the dwarf banana). Bananas also yield valuable by-products. In southern India, for instance, green leaves are sold to urban households, where they serve as plates. Elsewhere, dry banana leaves serve as roofing material. (3) By providing plenty of shade, trash, and permanent cover on slopes, the banana helps to maintain soil fertility and reduces erosion. It makes the application of large amounts of fertilizer worth while and responds well to fertilization with mulch, manures, and mineral fertilizers. Bananas may occupy a plot permanently. (4) The yield of bananas is produced relatively regularly, and where they are for local consumption there is a supply of food all the year round. In the special case of the Ensete banana, Ethiopia (Ensete edula Horan), the starch is stored in the stem, and, as with manioc, there are starch reserves available in the living plant.
214
(5) The labour requirement is high, varying from 0.5-3.0 man-equivalents per hectare (0.3-1 .0 man-equivalents in plantations), but there are no pronounced labour peaks, and the ratio between labour input and the return of calories is usually better than with most annual food crops. (6) Finally, the banana is compatible with other crops, facilitating both mixed cropping and its incorporation in a rotation. On the other hand, there is the danger of damage by the wind, and the high cost of controlling banana diseases. The Panama disease is reported to have increased production costs in several parts of Latin America to the point when export production is no longer competitive. The fact that the fruit is perishable is another drawback, and processing and conserving are scarcely worth-while. Whereas sugar-can' plantations tend towards monoculture, banana cropping usually leadb to diversified farming. In large plantations it is usual to plant tree crops like oil-palms, rubber, or cocoa immediately after a few years of bananas. Smallholders usually combine banana production for sale and subsistence with other activities, either in separate blocks or as mixed cropping. It is noteworthy that there seem to be rather pronounced complementary effects between banana production in smallholdings and animal husbandry. Cattle ard pigs feed on banana leaves, stalks, and fruits, and their manure is of high value in increasing banana production. (a) Banana estates in CentralAmerica. Until recently, bananas for export were grown mainly on large estates, because it is necessary to deliver large quantities of a perishable commodity of uniform quality at the right time. These conditions can be fulfilled more easily in large-scale than in smallholder produc, ion. The management characteristics of a large banana estate in Honduras are described by May and Plaza (1958). A typical banana plntation in Central America belonging to the United Fruit Company comprises a banana-growing area of 8000 hectares, together with extensive land reserves and waste land. Only the best soils are planted with bananas, and the workers are allowed to use the remainder as they wish, or it is left as waste land. The holding as a whole is split up into twenty-two or twenty-three sections, each of 350 hectares. The terrain is divided up like a chess-board and drained if necessary. The cleared vegetation is not burned. Bananas are planted in rows on the cleared land. Usually there are two pipe systems (which are permanent), one for sprinkler irrigation, the other for spraying pesticides. The aim is 750-800 fruit-bearing stems per hectare, with a yield per hectare of 27 tons stem-weight or 23 tons fruit-weight. The United Fruit Company works on an average loss of 25 per cent cf the harvest. Harvesting takes place at all times of the year, and the bunches of bananas are carried from the tree to padded carts or pack donkeys, after which they are taken to collecting points where there is a rail connection. The bunchL- are sorted, washed, dried, packed, and hung in special trucks to be transported. The permanent workers in one section with 350 hectares of bananas number 150. Harvesting all the year I Weight of banana bunches including the fruit stem.
215 CROPS SYSTItMS WITH PERENNIAL so that about eighty men are occupied sixteen round permits a high degree of specialization, with plant prottction, five with watering,
replaced by mainly rnediumIn recent years, large-scale units have been bananas. In Latin Amneri':a and West sized estates with 10-100 hectares of all the bannaars are i~creasingly Africa, the people who produce and box chiefly as rte.keting agent, and small entrepreneurs, the company acting Table 8.4 (p. 208) presents the providing loans and technical assistuace. in Ivory Coast. The high financial farm-management data of one such holding iz pai-ticularly noteworthy. outlay on fertilizers and disease control or Bukoba, Tanzaniaq (b) Banana-coffee smalffioldigs: the Wahaya of the main activities one markets is Banana production for food and local rainfall and temperature are sufficient of tropical smallholders, wherever is the banana-coffee culture of the for banana growing. A case in point because of the intelligent adaptation Wahaya, which is particularly interesting natural conditions of the location. of banana production to the difficult well-distributed rainfall averaging 200 The Bukoba. District receives ample and are badly leached and very poor. The millimetres annually. The soils, however, of which 0.80 hectares are hoinesteads but average holding comprises 1.4 hecta,"es, are usually used as subsistence food,yield which carrying mainly cooking bananas,far,hich away as Dar-es-Salaam. Bananas, coffe, which are also sold to markets Ps are interplanted with Coffea robusta rcivin, of hectares 0-6 more than half of the gro.3s includes holding the maize, and beans. In addition to the homestead, open, infertile grassland. The so-called homestead (1), Fig. 8.4 shows the layout f a Bahaya holding. ia mixed stands, is sitiated other crops a banana plot with coffee and several homestead there is usually a garden the of around the hut. At the edge At some distance from the homestead growing predominantly root crops (2). the farm. Here small plots are cultivated lies the grassland (3) belonging to to serves as psture and supplies mulch on a short-term basis. The grassland' everyone that belongs to the farm, the homestead. In addition to the grassland Within the homestead both a spatial cn make use of communal grassland. (c)): is recognizable (see Fig. 8.4 (b) and and a vertical order in plant growth cattle kraal, the holding focal point (1) Directly around the hut and the manure, a dense growth of bananas where the soil has received the most away from the hut, the less manure can be found. The greater the distance thiinner. More and more coffee tre-. is applied. The bananas become the bananas until the belt of bananas maize plants, and beans grow between gives way to a belt of banana-coffee around the centre of the holding of banana-coffee-beans is succeeded cultivation. Gradually the mixed stand or coffee-beans. In some cases, by one consisting of banana-maize-beans a plot with pure stands of new by the belt of mixed stands is supplemented
216
(a)
(a)
__-(a)
--
-(1)
- -(2)
la
b)(b)
Hut
(b) trees
%%ee-bea,
,~~nas
e~aa Pasture
[lut
Cc)U
(1CHut
(3)8anan,
(5)Maize
(7)Vegetables
FIG.
(2)Boina (4)Coffee (6) Beans 8.4. Land use in a Bahaya holding, Bukoba, Tanzania: (a)layout; (b) horizontal order; (c) vertical order. From Friedrich (1968, p. 175).
217
coffee. Land use is thus organized in concentric circles around the central hut, the bands ranging in width from a tew feet to several dozen yards. (2) Besides the spatial order of plant growth, there is a vertical one. On the heavily fertilized soil close to the house, the bananas reach unusual heights, casting dark shadows upon the ground, where only some arrowroots grow. Bananas close to the centre of the farm frequently grow twice as high as those growing on the less fertile edges of the holding. With the distance from the hut increasing and the supply of manure and mulch material growing less, and at the same time the heights and shading of the bananas decrezsing, more coffee trees can be found growing between the bananas. If the density of the banana-coffee vegetation permits it, the ground is covered with beans, so that, in accordance with their need for light, three crops are growing one above the other: beans, coffee, and bananas. Arrowroot, yams, tobacco, maize, and vegetables are aiso grown where possible, in such a way that on different plots different crops are grown, which are compatible with regard to their light requirements. At the same time, the ground is given complete protection by a thick layer of mulch. Naturally, the various belts around the hut are not kept strictly separate; instead, we find a gradual transition in the combination of plants, which may vary from one location to another, from one farm to another, and even on any one farm, according to the nature of the soil and the site of the slopes. The determining factors in the cultivation of the homestead are the fertility of the soil, domestic needs, the land available, and-occasionally-the influence of extension officers. The way mixed cropping is apportioned depends primarily upon domestic requirements. In cases of large families with small plots of land, bananas are given priority, and there is little room left for coffee. In contrast with this, there is proportionately much coffee where a family has at its disposal a relatively large homestead. Bananas at Bukoba are a permanent crop. Some of the homesteads have carried bananas for more than sixty years, and this permanency of farming is based on intensive manuring. The export of nutrients is lim;ted to some dried coffee. Instead, there is a substantial import of nutrients by the gathering of fuel, the purchase of food, the use of ample mulch collected from the open grassland, and the fact that the cattle are kept in stables and their manure systematically applied. The build-up in soil fertility is the more marked the older the homestead. The man farming an old plot, which in the course of time hr3 received much refuse and mulch, can rely on his bananas having a denser growth and a richer yield than a man cultivating a new plot. One-third of the farmers own cattle, usually 3-4 head. The cattle graze on the open grassland and produce mainly manure for the banana land. The production of milk and meat is very low. Friedrich (1968, p. 159) gives the following information: calving interval 25.5 months, weaning loss 30 per cent, daily milk production
SYSTEMS WITH PERENNIAL CROPS during the six-month lactation period about 1.5 kilograms, and 20 per cent of the cows are probably sterile. The average milk production per crw and per year, over and above calf-rearing, is thus estimated at 50 kilograms. The most important product of the cattle economy is obviously manure (see Fig. 8.5). On farms where
218
(2)
1000
I I I I I
5.0
Fo. 8.5. The relation between farm size and gross return in Bahaya holdings with and 2 without cattle. 2
(!) Farms with cattle: R = 0.98, n = 16. (2)Farms without cattle: R 0-51, n = 18. From Friedrich (1968, p. 204).
=
cattle are kept, the gross return per hectare of banana-coffee land is on average $200 above that of farms without cattle. If we assume-certainly not entirely correctly--that the additional a' zrage gross return on farms with livestock is wholly due to the manure, we arrive at an annual manure value per livestock unit of $78 (130 shillings). This figure may explain the Wahaya's interest in obtaining cattle, and thus manure. Table 8.6 shows the economic return of Bahaya farming, which is higher than in most other farming systems of Tanzania. The more successful farmers are quite obviously those who produce more per hectare. 58 per cent of their labour capacity is employed, as compared with the average of 37 per cent.
8.5.
Coffee and tea are the main representatives of a distinct group of crops that notanically are tree crops but because of their farm-management characteristics are termed shrub crops. The main difference in cultural practice between these and tree crops is the high input for weeding and pruning. Both aspects are tied to the shrub-like appearance of the plants.
SYSTEMS WITH PERENNIAL CROPS 219 8.5.1. Coffee holdings The production of arabica coffee in most countries was usually started on estates and then spread to smallholdings. A number of the farm-nianagrment features of coffee promoted this patiern of development: (1) Coffee growing is adaptable to wide differences in farining intensity and to various forms of mixed cropping and intercropping.
TABLE
8.6
Average of even
Workers per holding (ME) 2.12 Size of holding (ha) 1.50 Banana-coffee land (ha) Field crops (ha) Banana-coffee as % of total area cropped 94 LU per holding 4-3 Coffee trees (no.) 150 Economic rertrn ($ per holding) Gross return 539 Banana Coffee Field crops Cattle Expenses for purchased material inputs 7 Income 532 Costs of hired labour 4 Family income 528 Productivity Gross return of banana-coffee land ($ ha- ') 380 Gross return (S ME- 1 ) 254 Value added (S per ha of banana-coffee land) 373 Labour utilization as %of available labour 58 Labour input (hours per ME) (arable crops, permanent crops. and animal husbandry) 1780 Source: Friedrich (1968, p. 206).
(2) Coffee production at low levels of yield is not very susceptible to plant disease. (3) By far the most important input in coffee production is manual labour. (4) The produce may be dr'ed on the farm or processed in small pulperies. Transport of cherries to large pulperies is a problem, but by no means comparable with that of transporting sisal leaf or sugar-cane. (5) The introduction of coffee into subsistence holdings does not usually lead to pronounced peaks in labour demand.
220
Because coffee cropping suits small-scale farming, expansion in various parts of the world since the 1930s took place primarily in smallholdings, as in southern Brazil and East Africa. On the other hand, estate production has recently gained in competitiveness, owing chiefly to the more rapid application of technical advances like improved varieties, mineral fertilizing, plant protection, pruning techniques and processing.
(a) Coffee estates in southern Brazil and Kenya. The coffee estates in southern Brazil are mostly a few hundred hectares in area and are run on a wage or contract basis. Holdings normally have at their disposal: (1) a large coffee plantation, which occupies 15-30 per cent of the overall area; (2) a large pasture consisting of 40-60 per cent of the holding, supporting about 50 head of livestock per 100 hectares of grassland; (3) arable land for the workers as payment in kind; and (4) odd pieces of forest or waste land. In some cases we find arable land cropped with maize, non-irrigated rice, legumes, cotton, etc., a special crop like sugar-cane or onions, or a marketorientated dairy herd. The different branches of the holding are cultivated extensively. The coffee stands are mainly very old and yield low returns. Fertilization with animal manure and mineral fertilizers is limited to progressive holdings. When the one harvest takes place, ripe, unripe, and overripe cherries are all picked together and prepared 'dry'. A commodity of moderate quality is produced on a large scale, the coee yield per hectare being on average only 4 quintals, although in the examples in Table 8.7 it reaches 7-12 quintals. The efficiency of the grassland is that of rough grazing rather than pasture, and the milk yields per cow exceed 1500 kilograms in only a few cases. The holdings in Table 8.7 illustrate the ways in which this holding system is developing. In the case of the holding with 668 hectares, the purchasing power of neighbouring -owns has caused dairying to be expanded, so that only one-half of the holding return is derived from coffee. hi the holding with 454 hectares, the growing urban demand has led to a special crop, in this case onions. In other holdings, like the one with 716 hectares, coffee cropping has been modernized, a step reflected in younger stands and higher efficiency per unit area. The introduction of mineral fertilizing in conjunction with sprays has given coffee the power to compete again with other forms of land use. Coffee estates in other parts of Latin America and in Africa are usually less diversified than those in southern Brazil, where urban purchasing power is comparatively high. In Kenya for instance, most coffee estates are specialized undertakings. In the case study depicted in Table 8.10 (p. 229), half of the farm area is pianted with coffee, the rest being waste land, or producing mulch material.
221
Coffeecatta,,
CoffeeCoffeearable crops cattle Case studies 716 27% 2% 31 % 20% 200/ 454 19% 16% 44% 7% 14% 23 41 1200 1800 12k7 1237 468 109
17
22 900 n-a
1000
n.a.
361 n.a.
1515 309 68
90 92% 8 j 70% 2% 9% 29
177 93%
70/
205 44%
46%
94 500%/0
15%
Livestock Total expenses (ha) Wages Equipment Fertilizers, seeds, etc. Net return (ha)
61
122
J
66% 4% 20%
137
86
55
68
Productivity Gross return 94 205 177 90 ($ per ha utilized) 470 890 1040 900 ME-') ($ Gross return Value added 37 143 166 72 ($ per ha utilized) 185 623 975 720 ME-') (S added Value a Case studies nos. SF-40, LF-160, MF-2, MF-I. 250 man-days per ME. land + pasture, not The labour force is related to land utilized, i.e. coffee land + arable labourers. for land or bush including Almost all are catt!a. household, or renting Not including production on labourers' fields, food to the farmer's value of the farmhouse. Source: Food and Agriculture Organization (1960, part I).
222
SYSTEMS WITH PERENNIAL CROPS Considering the farm-economic data of the coffee estates in southern Brazil and Kenya (see Table 8.7 and 8.10), we may summarize their farm. management characteristics as follows: (1) A high degree of intensily per hectare planted with coffee, with gross returns far above most arable crops under similar circumstances, and usually also above most tree crops that suit the location. The duration of the cycle is an indication of the degree of intensity. With extensive cultivation, the coffee stands are replaced after twenty-five to thirty-five years; with intensive cultivation they are replaced after twelve to eighteen years. (2) Soil fertility is not as easily maintained in estates with monoculture as in smallholdings with a banana-coffee mixture. Intensively mulched and manured coffee plantations in Guatemala are reported to produce coffee for one hundred years without reduction in soil fertility. Extensive coffee production, however, frequently results in soil mining. (3) Manual labour and salaries are important input items, requiring about half of the total costs. The level of coffee output per hectare is closely related to the level of labour input, since the extra work that is involved in applying intensive husbandry techniques (see 8.5.1 (c)) may result in substantial increases in quantity and quality. (4) The capital input, except in trees, roads, and pulperies, is low compared with most othur estate crops. (5) Estate production with a igh level of yicld is exposed to considerable production risks: frost is a particular hazard in southern Brazil; coffee berry disease has brought great losses to the planters of Kenya. (b) Coffee smallholdings in Tai-aniaand southern Brazil. Both robustaand arabicacoffee are extensively grown in smallholdings, wheie coffee production is combined with various other activities. Three distinct types may be distinguished: (1) Coffee may be grown under a cover of natural frest, and rarely mixed with food crops. This is true fbr most of the robusta production in West Africa and the arabica production of Ethiopia. (2) Coffee may be grown in a mixture wiih bananas and various other food crops. This is true of some robistacoffee and of much of the arabicacoffee in Latin American and East African smallholdings. Mixed stands are frequently the outcome of introducing coffee into fallow systems. (3) Coffee may be grown in pure stands, except for some interculture in the early stages of plantation development. This is true of smallholder coffee in southern Brazil and Kenya. The following is an outline of the farm-management characteristics of holdings growing coffee in mixed and pure stands:
223
(1) Coffee-banana holdings on Mount Kilimanjaro, Tanzania. In the course of five decades an important and particularly interesting farming system has developed in the coffee-banana belt on the fertile, well-watered slopes of Mount Kilimanjaro (see Table 8.8). These are small farms of
fABLE 8.8
Tatzania Machame, Mt. Kilimanjaro 1961 Sample 100 7.5 2.1 161 1'3 1.1 1.0 0.7 0.5 0.1 0.9 1.2 0.3 b 179 309 263c.d 38% 18% 6% 38% 64 199 45 154 65 202 125 153 95
Brazil Sao Paulo 1958 Case study n.a. 4.3 35 12'1 8.7 0'8
}10
0.8 3.0 n.a. 890 n.a. 3345 n.a. n.a. n.a. n.a. 877 2468 143 2325 n.a. 276 778 204 574
Including family members receiving other income and 0.5 man-equivalents of non-family labour. Average of 1960 and 1961. The sample year 1961 was exceptionally poor. d Without renting value of farmhouse. Sources: Tanzania, Beck (1964): Brazil, Food and Agriculture Organization (1960).
224
SYSTEMS WITH PERENNIAL CROPS about 1.0 hectare, on which coffee and bananas are grown in mixed stands. In addition, a few miles lower down the mountain farmers cultivate a plot of maize. Most of the farms keep some cattle and goats, which are stabled throughout the year. The animals are fed with the leaves and stems of bananas. Grass and hay is brought from the lower steppe, the women carrying bundles of it on their heads for miles. The goats feed on the leaves of hedges surrounding each holding. The main purpose of this system of animal husbandry is to obtain manure. The verious farm activities are integrated to P returna able extent. Bananas supply food, the raw material for the brewing of bet;, some cash, foddcr for cattle, and mulching material. The manure for both coffee and bananas comes from the cattle, which also supply animal proteins. Planting is arranged according to the light requirements of each crop. Light permitting, beans are grown nearest the ground, and above them the coffee trees, which in turn grow in the shade of bananas. The smaller the farm the greater the preference given to bananas, because under present methods of cultivation they yield in terms of money the highest gross return per lectare. .eck's (1963) study of 1961, a particularly poor year, shows that on averase, a 1.2-hectare holding of this type yielded a gross return of $263. The r.ain outlay was the wage bill for outside labour. One-thire! of the families derived an income from outside the farms, usually by working on nearby estates. The average family income in 1961 was quoted at $219. According to another more widespread but less systematic enquiry in 1959, the returns from coffee and bananas were considerably higher, even on smaller farms. Woithy of note are the beginnings of commercial milk production. The income from coffee has resulted in a greater demand for milk. Some farmers are buying Jersey cows from the estates in order to sell milk to their neighbours. The annual milk yield per grade cow was about 2000 litres in 1961. Some farmers buy additional bran and pyrethrum cake; they also grow grass for fodder, either in terraces between the coffee or in separate plots. If the cows are pro I erly looked after-so some of the better farmers sayfodder grass shows greater monetary returns per hectare than either coffee or bananas. Hence grass becomes another cash crop. The farmers are encouraged to grow their own grass because the purchase of hay from the steppe carries the danger of introducing ticks and thus east coast fever and other diseases. Disease-prone grade cattle are kept healthy by stablefeeding with home-grown fodder all the year round. The fact that this Wachagga farming is commercialized to such an extent has also led to changes in the attitude to land tenure. Land has become the private property of individual farmers. It is openly bought or sold, and can be used as security against government loans. (2) Coffee sitios in southern Brazil. Where coffee is grown in pure stands,
225
most smallholdings are highly commerciaiized, and tht, same husbandry techniques as in estates are applied with varying degrees of care. Economically important examples of this art. ihe coffce holdings in southern Brazil,
which account for the major part of local production. Table 8.8 contains the data of a holding typical of this kind, whose efficiency is above average. The area cropped with coffee is 3.7 hectares, most of which carries pure stands. Young coffee plantations are also used for growing maize, rice, and beans. 3 hectares of pasture support 5 cattle. Arable fanning and cattle-keeping serve only to cover household food requirements. The holding is operated by family labour. At harvest time, which is a relatively short period in this area (unlike other coffee-growing areas of the world), a number of seasonal workers are employed. The result of using a relatively small labour force is an extensive form of coffee growing. Four plants are put into one hole. The branches are not pruned, so that work is limited to control of weeds and harvesting. The tree is stripped when most of the cherries are ripe, and they are then placed in drying sheds. in this way coffee is produced at half the price of coffee in the holdings on Kilimanjaro. The cropping area and size of the harvest is disproportionately larger. The good results from the holding shown in Table 8.8 derive from the use of a relatively young plantation that is heavily fertilized. This is not usually the case in Brazilian sitios, the yields averaging only 4 quintals of coffee per hectare. However, there are signs of improvement, such as the intensification of dairying, which produces animal manure, or, as in our example, the development of smallholdings with heavily fertilized pure stands of coffee. (c) Husbandry techniques of coffee production in various countries. The comparison of husbandry techniques in various coffee-growing areas provides an idea of the socio-economic adaptability of coffee p,'oduction. This can be illustrated further by a comparison of labouv input, ;osts, and returns of production of arabica coffee in three Latin American countries and in Kenya. (1) Traditional coffee cropping in Sao Paulo. The aim in coffee cropping in Brazil is to produce large quantities of moderate quality, using a large area of land and as few workers as possible. The traditional method of farming is relatively extensive. Coffee is sown in the field and not planted. 800 seed holes per hectare are the norm, and as the plants grow they are thinned out, leaving two to six plants in each hole. The struggle for light causes the individual branches to bend outwards, and fruit-bearing branches to be formed without any pruning. Hoeing the weeds is the only major operation. Coffee occupies the plot for a relatively long time, normally four years, before the first yield is produced. The plot is used
16
226
SYSTEMS WITH PERENNIAL CROPS over a long period: two-thirds of tile trees in holdings in Sac, Paulo are more than twelve years old and half are more than twenty-five years, whereas the maximum yield is reached in the seventh to ninth year. People prefer, however, to have large areas carrying old stands with a low yield to save the cost of frequent replanting. The cherries are not harvested selectively, but when most of them are ripe. All the cherries growing on TABLE 8.9
Labour input per hectare of established coffee in several countries (hoursper hectare),
Country Exploitation system Weeding and decouching Pruning Care of shade trees Infillings Mulching, fertilizer application, and erosion control Plant protection Harvesting Sundries Total labour (hours per ha) Wage costs ($ ha-)bC Yield, green coffee (kg ha-') Price (s t-1) Gross margin ($ ha - 1) Wage costs as , of gross margin Gross margin (5 per hour of work) Brazil Estates 209 15 6 50 28 218 50 576 92 446 450 201 45 0.35 Columbia Smallholdings 254 63 29 24 10 37 372 10 799 104 523 704 368 28 0.46 1-1 Salvador Estates 219 227 151 86 200 67 566 50 1566 157 660 955 630 25 0.40 Kenya Estates 708 552 16 259 120 912 48 2615 158 840 875 735 22
0.28 a Direct costs for upkeep and harvesting only, without general charges, overheads, or processing. h Total labour input valued at local wages: Brazil $0.16 per hour; Columbia $0.13 per hour; El Salvador $0.10 per hour; Kenya $0.06 per hour. o Rate of exchange: Brazil, 100 cr = U.S. $1.32; El Salvador, source in U.S. $; Kenya, I shilling = U.S. $0.12; Columbia, I pesos = U.S. $0.234. Sources: Food and Agriculture Organization (1958, pp. 24, 28, 43, 45, 49, 50, 79, 105, 117, 138); (1960, part 1, pp. 20, 48, part 2, pp. 48, 55, 61); Kenya Coffee Growers Association (1968).
the tree are shaken or stripped off, and dried in the air on drying floors or in drying sheds. Air-drying is preferred to the pulping and fermentation method, because the latter requires selective harvesting. This in turn requires a high labour input at harvest time, and in Sao Paulo harvesting takes place in a few concentrated months, a situation that distinguishes it from coffee-growing regions near the equator.
2271 SYSTEMS WITH PERENNIAL CROPS average input of only 576 manThe Brazilian method demands an are no appreciable costs of fertilizing equivalent hours per hectare. There i3 hand, the yields are low, the quality and planting material. On the other of cost per hectare is small. The i.verage, and the gross monetary retuvn the gross return, so that per manlabour amounts to 31 per cent of of $035 (see Table 8.9). equivalent-hour there is a gross return above are on the way out. ProducThe husbandry techniques described plantations, with plants raised in tion is increasingly coming from new holes-four plants to each holenursery beds and planted in fertilized and mulch ace applied. There is, along contour lines. Mineral fertilizer harvesting, and the produce is stil! however, little pruning, little selective sun-dried (Nixdorf 1968). Columbia, coffee is produced mainly (2) Coffee growing in Columbia. In comprise 10 hectares of usable land in peasant holdings, which normally are in mountain locations, where with 3 hectares devoted to coffee. They high and land is correspondingly scarce. the population density is relatively of intensive than in Brazil. Instead Coffee cropping is, therefore, more plants seedlings, but only some of the seeding the coffee, the farmers plant suckers are still commonly used are raised in nursery beds. Adventitious form of planting possible). The for replanting (this is the most extensive per hectare is very high. The growing plant density of about 2000 trees also produce a yield, but where there stands are shaded by bananas, which shaded by trees that compete less are older stands they are sometimes than do bananas. There is not a great with the coffee for light and nutrients and and the only cultivation is weeding deal of pruning. Fertilizing is rare size. down to a manageable annual topping to keep the trees the year, and it is selective according Harvesting takes place throughout before undergoing fermentation, to ripeness. The cherries are pulped investment than the air-drying methods. which demands more labour and the necessary hand-tools and facilities. The holdings are equipped with per method, an average of 799 hours The labour input in the Columbian counterbalanced is input but this higher hectare, is far above that of Brazil, better quality. The labour costs considerably or yield by a somewhat higher return, and a gross return per manconstitute 24 per cent of the gross equivalent-hour of S0.57 remains. has Coffee cropping in El Salvador (3) Coffee cropping in El Salvador. great with raised of intensity. Plants are attained a particularly high degree bed. in a seedbed and then in a nursery care and a high labour input, first they when and they are planted out Only selected young plants are used, is this of is comparatively late. The point are one or two years old, which and period when no yield is produced, to reduce as much as possible the replanting can be avoided. The seedlings this is also how total clearance and chopped down, or they are planted replace trees that have already been
228
near trees with a poor crop, which are chopped down as soon as the new plants begin to bear fruit. In this way, stands are being permanently rejuvenated. Except in particularly high locations, the coffee grows under shade trees, like mangoes or avocados, which themselves often produce an economic return. The young plants are put in holes fertilized with organic or mineral fertilizer. The coffee trees are also treated with compost and mineral fertilizer during their fruitless early years. They are frequently and carefully pruned, erosion is controlled by terracing, harvesting is selective, and the wet fermentation method is used. The labour input of 1566 hours per hectare is three times as high as in Brazil, but compared with the data from Brazil it is compensated less by higher yields than by better prices. The wage bill accounts for 24 per cent of the gross return, so that the gross return per man-equivalent-hour is $0'40. (4) Coffee-growing techniques in Kenya. Coffee growing in Kenya is even more labour demanding than in El Salvador, although the production techniques are similar. The labour spent on weeding and decouching is particularly high. The pruning requires twice the number of work-hours as in El Salvador. The same is true of selective harvesting. The data thus indicate that the labour efficiency per man-hour is much lower in Kenya than in El Salvador. The higher input in man-hours is offset by lower wages. Toal wage costs in Kenya and El Salvador are almost at the same level. The same applies to prices. The gross margin per hectare, however, is higher, while the gross margin per hour of work is much lower. The data in Table 8.9, comparing labour input and yields from coffee plantations with different growing methods, must be treated with caution. They are average values from surveys of the Food and Agriculture Organization dating back to the late fifties. There are great variations of performance within each area. Moreover, the rates of exchange have introduced their own distortions. The value of the comparison does not lie, therefore, in any of the details, but rather in the demonstration of the gre.-t numbers of possible methods and in the fact that there are definite reasons for the cropping methods that have been chosen. 8.5.2. Tea holdings Growing of tea in suitable areas depends on the processing capacities of tea factories. In Ceylon and Assam, there are still many small factories in operation with a processing capacity of less than 110 000 kilograms of made tea a year. The tea area required by a factory of this kind is 50-100 %ectares according to the level of yield. The size of these holdings is typical for the indigenous owner-planter. The tea estates belonging to firms have larger processing capacities and larger plantations. In the highlands of Kenya the smallest size, according to McWilliam (1959), is a processing capacity of
229
25 000 kilograms of tea and 200 hectares. The optimum estate size is twice this. Besides the estates, we also find occasional tea-growing holdings whose harvest is sold to a neighbouring factory. In Kenya, Ceylon, and Java, these are run chiefly by small-scale producers. In Japan and Taiwan, the smallho!ders process their own tca by traditional hand methods. (a) Tea estates in Ceylon. Table 8.10 shows data of input and yield on a tea. estate in Ceylon, where most of the tea is more than 50 years old. A
TABLE 8.10
Coffee,
Kenya highlands Model 1968 162 50% 44%
100
Teab
Ceylon, high level Case study 1964-5 217 79% 12% 9%
178
14851
840 d
n.a.
914#
735 624
50% 14% 7% 7%
22%
576 518
24% 22% 13% 24%
11%
Overheads
111 49
735
58 60
576
735 417
417
323 370
208
*Estate average: 600 kg. Planning figures for a newly established estate. Net of interests. Sources: Coffee, Kenya Coffee Growers Association (1968); tea, estate accounts, personal enquiries.
230
special advantage from the angle of' farm management is the regularity of the labour operations. In Assam and East Africa, 80-90 per cent of the harvest of leaf occurs in 5-6 months, whereas in Ceylon it occurs in 7-8 months. Tea estates are generally characterized by the following farmmanagement features: (1) Tea is an intensive form of tropical land use. In the example in Table 8.10, the yield is 914 kilograms of made tea per hectare, i.e. $576 per hectare. Improved planting and plucking techniques, in conjunction with mineral fertilizing, have made a vital contribution to the increase of average yields in north-east India from 625 to 1075 kilograms in the last forty years. Fields with new clones, in a good rca climate, planted at high densities, and with high applications of mineral fertilizer (320 kilograms nitrogen per hectare), yield in Ceylon 30V(--4000 kilograms of made tea per hectare. In addition they are earlier yielding than traditional tea. (2) A tea plantation is permanent and the yield is reliable. A plantation produces regularly with only minor fluctuations from year to year. It is common to find tea plantations being used continuously for fifty years, and in Japan plantations are still cultivated that arc over 150 years old. Replanting becomes necessary when there are too many missing bushes, or when the productivity of the established bushes is low in comparison with that of improved clones. Conservation of soil fertility on tea estates is a problem similar to that of coffee cropping, in that it depends largely on husbandry techniques. Modern plantations with li.!le space between the plants, growing along the contours, showing a dense table of tea leaves, and with heavy fertilizing may well be regarded as a soil-preserving type of land use. (3) The main costs are those of labour. Practically 80 per cent of the costs of establishing a plantation without a.factory and 60 per cent of the costs of a mature plantation are wages and salaries. In Ceylon. we find a labour input of 5690 work-hours per hectare (see Table 8.11), 44 per cent of which is spent on harvesting the leaf (4000 shoots have to be picked for one kilogram of made tea). 19 per cent is spent on hoeing weeds and 8 per cent on work in the factory. The labour demand hovers around 2 man-equivalents per hectare of tea, and almost half is female labour employed in plucking. Tea plantations are very sensitive to wage levels, and success depends on the availability of plenty of cheap labour, since operations cannot be easily mechanized, mainly because of the inability of the plucking machines to discriminate between tea leaves for quality, and to avoid damaging the tea bushes. In addition, it is difficult to mechanize tea planting on sloping land. In Ceylon, for instance, the employment ratio for iabour-approximately 2.2 man-equivalents per hectare of tea-has changed little over the last forty years (Lir 1965, p. 18). Plucking shears,
231
with which a yield per day of 10-15 kilograms of green leaf can be increased to 100-150 kilograms, are used in Russia, Japan, and Taiwan, but only occasionally in Assam or Ceylon. They do reduce the quality of the leaf, however. Another feature of tea production is the rigid timing of operations, because the leaf has to be plucked at regular intervals and processed a few hours later. Thus, tea plantations can be disrupted much more easily by strikes than can sisal or rubber plantations.
TABLE 8.11
(4) High initial investments are necessary. The replacement of a plantation of one hectare in Ceylon costs from S1000-1500. In Kenya, McWilliam (1959) quotes from S2000-4000 per hectare for a new plantation and $87 500 for a tea factory serving 200 hectares. Thus a 200-hectare plantation costs at least S500 000, and, since there must be this high initial investment, there is a burden of high fixed costs. (b) Smallholder tea. Tea is a typical estate crop for various reasons: firstly
because the green leaf must be properly plucked and processed within a few
hours, and this is very difficult to organize with numerous smallholdings, and secondly because of the importance of refined techniques in plant raising, plant establishment, fertilizing, plucking, and in particular pruning, all of which have to be applied for optimum yields. Smallholder tea in Ceylon yields significantly less than estate tea. Nevertheless, smallholder tea has been introduced on a large scale in Kenya, and is also being introduced in various other countr;es. The key to the striking initial success in Kenya is the very thorough organization, the supervision of planting, husbandry, plucking (bud and two leaves only), and delivery of green leaf. Tea is grown in smallholdings that include numerous other activities like maize, potatoes, coffee, and milk production with grade cattle. There is, however, almost no interplanting, pure stands being cultivated according to closely prescribed rules adapted from the estate economy.
SYSTEMS WITH PERENNIAL CROPS Smallholder tea in Taiwan is quite diffei,.'t, because the production aim is a tea of low quality but high return per hectare (up to 4000 kilograms of dry tea). Harvesting with shears take.- place four tirr: 3a year. Mineral fertilizers are applied, together with mainly pig manure, and large areas of tea in Taiwan are intercropped with sweet potatoes and groundnuts. 8.6.
232
Estates and smallholdings with tree crops Crops like cocoa, rubber, oil-palms, and coconuts are tree crops proper. They differ from the shrub crops dealt with above in that they need more years to reach !he bearing age, their vegetation cycles are longer, and they present a less intensive form of land use. High costs for establishing plantations and for harvesting and low costs for plantation maintenance are the most obvious characteristics of this group. 8.6.1. Cocoa holdings Cocoa is a typical smallholder crop, mainly because cocoa can easily be established in combination with arable farming. It fits into various forms of mixed cropping and interplanting, husbandry demands are comparatively low, and processing can be done with simple equipment on the holding, Estate production is, however, practised in Brazil, Ceylon, and New Guinea. Jor instance, and is usually aimed at quality produce, which requires more refined processing techniques. (a) Cocoa estates in New Guinea. The initial stage of tree-crop development in the tropics is usually tied to the estate backward population. This is true for system, particularly in areas with a cocoa in New Guinea. Table 8.13 on p. 235 shows the farm-management data. Production is far less intensive than on tea estates. Plantation maintenance, harvest, and amortization are the main cost items. A special feature of the product ,'oa in New Guinea, where Australian personnel are employed, is a high salary for the manager. (b) Cocoa smallholdings of the Yoruba, Nigeria. More important than estate production is smallholder production, which predominates in West Africa. In the tribal area of the Yoruba, in the last few decades, more than 150 000 largely commercialized small-scale producers have established themselves. In these holdings we can distinguish sistence farming. A typical cocoa holdingclearly between cocoa cropping and sub0.5-1.5 hectares of land carrying arable consists of 1.0-1.5 hectares of cocoa and land belonging to the holding usually crops (see Table 8.12). The fallow and waste amounts to twice or four times the area the arable land. Stock consists of a few of goats, sheep, and chickens. The Yoruba's land-use system has evolved from shifting cultivation and has become partly
233
permanent and partly semi-permanent.' After clearance, yams, maize, and manioc are cropped for two to three years. During arable cropping, cocoa is sometimes sown and sometimes planted among the subsistence crops. Manioc, bananas, and papaya create shade at the beginning, and a plantation of cocoa develops, mixed with cola trees and oil-palms. Land planted in this way cannot be used for arable. cropping for a long time. The yields of cocoa that this method produces are very TAULE 8.12 Farm-management data of Nigeria cocoa holdings
Location Year Method Number of holdings ME per holding ME per 100 ha farm land Size of holding (ha) Total land claimed Arable crop land Tree land (mainly cocoa) Fallow land Livestock (mainly goats and poultry) (LU) Yields Cocoa (kg ha-1) Cocoa ($ ha-') Food crops ($ ha' 1) Economic return ($ per holding) Gross return Arabic farming Tree crops Livestock Expenses for purchased material inputs Wages for hired labour Family income Yoruba country 1951-2 Sample 70 3.3 48 6.9 1.3 0.8 4.8 0.3 533 223 36 324 20% n.a. n.a. 13
-,
Alade (Yoruba country) 1963-5 Sample More than 30 3.8 49 7-8 1.1 2.8 3.9 1.5 n.a. 156 130 525 27% 68% 5 0-1 62 463
311
Non-agricultural income
Productivity Gross return (S per ha total land) Gross return (S per ha cropped land) Gross return (S ME-') Value added ($ per ha total land) Value added ($ per ha cropped land) Value added ($ ME-') Hours of field work per ME
n.a.
47 154 98 45 148 94 n.a.
27
67 134 138 n.a. n.a. n.a. 921
Sources: Galetti, Baldwin, and Dina (1956); Upton (1967). ' In Ghana, the place where the most cocoa is produced by smallholderz, the conditions were originally different. The rapid expansion of cropping is due to the activity of small entrepreneurs settling in the area, who move on and buy new land as soon as cropping is no longer profitable.
SYSTEMS WITH PERENNIAL CROPS low at 533 kilograms per hectare (the average yield in Ghana is also low, with about 300 kilograms per hectare), since a reasonable yield expectation in modern cocoa plantations is 2000 kilograms per hectare. On the other hand, the costs of establishing the plantation and cultivating it are also low. 234 The arable branch of the holding continues alongside the tree crops with practically no modification. The same plot carries a variety of crops arranged more or less in a vertical order. The ground layer consists of creeping crops such as groundnuts, pumpkins, gourds and calabashes, melons, and creeping beans; the lower erect layer comprises manioc, maize, peppers, and erect beans; and the upper climbing layer is made up of yams and climbing beans. The extension of cocoa growing naturally leads to a reduction in fallowing. The ratio of fallow years to cultivation years decreases from 3:1 in less densely populated areas, to 1:1 in the middle of the cocoa region. At the same time, subsistence farming undergoes a change. Manioc is given priority, because it makes relatively low demands on soil fertility and produces more calories per hectare and per work-hour. Galetti, Baldwin, and Dinar (1956) calculated that in 1951-2 the gross return of a holding like this in the region of Ibadan was $323. Upton's more recent estimate (1963-5) amounts to $525, of which a good half is derived from cocoa (Upton 1967). Material inputs are low, and capital, apart from plants, consists only of a few hand implements and a rack for drying the cocoa beans. The problem is how to meet the labour demands that the arable branch of the holding presents. Cropping more than one hectare with subsistence crops usually fully occupies the labour force of a family of hoe farmers. The Yoruba farmers employ, therefore, a good deal of seasonal labour. 8.6.2. Rubber holdings Rubber is grown extensively by both large and small producers, although the objectives and the techniques of production differ. Production in estates is geared to the principle of maximizing financial returns, while smallholder rubber is part of an integrated farming system, with rubber serving as a supplementary source of income. (a) Rubber estates in Malaysia. Rubber estates are mainly situated in Malaysia, Sumatra, and Ceylon. Table 8.13 provides some information about the economics of estate production. The main farm-management characteristics are as follows: (I) Rubber production, although yielding much more than rain-fed arable farming under similar circumstances, is less intensive than, for instance, tea or coffee production. The established estate depicted in Table 8.13 yields 900 kilograms per hectare, valued at S212. With traditional varieties it is about seven years from planting to tapping, which is a significantly longer unproductive period than with the permanent crops dealt wi:h so far. (2) Soil fertility is well preserved in rubber plantations, which have the appearance of a forest in temperate climates.
235
(3) Manual labour is the main item of cost. Salaries and wages amount to 76 per cemt of the total costs (see Table 8.14). 3.3 hectares per man are reported itt our case study, whereas 1.5 hec'ares per man is the national average for estates in Ceylon, and the raio has remained almost unchanged
TABLE 8.13
Cocoa New Guitea 1958 Modela 400 90 750 (dry beans) 1000 69%
31 %
Rubber Malaya 1961 Case studies 40v 30 goo (. rubber) 10009 71%
29%
200 4400 30 25 1260 (copra) 2500 (oil) 600 (kernels) 7300 (nuts) 1580, n.a. rT.a. n.a. n.a.
n.a.
Economic retlrn (Sper ha planted) Gross return" Costs, Upkeep of plantation Harvesting, transport) Processing" Equipmentk Depreciation Overheads, inc!uding management Net return ($ha-)
Productivit)y 165 467 212 268 Gross return ($ ha.-) 550 1868 707 298 Gross return (S ME-') Prepared model for estate investment, based on estate records. Project plan. All data refer to mature stands only. Including factory labour. Including factory for processing. Estimate of investment requirements. Book value, below market value. h At estate, riot including transport to harbour,net of export tax. Net of interests. Including fertilizers and pesticides. " Running costs only. Sources: Cocoa,Wood (1963); rubber,Westgarth and Narayana (1964), oil-palm, E.W.G.S.O.D.E.P.A.L.M. (private comm unication) (1968): coconut-palm, Ceylon Department of Census and Statistics (1965, pp. 67-70).
236
for decades. Mechapizaton is still in the experimental stage, and taipping will always remain ianual work. There is no pronuunced peak in labour demand. Tapping, the mcst important cost item, is carried out regularly all through the year, with fixed assignments for each worker. This makes it possible to work with an all-the-year-round division of labour, one group being engaged in tapping, another in weediig, and so on.
TABLE
8.14
0.48 Costs ($) 25% General charges Salaries Medical care Insuranca., tax. contribution, etc. Upkeep of buildings Contingencies 57 ' Plantation Wages for weeding, pruning, and harvesting Wages for terracing, transport, etc. Materials for fertilizing, pest control, fencing Other materials 18% Processing Factory labour Machinery upkeep Fuel, packing materials, sundries 100% Total 66 Salaries and wages as %of total 290 Costs ($ ha-I)c
'
I%
Estate accounts in Ceylon do not include depreciation of mature stands, buildings, and the factory. replacement costs are used instead. Replacements in the year 1963 were probably below an appropriate depreciation charge. Costs depicted above are therefore likely to underestimate capital costs. I Ceylon rupia = $0.1565. Yield level: tea 600 kilograms per hectare; rubber 600 kilograms per hectare.
(4) High costs in plantation establishment in relation to the returns per hectare make depreciation one of the most important cost items. Thus
rubber production requires a lot of capital per unit of output compared with other perennial crops.
(5) One of the important advantages of estate rubber is the ease of introduction of innovations. New tapping techniques and new clones radically chanred the economic prospects of rubber production, which had looked
237
New clones yield-under the conditions rather bleak due to falling prices. 0 -1 60 0 kilograms of latex per hectare per year. of estate production-420 production Economies of scale are very pronounced with increased rubber higher mean yields per hectare (see Fig. 8.6). With tea the increased
0*S0
X
0.70 KC
,5 0.0
x
0.50
0 '4C
I I
1
0-40-
_ ._ I
I.
500
750
1000
1250
1500
1750
pound in seven rubber estates in FIG. 8.6. Relation between yield per acre and costs per 1969). Hansen (from Ceylon
remain almost plucking costs. With rubber, however, costs of tapping place earlier. take can planting constant irrespective of the yield. Moreover, coild begin, tapping Plantations traditionally required seven years before years. 5. after start but with the new planting material tapping may or coffee bush, which (b) Rubber snallholdings. Compared with the tea in order to keep regularly plucked has to be manured, weeded, pruned, and undemanding. remarkably is tree it in proper productive condition, the rubber can be smoking and Rolling Processing is not tied to the producing farm. attracted therefore Rubber assigned to neighbours or larger processing units. to enter into comsmallholders in Ceylon, Sumatra, Malaysia, and Nigeria mercial tree-crop production. with trees grown Rubber estates generally present a picture of orderliness, cutting. Traditional in neat rows and the undergrowth kept down by constant
238
smallholdings usually have a different appearance. The situation in Malaysia is depicted by Courtenay (1965, p. 124) as follows: the valley floors are occupied by rice fields, irrigated by channels led along ... the hill foot from the central stream. Abuve ... these canals run the main footof paths, just beyond which are located the Malay houses under the shade are hillsides the houses, the above and coconut-palms and fruit trees. Behind planted with rubber, which may often be tapped by the rice farmer's wife or children. Smallholder rubber is often interplanted with fruit, coconuts and arable crops. The trees do not always form continuous stands but may be in clumps separated by other vegetation. Beyond the environs of the smallholder's house, the mixed stand of rubber and other crops usually gives way to a pure stand of rubber in the midst of tall undergrowth. Smallholder rubber plots in Malaysia average between I and 3 hectares. In contrast with estate planting (300 trees per hectare), final-stand planting on smallholdings is usually very dense, with 500-900 trees per hectare, the reason being that smallholders aim at high yields per hectare, disregarding higher labour input, while estates try to economize on labour by aiming at high yields per tree. Rubber tapping is not necessarily a regular task, and this makes rubber such an attractive crop for smallholders, since during the peak rice-planting and rice-harvesting seasons little labour need be devoted is to rubber. Smallholders prefer to tap when time permits or when cash cases, some In returns. average needed, although irregular tapping may lower smallholders tap only during periods when schools are closed-weekends and vacations. Another advantage of rubber for smallholders lies in the possible employment of female and child labour at low costs. Smallholders use unselected seedlings usually. They do not receive maximum yields because of poor cleaning, lack of manuring, irregular tapping, overtapping where cash is needed, and often wasteful tapping techniques. They are, however, low-cost producers, and as such competitivc with estates. Whereas estates are severely hit by slumping prices, smallholders may cultivate other crops more intensively, and in the meantime the rubber trees may rest for 'rejuvenation', to produce larger yields when rubber prices rise again.
work. I The family labour is mainly employed in off-farm b 159 coconut-palms, 70 various fruit trees, 100 pepper vines, 200 plantains.
c Vegetable garden.
dlntercropped under coconut-palms. The price per kilogram of rubbei is lower than in estates, because rolling and smoking Including return from various fruit trees, pepper, plantains, etc. Including minor chzrges for tax. s Including bullock laboar. Sources: Dahomey, S.E.R.E.S.A. (1959-60); Ceylon, personal enquiries.
'
TABLE
8.15
Rubbercoconut-rice Ceylon
Kaduwela 1969 2.0 178 1.18
-
Oil-palmmaize Dahomey
Cercle d'Allada 1960-1
Rubbercoconut-rice Ceylon
Kaduwela 1969 0.51 2.2 112 2.42
-
Coconut-rice Ceylon
Candy 1969 0.51 2.2 123 2.2
-
ME per holding
3 Family labour n.a. Hired labour 19 ME per 100 ha farm land Size of holding (ha) 15.80 Total land
Oil-palms
220
Rubber, mature Rubber, immature Coconut Other perennial crops Total perennial crops Rain-fed arable crops Irrigated rice Livestock (no.) Goats, sheep Poultry
--
0.41 0.81
0.80
2.02 0.40 4
Yields oil-palms, fruit bunches 10000 (kg ha- t) Rubber sheets (kg ha- 1) Coconut (nuts per ha) 1000 Maize (kg ha- ') Rice, paddy (kg ha-1) 100 Oil-palms ($ ha-') Rubber ( ha- ') Coconuts (S ha- ') 32 1) Maize ($ ha- 1 Rice($ ha ) Economic return (s per holding) Gross return Tree crops Arable crops Expenses for purchased
78 99 n.a. n.a.
118 69 108 63
240
Table 8.15 illustrates the farm-economic situation cf two rubber-growing smallholdings near Colombo, Ceylon. Rubber is the major cash crop and the plantations are considered as a capital asset. In both cases, new plantings with improved plants have replaced the traditional mixed,stands. Coconuts are primarily consumed in the household or sold locally. Rice provides the food for the household. 8.6.3. Oil-palm holdings Until recently, most of the oil-palm growing took place in West African smallholdings and was part of subsistence-food production. However, a genetic breakthrough has made oil-palms an economically worthwhile estate crop. With oil-palms, the economic return per hectare during the vegetation cycle depends on several variables, almost all of which have been influenced by technical innovations. (i) New improved varieties are available that mature after three to four years, instead of ten years as with old varieties. New varieties are easier to harvest, because the height of the stem is only a quarter of that of unimproved palms. (2) New varieties yield heavier bunches, wit' heavier fruits. Two types of oil are produced from the fruit: palm oil from the pericarp and palmkernel oil from the kernel. New varieties have much more fatty pericarp. Traditional production yields about I ton of oil per hectare of mature palms. New varieties yield 3-5 tons and more. (3) Oil milling and the transport of fruit bunches to the mill have benefited from innovations. We consequently still find extensive areas where traditional oil-palm production takes place, but increasingly both estates and smallholdings are planting new varieties. (a) Oil-palm estates in West Africa. Oil-palm estates have mainly been established in Malaysia and the Congo (Kinshasa). More recently Ivory Coast, Dahomey, and Cameroon have eatered large-scale production through para-state bodies. Modern oil-palm cultivation lends itself to estate production. New varieties, and the corresponding husbandry techniques, can be applied ad hoc on large areas. The perennial nature of the oil-palm, its significant response to weed and bush control, its constant employment of the labour force on continuous harvesting, and the need to move the fruit to the mill as quickly as possible after harvesting in order to avoid loss in quality are factors that favour large-scale p:oduction. The estates are generally very large, varying between 500 and 2000 hectares in Malaysia and 1000 and 4000 hectares in West Africa. The farm-management characteristics of oil-palm estates may be summarized as follows (see Table 8.13): (1) The initial capital investment is higher than with most other tree crops and is at about the same level as for tea and coffee. Establishment
SYSTEMS WITH PERENNIAL CROPS 241 of the oil mill is the most important capital item; amortization charges are correspondingly high. (2) Gross returns are higher than with most other tree crops, but are lower than with shrub crops like coffee or tea. However, the natural risks of production (disease or other natural calamities) are low. (3) Oil-palm plantations offer an almost complete cover for the soil. They may well be rega ,ded as a soil-conserving type of land use. (4) The labour input in running an estate is perhaps even smaller than with other tree crops. Harvesting is a regular affair that is helped by truck transport. (5) Large estates organized by state firms, as is the case with oil-palm estates in West Africa, are usually burdened by high overheads and management costs. (b) Oil-palm smallholdings in Dahomey. Most West African oil-palm production is still based on unimproved, 'wild palm groves'. According to the analysis by Zeven (1967), there are three main types of oil-palm groves: secondary rainforest containing oil-palms, open palm bush with plant densities from a few palms per hectare to more than a hundred, and dense groves of palms. In addition there are here and there small and medium-sized plantations established by small entrepreneurs. Smallholdings in the Cercle d'Allada, on the coast of Dahomey, where the rainfall is marginal for oil-palms, cover rather large areas, averaging 15.8 hectares covered more or less irregularly by oil-palms with plant densities varying from 20 to 200 palms per hectare (see Table 8.15). Harvesting, and transport of the bunches to the collecting points is the only work necessary with oil-palms. Labour-demanding processing by small hand-presses has already been replaced by proper oil milling. The return per work-day is about $0.50, five times higher than with maize, and this in spite of low-yielding varieties. In addition to oil-palms, small neglected patches with robusta coff,,e are to be found close to the villages. Most of !he working time is absorbed by growing maize and manioc for household needs. Two to five years of cropping are followed by five to seven years of fallow. The smaller the area in relation to the number of household members, the higher the food demand, and so the fallows are shorter and more manioc is grown instead of maize. In 1959-60, the average gross return of these holdings was estimated at $297, most of which came from oil-palms. The labour input of 945 hours of field work per man-equivalent yielded a return of $99 per man-iquivalent. The problem of smallholder production is one of reorganization of farming, to take advantage of the genetic breakthrough, which is already exploited by the estates. An important effort to solve the problem has been made in Ivory Coast. Nucleus estates are increasingly surrounded by smallholders, up to 20 kilometres away, who practise modern techniques of oil-palm production with the help of credit schemes and close technical supervision. The fruits are collected and processed by the mill of the estate. Initial experience shows
17
242
that plantation establishment in smallholdings is much cheaper than in the estates, but hushandry is not usually as good. 8.6.4. Coconut-palm holdings Most commercial coconut production takes place in the numerous smallholdings of the Philippines, Malaysia, Indonesia, Ceylon, and the East African coast. Intermingled with the smallholdings, there are coconut estates. (a) Coconut estates in Ceylon. Coconut estates in Ceylon vary in size from a few hectares to several hundred hectares. The characteristics of production are largely the same irrespective of size. Plant density is about 160 palms per hectare. lntercropping is rare. Some plantations are grazed by local cattle, but most managers prefer to farm without cattle. In the case study depicted in Table 8.14, each palm yields an average of forty-six nuts a year, which makes 7300 nuts per hectare. About 5900 nuts are required to produce one ton of copra. The nuts are picked five or six times a year, and the husk is separated from the nut in the field. The husks are used as mulching material. The kernel and the nutshell are separated at the farm. The flesh is dried in a copra kiln, the shells being used as fuel, after which the ash is scattered on the plantation. Besides wages, the fertilizer input is the main item of expenditure, amounting to 28 per cent of the total cost. Palms with a low yield, of less than thirty nuts a year, are interplanted with new palms, and chopped down as soon as the new palms begin to produce. There is a high degree of specialization. Permanent labourers are employed for weeding, transporting, and processing. Harvesting is done by outside task workers. We may summarize the essential characteristics of coconut-palm estates as follows: (1) The intensity of land use in coconut-palm estates is lower than with other tree crops. In the case study in Table 8.13, gross returns amount to $165 per hectare. (2) In addition, we have to take into account that the lapse of time between planting and the first and the full crop is particularly long. Traditional varieties take eight to ten years before the first crop is picked. The full yield can only be expected after fifteen years. The economic cropping time then may well extend to over fifty years. (3) Cultivation requires comparatively few workers. The average figure for Ceylon is 20 man-equivalents per 100 hectares of coconut-palms, which is lower than is noted in the case study (Lim 1965, p. 121). Labour costs are nevertheless the most important item of cost, because few other inputs are required to run a coconut-palm estate. There are no peaks in labour demand. The main work, harvesting and preparing the nuts, occupies about 60 per cent of the total labour and is spread over the whole year.
243
When estates establish plantations, they have to reckon with high costs, amounting, according to Piggot (1964), to $750--2000 per hectare, largely because of the long lapse of time between p'antii , and the first harvest when the plantation has to be maintained without yield. The capital input is consequently high in relation to output and employment in mature plantations, but the day-to-day running of the estate requires little capital. The replacement of old palms is, compared with the costs of new plantations, relatively inexpensive, because young palms can be intcrplanted in old stands to replace them gradually and without a break in production. Even the processing facilities, copra kilns, do not require a high financial input. The size of kiln may vary according to estate production, so that estate size does not depend on the capacity of the processing facilities. (4) Estates usually carry pure stands, the soil being kept free from bush by slashing or cultivation. Coconut plantations in high-rainfall areas, however, lend themselves to mixed cropping or intergrazing, and this need not reduce nut yields provided the interplanted crops do not compete for water and nutrients. Thus interculture has to be restricted to seasons with high rainfall, and the crops have to be fertilized. Cashew, mangoes, bananas, pineapples, and various arable crops, like manioc or rice, may be found intermingled with coconut-palms. Grazing under coconut-palms is extensively practised in Ceylon and in the south Pacific. Fox and Cumberland (1962) report from an estate in Samoa that Mimosa pudica as a fod ler crop under palms may carry 2. livestock units per hectare. (5) Considerable technical advances have been achieved in growing coconut-palms, as in the case of most other tree crops. New varieties, which yield after as little as six years, and which produce 2000-3000 kilograms of copra per hectare of mature palms, are now available. Fremond wrote in 1968 about Ivory Coast: 'We know that yields of definitely even more than 3000 kilograms of copra per hectare, with nut-producing after 4 years, perhaps even less, are henceforth possible.' (b) Coconut-palm smallholdings. Coconut-palms lend themselves to smallscale production and are mainly grown in smallholdings. They are suited to interculture, almost all the operations in coconut growing are performed by manual labour, and there is no problem of overcropping as with rubber. The uses of the coconut are many and diverse, including copra, fresh nuts, fibre, fuel, leaves for roofing, and sap for beer. The dependence on large-scale processing facilities is not very pronounced. The main advantage of smallholder production arises from interculture in the early stages of plant development, which makes it much cheaper to establish stands of coconut-palms than in estates, where the young stand must be weeded for eight to ten years before harvesting begins. Table 8.15 shows the farm-management data of three smallholdings with
244
coconut production. In all three cases, coconuts are not a major cash crop, but are primarily grown for household consumption, while rubber and rice are mainly grown for cash. Another striking feature is the preference for non-
farm work. In two cases, family members are employed outside the farm, and hired labour is employed for the farm work. All three major crops grown in the holding--coconut-palms, rubber, and rice-lend themselves to task work, which can be carried out without much supervision by the farmer.
crops
The natural production conditions of the humid tropics are particularly good for the cultivation of perennial crops. Consequently, problems in production are not primarily due to the natural conditions and the inherent characteristics of the various crops, as is the case with rain-fed arable farming, but to the socio-economic organization of the production process. Both major types of exploitation, smallholdings as well as estates, show certain weaknesses, which may be summarized as follows: 8.7.1. Weaknesses of srnallholderfarming with perennial crops Perennial crops require a high input of labour and relatively few material inputs, which would indicate that they could be grown more economically in smallholdings than in estates, where labour is more expensive. Experience shows further that tropical smallholders are usually very skilful in choosing optimum crops, crop sites, and crop mixtures. They are, however, inherently weak in husbandry in comparison with the performance of estates. Land, labour, and capital would frequently bring much higher returns if perennial crops were properly taken care of. Lack of proper husbandry is often accompanied by mixed cropping and interculture, although it is certainly not a necessary feature of these practices. Farmers on Taiwan show convincingly that very high levels of husbandry can be attained with interculture. However, few smallholders in the tropics cultivate their perennial crops at similar high standards. As a rule they can hardly be induced to cut down old fruit trees. Weeding, bush clearing, and mulching are neglected in many cases. Pruning, which is not heavy work and offers much latitude in timing, is carried out negligently if at all, and this in spite of the fact that seasonal underemployment prevails. The list of short-comings in husbandry could easily be expanded, but it is much more difficult to explain why most of the smallholders are poor in husbandry. There is certainly no single reason, but rather a host of possible explanations. Those most frequently put forward by the farmers themselves include the following: (1) The production of seedlings is usually a specialized task that can hardly be carried out efficiently in smallholdin3. Optimum sniallholder
245
development consequently depends on specialized nursery producers supplying the seedlings and the necessary loans. (2) Smallholders almost invariably say that husbandry falls short, even of their own standards, because of peak demands in labour and lack of the necessary cash for equipment. (3) Some difficulties are due to the length of the vegetation cycle of perennial crops. A smallholder usually inherits or acquires a mixed stand of trees and shrubs of varying age and not planted in rows. Rcplanting in rows in one part of the plot after the other is difficult to organize. Infilling is a much more convenient way, but perpetuates a shrub and tree garden where yield-increasing innovations are difficult to apply. (4) Other losses in yield are due to the relatively short time-horizon of smallholders. Practices that reduce the total yield of the perennial crops but bring about early returns, like heavy interculture in young plantations and early tapping of rubber, are often preferred, because smallholders more or less corsciously reckon with higher discount rates than do estate owners. (5) In many cases, husbandry is poor because the returns from an additional hour of work are not considered worth the additional effort. Most smallholders are still inhibited by the idea that their soil is not good, that the climate is adverse, that improved practices would not work, and that, consequently, increased efforts are not worth while. Because of this attitude, husbandry is generally poor, and poor husbandry results in low returns per hour of work, which again leads to even poorer husbandry. Although these and many other arguments are valid, it is nevertheless clear that they do not afford a complete explanation of the lack of proper husbandry. If labour were indeed the main limiting factor, it should express itself in high marginal returns and the existence of a significant relationship between the labour available and the gross returns. This is not usually the case, however. The shortage-of-labour argument often offers no rational explanation, because a small but properly cultivated area probably yields higher returns per hour of work than larger, neglected fields. The small but significant number of better farmers who work more intelligently shows that with less effort per unit of output, but with greater care, more money can be earned. Moreover, those smallholders who are not willing to take on additional work can usually employ seasonal labour for fairly low wages. It would seem that the main reason for poor husbandry is not any objective factor, but primarily a lag in cultural adaptation. Smallholders in many parts of the tropics, in particular in Africa, experienced within a few decades the change from shifting ciltivation with temporary homes to sedentary farming with perennial crops. Shifting cultivators know how to adjust to changing
246
natural conditions, and so in this respect smallholders are very knowledgeable. But they are less well prepared for the tasks of transforming natural conditions by conscious human effort and of improving yields by the proper maintenance of the plants-something that farmers with centuries of land shortage or decades of cash cropping behind them have learned how to handle. The quality of the husbandry found in especially worthwhile cash activities, and in areas that have had a land shortage for a very long time, indicates the starting points for better husbandry: (i) there must be a very strong incentive in the form of a high cash return per hour of work; or (2) there must be a serious, long-term land shortage, which leaves no choice but to take care of plants, soils, and animals; or (3) there must be production under close supervision. The fact that most plantations are poorly husbanded may be considered cn indication that, as a rule, neither the pressure nor the incentive has passed a certain threshold value required to get the smallholders on the way to proper husbandry. rhe fact that peasant production can still compete with estates, despite obviously poor cultivation and husbandry techniques, is explained by the cheapness of production and the lower susceptibility to risks. Investment in an estate consists of the cost of: (1) the factory; (2) the establishment of the plantations; (3) workers' houses, a hospital, roads, etc.; and (4) bridging the initial period when there is no yield. The advantage of peasant production is that the financial outlay is low apart from building the factory, or can be raised by the farmers themselves. Smallholdings have no production overheads. The recurrent cost schedule is different, since up to a certain level most of' the labour that is employed is either unpaid fam."y labour or receives far less than the rates laid down by unions that apply tc estates. Instead of having to pay back interest, a family is prepared to forg, the full income from the land until permanent crops produce yields. But it is not just a question of different demands on cash, or of the availability of labour that can scarcely be employed anywhere else; the whole situation is different as far as costs are concerned. A large-scale estate must bear the costs of clearance work, whilst in the family holding the land is in any case cleared to grow subsistence crops. It costs little extra to plant seeds or to set plants of a future perennial crop in the cleared land. The mixture of perennial crops with arable crops in the early years bridges the period when there is no harvest. A plantation that competes with arable crops grown on the same land certainly does not produce the same yield as
247
a pure crop, but it is comparatively cheap and simple to set up.' These combined benefits give the smallholder a distinct cost advantage. 8.7.2. Weaknesses of the estate economy The farm-management advantages of growing perennial crops on a large scale do not lie chiefly in the labour economy. The use of labour-saving equipment and transport is important only with perennial field crops like sisal or sugar-cane. The competitiveness of estates growing shrub or tree crops with the farmer's cheap production methods is based mainly on: (1) the rapid and consistent use of technical advances in plant production; (2) the more efficient organization of delivery of the crop to the processing factory; (3) the more efficient processing of the product; and (4) the better access to markets and capital. Most managers of large-scale holdings, particularly the employees of large plantation companies, take advantage of modern technology. They harvest larger quantities per hectare and the yield is of considerably better quality. The processing of the crop is better, and results in a quality product that can be sold at higher prices. Moreover, by-products and residues are used more efficiently. All told, a return or income per hectare and per worker is obtained that is often higher than that of a smallholder. Estates, therefore, produce a high, taxable income. In addition, the proportion of the income charged as tax is much higher than in the case of smallholders. On the other hand, there are a number of disadvantages: (1) The estate economy is exposed to high risks and uncertainties. Estates have to meet high interest charges on borrowed capital. The production is usually devoted to only one crop, and a shift in emphasis from one crop to another is a long-term business. The return from estates fluctuates widely because of natural hazards, for example wind damage with bananas, and because of changing prices on the world market. In the developing countries, the private-estate economy is threatened by nationalization, which acts as a barrier to investment. Semi-commercialized smallholdings are much better protected from risk: subsistence is guaranteed by cropping plants for immediate consumption, the proportion of the overall costs paid in wages is low, and there is normally no political danger. (2) Estates with perennial crops depend on an ample supply of cheap labour. The fact that the tropical countries can supply plenty of cheap labour is one of the reasons why large-scale estates with perennial crops I Thus in Kenya the cost of establishing a large-scale coffee estate before the first yield is $750 per hectare, whereas the research of MacArthur (1966) in neighbouring smallholdings shows a cost of $437 per hectare (including the wages for workers belonging to the family).
248
SYSTEMS WITH PERENNIAL CROPS products were established there, and explains their power to compete with Rising from temperate climates or products from the chemical industries. wages, which are spreading with the organization of unions, hit large-scale be mechestates with perennial crops very hard, because operations can anized only with great difficulty or at unfavourable rates of substitution. workers, Smallholdings operate with family labour and local seasonal those of than lower disproportionately who are prepared to accept wages estate workers.
rise, As risks in the marketing and political sectors increase, and as wages smallholdings are becoming more competitive than large-scale producers, are more particularly with perennial crops rather than arable crops, which stategeneral this Naturally mechanization. amenable to motorization and crops perennial various The cases. ment needs to be qualified in individual largefor suitability their in differ and advances in agricultural technology scale or small-scale production. As a rule of thumb, estate production is more competitive than smallholder production: this is (1) When the production process is technically demanding. Where likely is smaliholdings and estates the case, the difference in yield between change cultural where case the to be particularly great, and this is especially among the smallholders is slow. cjsts. (2) If the produce of the land is bulky, requiring high transport holding, the on Here, the processing factory should preferably be located with the fields as close as possible. processing (3) Where the produce is perishable and has to be delivered for or sale in large quantities of even quality at definite times.
8.8.
There is a great deal of scope almost everywhere in the tropics for increase in the plantation area, in particular to replace shifting and semi-permanent and cultivation. Increases in area depend only partly on capital, labour, and smallholdings technical infornation. Plantation establishment-both in However, in estates-is largely a function of prices and price expectations. are the perhaps even more important than the scope for extension in area connumerous possibilities of increasing output per hectare. The following important. siderations are among the most 8.8.1. Increased application of technical innovations field Organized agricultural research in the tropics began with perennial was and crops, shrub and crops like sugar-cane, progressed to arable crops
SYSTEMS WITH PERENNIAL CROPS 249 concerned only at a relatively late stage with tree crops. Growers are now in a position to reap the benefits of a great number of breakthroughs, particularly with tree crops. With traditional varieties, 30 per cent or less of the shrubs or trees produce 70 per cent and more of the crop. The development of high-yielding varieties and clones has significantly changed the picture. With various tree crops a doubling of yields is within reach. In addition, numerous other agronomic innovations are in the process of being introduced. They are summarized by Webster and Wilson (1967, p. 217) as follows: (a) the use of improved planting material, selected to give high yields of good quality produce, but possessing good secondary characteristics; (b) cheap and practicable methods of propagating the improved planting material on a large scale; (c) efficient transplanting and early care of the young trees, in order to obtain a full stand of plants that grow vigorously and reach the productive stage as quickly as possible; (d) good horticultural practice in respect of such matters, as spacing, shade management, windbreaks, pruning, etc.; (e) methods of soil management, including soil conservation, cover crops, weed control, nulching and manuring, which will maintain soil fertility and sustain high yields; (f) control of pests and diseases; and (g) improved methods for the exploitation of the crop, such as modern techrques of tapping and yield-stimulation in rubber, or the use of synthetic growth substances to induce early and uniform fruiting in pineapples. The relevant economic aspect of these innovations is that rising wages can be absorbed by higher yields per hectare. In many cases the quantity of product required for the processing unit may in future be produced more economically on a fraction of the original land. The remaining land then becomes available for other uses. 8.8.2. From mixed cropping to pure stands and interculture Yield-increasing innovations are best applied in pure stands, where they may be adapted to the special requirements of the crop. Mixed cropping is a handicap to mechanization alzo. We may assume, therefore, that mixed cropping is likely to decline. However, where land is scarce but where labour, capital, and technical knowledge are relatively plentiful, we can assume that various forms of interculture will be extended and intensified. This trend is seen clearly in Taiwan for example. 8.8.3. From holdings with one cash crop towards diversified commercial production The tendency towards pure stands in order to apply effectively innovations specifically suited to one plant need not necessarily result in monoculture. This may be true of some perennial crops, but it is certainly not true of all.
250
In high-potential areas with high population densities, it is much more likely that production development will lead to even more diversified holdings. The introduction of intensive animal husbandry is of particular importance in this connection. The development of Kikuyu agriculture, as depicted in Table 1.16 , are a case in point. Originally, shtifting cultivation with millet, maize. and ovans prevailed. Increasing population pressure led to semi-permanent and pirraanent farming, reduced available grazing, and resulted in a process of involution. The. efforts of the Swynnerton Plan brought a sweeping change in the farming systero.s. Coffee. and other cash crops were introduced. Commercial milk production with grade cattle began, and some leys were established. Tea production, hybrid maize, potatoes, and the application of mineral fertilizer were already spreading in the mid-sixties. It might well be expected that the impetus in smallholder development, which begaa with coffee, would facilitate the introduction of further innovations such as fodder growing and zero-grazing. TABLE 8.16 Year Evolution , f the farming system in the Kikupu districts of Kenya Land rights Livestock Cropping Type of economy pattern farming Shifting cultivation Maize, beans, mixed cropping Ample grazing, Zebu cattle, goats Limited grazing, Zebu cattle, goats Ample land, rights of land .i
Limited land, rights of land use, communal grazing Rights of land use, Roadside Maize, beans, Permanent About 1950 sweet potatoes, grazing, mainly most communal cultivation grazing turned banana, wattle goats, some into individually Zebu cattle cropped plots Private property Roadside Permanent crops Coffee, maize, About 1960 grazing + some rights; land can be beans, sweet + permanent leased and grade cattle potatoes, cultivation + mortgaged banana, potatoes some ley vegetables, farming pineapple, lays Private property Gradc cattle, Permanent crops Coffee, tea, Estimated rights; land can mostly zerobeans, hybrid expansion path + permanent grazing, poultry be leased and maize, sweet cultivation + for 1980 mortgaged; an pigs some sprinkler potatoes, active land potatoes, fodder irrigation market grass, application of mineral fertilizer and stable manure Semi-permanent Maize, beans. sweet potatoes, cultivation mixed cropping This table is nothing more than a rough guide to the developments which took place and which are expected in the Kikuyu areas. See Rntthenberg (1966, p. 217).
251
food andfruit crops 8.8.4. Increasing importance ofperennial Bananas are one of the staple foods of tropical cultivators. With increasing population densities and with increased cropping on sloping, easily eroded land, it might well be expected that bananas, in particular cooking bananas, will gain in area and economic importance. Industrialization and urbanization might also specifically favour fruit trees. Firstly, there is tle influence of the market, demanding citrus, mangoes, papayas, bananas, etc. Secondly, there is the tendency for people living in mixed rural-suburban communities to plant fruit trees, coconut-palms, and bananas in their extended gardens. Tree plantations are particularly favoured by part-time farmers. 8.8.5. Integration of estate andsmallholderproduction Of particular interest are the tendencies to combine in 'nucleus-estate' projc'ts the advantages of estate production-the rapid application of innovations in production and processing-with the advantages of smallholder production--the low labour costs in plantation development and the smallholder's preference for investment in tree or shrub crops. Schemes of this nature show a promising start with oil-palms (Malaysia and Ivory Coast) and tea (Kenya). The idea is that commercial estates are surrounded by a ring of smallholders, who obtain credit, advice, material inputs, machinery, and markets through the scheme management and who are subject to close supervision in their husbandry practices. Nucleus estates, combined with supervized smallholder production, seem to be a major alternative to the dualistic plantation-smallholder economy that still prevails in extensive parts of the tropics. It is likely to be more popular because: (1) capital costs are lower than with state-financed estate development; (2) production is less sensitive to rising wages and labour unrest; and (3) projects of this nature are in line with the political philosophy of numerous countries.
9. Grazing systems
9.1. Definition and geographical distribu'.on of grazing systems
in the tropics as the main activity of any enterprise or household is more or less confined to various types of nomadism and to ranching.
STOCK-KEEPING
(1) In total nomadism, the animal owners do not have a permanent place of residence. They do not practise regular cultivation, and their families move with their herds. (2) In semi-nomadism, the stock owners have a permanent place of residence, which is kept for several years. In the vicinity of the residence, supplementary cultivation is carried out. Semi-nomads usually travel for long periods of time with their herds to distant grazing armas. The husbandry practices of partial nomads, i.e. of farmers in permanent settlements, who live mainly by arable farming but who own cattle grazing on the communal land, are similar (see 4.3.4). (3) In ran:hirg, stationary animal husbandry organized according to commerciL! principles is practised on large stretches of land. In all these types of land use, the stock-keeper has to adapt his stock to the fodder produced by the natural vegetation. Climate, soil, vegetation, and animals form an integrated system, which the stockman, if he wishes to operate profitably, can influence only indirectly through such matters as the way he keeps his animals, the composition of the herds, animal density, times of grazing, and other considerations. Such considerations are especially important in dry regions, where the rainfall is not sufficient for arable cropping or tree crops, for example in the marginal zones of the Sahara and the semi-arid regions of Australia and East or southern Africa. Nomadism and ranching in the humid savannas, on land that is -ilso suitable for arable farming, are practised principally in areas with low population densities, as in Madagascar, for example, or in various parts of Latin America. Commercial ranches are almost always very large. The establishment of extensive holdings with permanent stock-keeping was considerably helped by technical advances in fencing, water supply, and vehicle transport, and holdings of this kind are found principally in America, Australia, and various parts of Africa. The common aim of ranchers is the conversion of naturally
252
GRAZING SYSTEMS
253
is achieved by grown herbage into marketable produce, and this objective In subsistence livestock. of herds employing small numbers of people for large by numerous kept are livestock of herding, on the other hand, large numbers in grasses, stored energy the households, and the aim is above all to transform food. of supply herbs, and shrubs in the tribal area into a regular of using grassRainfall is the primary factor determining indigenous ways rainfall averagland (see Table 9.1). Where there is less than 50 millimetres
TABLE 9.1
years, we find only per year, which means in effect no rain at all in ma9n and the dry occasional visits by nomadic herds, but in the semi-desert is usually systemsteppes, with between 50 and 200 millimetres of rain, there example, in the for found, is as such atic nomadic grazing with camels, 200-1000 milliof rainfall a is Sahara and northern Kenya. Where there economic greater much of are metres, semi-nomadism and partial nomadism savannas. African the in importance, and we find this kind of economy chiefly economic better a is Where there is a higher rainfall, permanent stock-keeping milli1000 than proposition, and semi-nomadism where the rainfall is more of form practical metres is a matter of tribal tradition rather than the only land utilization that conditions will allow. according to place Animal densities in grazing economies fluctuate greatly 50 hectares and type of farming. In areas with a rainfall of 50-100 millimetres, cow or one of equivalent (the unit and more are needed for one livestock is often seven sheep). In regions with 200-400 millimetres, the requirement are hectares 6-12 millimetres 10-15 hectares, and where there are 400-600 and distributed, evenly more necessary. As the rainfall increases and is Grass grows farming becomes more intensive, the animal densities increase. rainfall of high a is very extensively in tropical grazing areas where there an even with 1000-2000 millimetres a year. Where high rainfall coincides
254
GRAZING SYSTEMS distzibution of rain, such as occurs in some areas of Central and South America, animal densities of up to one livestock unit per hectare are possible. 9.2. Total nomadism 9.2.1. General characteristics of total nomadism Total nomadism is usually carried on under decidedly marginal production by conditions, or where it is not worth trying to overcome natural difficulties applying cultivation techniques. The basic principle lies in as much adaptation as possible to natural conditions. From the point of view of agricultural evolution, nomadic stock-keeping belongs to the same primitive category as shifting cultivation, but this l-y no means prevents the type of adaptation being an extremely complex and carefully calculated system. are The nomad's animals have to keep moving because water and fodder consequence A place. one not sufficient for them to be kept permanently in of the animals' movement is the movement of the household, since the people live directly off animal products, particularly milk, and they are thus obliged to stay near the herd. This permanent migration precludes regular arable it farming. The nomads' production is not tied to one particular place, since comprises only animals and the implements necessary for animal-keeping. the Thu pastures grazed by the animals are either freely available to all them. rent to have nomads the or group, members of the tribe or the The nomads who practise horizontal migration, as in the Sahara or Arabia, chiefly keep large animals (camels) and trek with their animals to where rain falls and fodder grows. On the other hand, there are nomads who practise verticalmigration. They are rare in the tropics, although typical of nomadism possess in the subtropics of the northern hemisphere, where people who temporarily that principally sheep and goats migrate to the various altitudes of carry fodder. Their routes are mostly shorter than those of the nomads to helps the plains, since the seasonal growth of fodder at various altitudes ensure an even supply. and Nomadic groups operate chiefly in areas with little access to a market, as fully aim at diversified production in order to cover personal consumption stock-keeping, namely production, of line as possible. They have only one With the result that this activity has to perform a disproportionate number of or comroles compared with that of any single activity in arable holdings mercial stock-keeping. It has to provide: (1) milk and meat for the household; (2) wool and skins for the household; (3) dung for fuel; (4) mounts and pack-animals; and plant (5) cash products to pay for the purchase of crop foods, other products, and non-agricultural consumer goods.
255 GRAZING SYSTEMS These aims of stock-keeping must be fulfilled as regularly as possible, because storage presents much greater problems to a nomadic than to a stationary in household. The natural conditions in a dry climate favour specialization and heat tolerating of capable are which animals such as camels and sheep, that trekking long distances. But the various needs of the household mean and mules, horses, by the dominant type of animal is often supplemented more a for cattle sometimes ,,unkeys for transport, and by goats, hens, and and diversified food supply. Cattle require a good deal of water and fodder not are they reasons are not so mobile as camels or sheep, and for these populastationary characteristic of total nomads but of semi-nomads and the milk, tion. The nomadic household depends especially on a regular supply of in supplies which is the staple food. The importance of the continuity of food blood nomadic systems is so emphasized that live-animal products (milk and tapped from live animals) are the ones most utilized, with terminal products only on (meat and the blood of slaughtered animals) being mainly consumed the like animals Thus 76). p. 1969, Hudson special occasions (Dyson and camel female the for preferred, are camel, with a long lactation period, calf supplies about 2-4 kilograms of milk daily more than is needed by a H. L. 303; p. 1962, during a 12-18 month lactation period (Capot-Rey Brown 1963, p. 9; Schickele 1931, pp. 35 ff.). fairly To ensure that animals, and particularly lactating animals, have a available the after regular supply of fodder, the nomadic household treks is used vegetation, the camp being shifted as soon as the fodder in its vicinity but move, up or becomes too dry. The nomad is by no means always on the camp, his mode of living and farming is characterized by a frequent change of is in any which in turn makes it practically impossible to crop fodder. There and case no permanent base at which to store fodder reserves, so that when nomad the because only overcome is where fodder is scarce the shortage portable selects and breeds animals with greater ability to store fat, as with areas that so distances, long reserves in the animal's body, and to travel little fodder can be crossed. per The aim in breeding is not so much a high yield of meat, wool, or milk disease, cold, heat, animal as a high level of resistance to trekking, drought, inand periodic shortages of food. Advances in breeding geared towards can feeding creased output per animal and requiring improved animal production scarcely be applied by nomads, who have to be prepared for a high of animal spread the to frequently leads migration risk, particularly because There are diseases whose control is made more difficult by the need to travel. conditions, weather extreme from animals no stables or shelters to protect shortage theft, or predators, and stock losses during droughts with extreme most that shows Experience of fodder and water are particularly serious. their of proportion iarge a nomads can expect that at irregular intervals animals will die.
256
GRAZING SYSTEMS
herds Because of the high production risks, the nomads try to accumulate but gain, financial not is aim that are as large as possible. The production emergencies, possible against rather the supply of household needs, provision of and increase in social status, which rises in proportion to the number conof assets animals owned. Nomads frequently have, in their livestock, as siderable value, but these cannot be regarded as economic capital, only are they hoarded wealth. Livestock do not represent or indicate wealth: satisfacwealth. Moreover they are a means of social fulfilment and personal tion, whereas comparable satisfaction cannot be obtained through money on (Dyson and Hudson 1969). The society of the nomads is not organized is but farmers, arable among find often we egalitarian principles, such as of possession The owned. animals of structured according to the number tribe. animals is by no means evenly distributed among the families of a animals. the of cent per 60-9G Normally 10-30 per cent of the families own it possible On the other hand, the natural conditions of production make weidthy the For scale. for the poor to acquire animals and rise in the social frequent A herds. nomad there is little sense in migrating with very large for the change of camp would be necessary, and to avoid this it is usual in farmers owner of a large number of livestock to copy the practice of arable by size adjoining areas and have his animals tended in groups of limited herdsmen, who are paid with a proportion of the young animals. of how The investigations of ethnologists and geographers give us an idea interan is example following the and nomadic stock-keeping is organized, (1970). Schinkel see nomadism, of esting case. For an informative study 9.2.2. A case study,- camel nomads in the Sahara picture, The nomads of the Sahara, of whom Chapelle (1957) gives a vivid milli150 than less live mainly by grazing camels on pastures that receive in and camels, metres of rain a year. The families each own four to fifteen In prosperity. of some cases even more than thirty, according to thei, level marginal addition they keep a few hens, sheep, goats, donkeys, and, in the following areas with more rain, cattle. In summer they nmigrate southwards, places. watering at camping northwards. go the rainfall, and in autumn they night, at out carried are at," common, are Journeys of 150-220 kilometres southern In far. as -',ice travc;! they but when there is little fodder available kilometres a Algeria and in the Fezzan, the camel nomads normnaiiy trek 500 kilometres. year, and in extremely bad years even 700-800 place depends on How the pasture is grazed in the catchment area of - watering zones are formed. grazing that so the distance the various types of animal can travel, and in days, three or two every only watered be to need In the hot season, camels of tolerant remarkably are the cool season only every two to three weeks.' They the northern I Webster and Wilson (1967, p. 337) write in this coanection: 'During by the browsed bushes and trees the of winter, or rainy season, the moisture content
GRAZING SYSTEMS 257 water with a high sodium and magnesium content. They rarely stray, and can defend themselves against small predators like jackals, and, therefore, when they are not needed near the camp for milk or as pack-animals, they graze in herds of thirty to forty at distances up to 80 kilometres from the camp. Where there is no danger of theft, camels are left to themselves, and they are therefore able, without any use of labour, to eat fodder that cannot be reached by other types of domestic animal because of the long distance from water. Table 9.2 illustrates the input and return of camel-keeping in Egypt, where, on account of the proximity of markets, favourable prices are obtained. According to the estimates of Alwan (1968), a family with ten camels would have a gross income of $327. Shepherds' fecs are a cost item only for those who own large herds.
TABLE
9.2
Gross and net annual cash income from 100 head of camels in Egypt
Gross return ($) Sale of young camels (20 at S103) Culling sale (12 aged camels at $91) Hair, hides, etc. Total gross return Expenditure ($) Shepherds' fees Watering costs Miscellaneous Total expenditure Net Cash Income ($) Note: E t = $2.29. Source: Alwan (1968, p. 33). Where cattle are kept, they too may graze in complete freedom, but only up to 4-8 kilometres I-om the watering place. In the hot season, cattle must be watered twice a day, and in the wet season every two to three days. Pastures that are a long way from the watering place cannot therefore be used by cattle. Goats and sheep graze close to the camp, but, since they ter.d to stray, and since they cannot defend themselves against jackals, they are guarded by day and put in kraals at night. Goat-keeping in this system involves an incipient form of feeding, since it is normal practice for the herdsman to lop branches off certain trees and to throw them onto the ground for the goats to eat. The nomads of the Sahara who keep large animals supplement their income from stock-keeping by collecting ed*le fruits and by hunting. Some species of gazelle can live without access to water, since they extract the moisture they need from fodder, so that by hunting the gazelles the nomads are able to make use of pastures beyond the range of their own camels. Nomads in general are reluctant to become settled, which is the prerequisite of arabic farming, but on the other hand they do need crop food. A household of five peo ', requires about 300 kilograms of grai'n and 500 kilograms of dates annually. dromedary is quite sufficient to supply the total water r.quirernent of the animal, and water may not be drunk at all during this period. In the dry season the vegetation is desiccated and extra water, over and above winter requirements, is required for heat regulation.'
18
258
GRAZING SYSTEMS
various combinations of The need to obtain grain without spending cash leads to stock-keeping: arable farming and nomadic without soil preparation. (1) Millet is sown here and there in damp hollows, time, which means that harvest at After sowing, the nomads move on and return implements. they practise a form of shifting cultivation without were subjugated dependants (2) Nomads employ settled workers whose ancestors
and who cultivate arable crops for them. frequently own land that (3) In the marginal areas of their territory, the nomads for payment in kind. return in out it hire can support arable crops, and they and oases are a further source Trade and transport between arable-farming areas to buy grain, meat, and skins, of income. The oasis dwellers produce dates but have Finally, the nomad
grain. whereas the arablc farmers need dates and can supply and the arable farmers' dates dwellers' oasis the for products exchanges his animal thus he isengaged in and transport, with grain. At the same time he provides them transport and trade as well as in stock-keeping. 9.2.3. Weaknesses of total nomadism could not be used by a Nomadism allows the use of areas of land that and a system of ranching. perman -' population without motor transport sums in tracks, vehicles, Compared with the rancher, who must invest large at a low level of expenditure fencing, and watering places, the nomad operates represents therefore and produces cheap animal products. Total nomadism situation, and that pre-technical a to an agricultural system that is adapted where price conditions might still be considered the best economic choice Total nomadism, ranching. in and natural conditions do not justify investment that are progress technical however, is not open to the opportunities of a check keep to expensive available to ranchers. It is technically difficult and productive more with on the health of wandering herds, and cross-breeding is not modified. Because types causes more harm than good if the fodder basis it is scarcely worth their of the marginal nature of the nomads' territories, fodder, sowing more while to try to improve the feed situation by cropping productive gracses, and rotating the pasture. however, with Static production techniques in total nomadism contrast, where they live. This economic and social changes in and around the area which include: presents the nomads with P_number of problems, is spreading into (i) Diminishing grazing areas, because arable farming the peripheral zones. a rapidly expanding (2) Loss of pasture in arable-farming areas, because on an increasing arable-farming population is practising stock-keeping scale. with traditional (3) Loss of income because the motor vehicle is competing reduces the sale animal transport-the use of draught cattle and tractors for draught camels have of draught camels, and in many places prices dropped to a fraction of their former level.
GRAZING SYSTEMS
259
(4) Loss of income because urban traders are competing with the nomads in their traditional trading function.. (5) Loss of the political control that they were traditionally able to exert over entire regions. (6) The fact that farmers who used to supply the nomads' need of plant products are gaining their independence. (7) The attempts that development administrators are increasingly making to control the incidence and limit the degree of movements of the human population, which is one major eement of present policy towards nomads. Movements of herds and herdsmen are, however, almost unavoidable in a typical nomadic setting. The nomads have had to accept a considerable reduction in status, employment, and income at a time when new requirements have established themselves, and the consumption of tea, sugar, and alcohol has become common. Moreover, the population of nomadic tribes has increased rapidly in recent times. Measured in terms of traditional methods of land use, their territories are proving to be more and more 'relatively overpopulated'. 9.2.4. Development paths of total nomadism Since nomadism is found mainly in areas where ranching is not a paying proposition, and since these particular people have little desire for change, it is usually advisable to leave them undisturbed, that is to say in the anteroom of economic development. It is particularly difficult and expensive to make nomadic tribes stationary and to integrate them into a modern ecoto nomy. The natural habitat does not favour the transfer from nomadism semi-nomadism or partial nomadism, where herds are concentrated near the camp, the surrounding pastures are overgrazed, and parasites and deficiency diseases increase. It costs $6000-15 000 per family to make them stationary within the framework of irrigation projects. The nomads' attainments in arable farming are mostly poor, unless every operation can be closely supervised, as in the Gezira scheme in the Sudan, but as a rule there is insufficient state control and state authority. Nomads are particularly reluctant to farm small areas intensively, and tampering with water is a thing that they very much dislike. Experience has shown that nomads can be induced to become .. tationary much more easily through mechanized grain cropping than through irrigation farming. Where this is not possible, towns and industry offer more incentives to become stationary than does crop production. Where employto ment in non-agricultural pursuits and in arable farming is insufficient wise socially and economically probably is it absorb the nomad population, to let them continue their traditional way of life instead of wasting funds i, trying to improve a land-use system that will anyway gradually disappear the course of economic development.
260
GRAZING SYSTEMS
9.3. Semi-nomadism
9.3.1. General characteristics of semi-nomadism section, is restricted to decidedly Total nomadism, as depicted in the last measured in terms of both marginal climatic zones. Its economic importance, very slight, whereas semi-nomadism subsistence and marketed production, is particularly in the African savanand partial nomadism, which are practised stock-keeping, with little -or no nas, are of greater significance. We find the semi-nomads of West Africa supplementary arable farming, among (Masai, Turkana, Pakot, (Fulbe), Madagascar (Sakkalaven), or East Africa are often quite large there but etc.). The principal animal kept is the cow, more important and even is herds of goats and sheep. Partial nomadism African tribes that practise a concharacterizes stock-keeping among those siderable amount of arable farming.' savannas is fairly uniform. The The organization of stock-keeping in the pastures, fallow, and harvest fodder basis consists of open bush, untended knowledge of their pastureland. residues. The herdsmen usually have expert As Allan (1967, p. 4) writes: species, which they They know the feed value of the different grazing-and-browse associations, ecological they recognize usually distinguish by specific names, and capacity at stock-carrying and value or pasture types, and can assess their in Kenya and Tanganyika pointed out different times of year. Masai herdsmen they regarded as good for supplying to me various species and associations that for improving the potency of bulls, animals, mineral deficiencies, for conditioning and for making milk and beef. to the growth patterns of different They often show great skill in adapting and to the different feed values of fodder plants in the rainy and dry seasons knowledge of the position, quality, shrubs, trees, and grasses. They have exact traditionally operate under technical and capacity of watering places, but they result in low output and make the conditions and with economic aims that or even impossible. introduction of technical advances difficult to watering places and to pastures, for The livestock have to go long distances wild animals and thieves. This prevents at night only a kraal can protect them from time. During the day, the night-grazing them from taking advantage of valuable provides better protection against wild herds often graze close together, since this continuously on the move. During the animals, but they are consequently almost in the dry season there is little. Annual rainy season there is plenty of fodder, but fodder for the dry season and help to grass and bush fires reduce the reserves of other hand, burning helps to control the impoverish plant growth and the soil. On grass growth in the dry season, stimulate to grasses, old remove growing bush, to is therefore a part of the system Firing ticks. and to eradicate parasites, particularly I Partial nomadism is further considered in 4.3.4.
GRAZING SYSTEMS
261
grasslands of the that can scarcely be dispensed with. Most of the extensive tropical raing. and tire of present day are man-made, uy the joint action in the dry season. There isneither the technical nor the economic basis for conserving fodder (straw, hay, and silage), nor isthere any such tradition. Consequently, rearing takes a long time. Fodder and mineral deficiencies cause long intercalving of the periods, and only about ahalf of the cows calve in any one year. Moreover, fewer. even places many in and survive, cent per 50 about only born are that calves A basic feature of semi-nomadism is the tendenry to hoard animalsmore possible than in total nomadism because of the better fodder basiscombined with the communal use of grassland, which allows everyone to send any number of animals to the tribal or communal pastures. Thus a tendency to overgrazing, i.e. the maintenance of animals on a piece of land to the detriment of its carrying capacity, isinherent and particularly obvious in the system, for the more animals an individual possesses the greater ishis share in the communal pasture. Onc factor inducing the hoarding of cattle is the high risk of production. In this connection, French (1967, p. 153) writes: The retention of aged beasts in the herd isdeliberately based on the knowledge imthat such older animals have usually recovered from and developed some than frequently less reproduce may cows older munity to several diseases. Such younger animals but they do provide an insurance against recurrent disasters. This retention of older animals and of mature males in excess of those required for breeding or work purposes, together with relatively high mortality rates of conditions must be associated with achange in herd compositions if their maximum productivity isto be realized. Even more important are the social factors. Cattle are property, and accordingly they represent variable degrees of wealth, of social status, and of community influence. They are a man's legacy to his son. They can be exchanged to symbolize formal contracts of friendship and assistance. The transfer of cattle from the groom's family to the bride's isneeded to validate a marriage (Dyson and Hudson 1969, p. 78). Herdsmen usually tend to hoard cattle in order to build a system of human bonds aimed at increasing individual security. This may be done in the following ways: (1) Some of the animals of relatives and acquaintances are kept in a family's own kraal and they give some of their animals to other people, so that in case of disease the losses for any one family are not total. (2) Herdsmen lend animals to a neighbour or relative who has lost his animals through disease or theft, and thus ensure his help in their own times of need. The tendency to dispose of animals by lending isencouraged by the fact that large herds entail a rapid consumption of grass and necessitate long treks.
young stock, result in herd compositions which are very different from those in developed aieas. Any attempt to improve pastures, feeding or management Animals grow up alternately gaining weight in the rainy season and losing weight
262
GRAZING SYSTEMS
not want to keep them (3) The owners of large herds, who in any case do of the tribe, and in in one place, lend some animals to poorer members this way guarantee their allegiance. to the bride's father. In (4) If a man wants to marry, he has to give cattle in rich families to ten cows poor families this amounts to one or two, and husband, she can return and more. If the woman is treated badly by her the cattle. Conversely, the to her father without his being obliged to return and demand back his cattle husband can send his wife back to her father both parties have a if she behaves badly or if she is infertile. Consequently, but equally both are material interest in the success of the marriage, in order to meet their lives their of obliged to hoard animals for some part obligations if the case arises. populations, it is crucial to Thus when an assessment is madr of animal as the motive to maximize remember the social role of animals as well in Sukumaland that productivity. Rotenhan (1968), for example, established animals were tied as bride40 per cent of the Wasukuma's partially nomadic conditions prevail among prices of 10-20 head of cattle per marriage. Similar semi-nomads. by the fact that all The tendency to overgrazing is further encouraged People who pasture. communal members of the tribe have access to the as seasonal work who or acquire household surpluses from arable farming, inclined therefore are labourers in the mining industry or on the plantations, arable Permanent as prestige. to hoard cattle in order to gain security as well in Fulbe the like farmers tend to lend out Eheir animals to semi-nomads means no by are West Africa. Thus animal-keeping and animal-owning sharply inflected particularly find we identical, and in animal ownership in a few families Lorenz curves, which indicates that stock is concentrated (see Fig. 4.4). Uganda 9.3.2. A case study: semi-noinadism of the Bahirna, graze grasslands in northThe Bahima have long-horned Ankole cattle, which with a rainfall of 700-900 and level east Uganda, 1000-1300 metres above sea than that of many other pastoral millimetres. Their economic position is better which receives plenty of rain and tribes, because they have at their disposal land the lack of watering places and a to due largely for the present is not overgrazed, farming. In the tribal area only bimodal pattern of rainfall, which prevents arable pastoral regions of East Africa, other In I I livestock units occupy 100 hectares. two to three times as high, where less fodder grows, the animal density is sometimes spared the devastating been yet as have resulting in overgrazing, but the Bahima causes. damage that overgrazing The head of the family The animals are the private property of the families.but the pastureland is the discretion, cwn his at bequeaths the animals to his sons of the tribe can graze as many communal property of the tribe, and every member has dug them for as long as whoever to animals as he likes. Watering places belong
GRAZING SYSTEMS
263
they are occupied; or:e a watering place is abandoned, anyone can use it. The herds of the Bahima are comparatively large, and on average a typical family of ten people owns fifty-three animals, which are practically all cattle. In general, several families live and migrate together. Temporary kraals and huts are built at seasonal watering places. Where plenty of water is permanently available, larger and more permanent kraals and huts are built, and are shifted about every two years. The Bahima have plenty of grass, so they dc, not move the kraals or construct temporary secondary kraals because they are in search of fodder but because they are: (1) searching for water; (2) trying to escape disease through movement; or (3) complying with the custom of abandoning a place where an adult has died and is buried. Calving takes place in the rainy season, and normally one hundred cows produce seventy calves. Most bull calves are slaughtered immediately after birth, so that there is as much mil, ;,s possible for human consumption. Of the calves that the Bahima try to raise, about 20 per cent die in the first "ear, so tha from one hundred cows twenty-eight calves are brought up each year. A family owning fifty-three cattle of which twenty are cows, as shown in Table 9.3, could have five to six new animals entering the herd each year. However, account must be taken of high TABLF 9.3 Returns in cattle-keeping of the Bahima, Uganda, 1964 Persons per household Stock numbers (LU) Head of cattle Number of cows included Area available per head of cattle (ha)' Area available for the total stock (ha)" Economic return per household Cattle (kg live.weight)b Milk (kg)o Blood (kg) Gross return of cattleand milk ($)d Proceeds from sellhg Proceeds from selling as % of economic return Pe-oductivity Cattle (LU per 100 ha) Cattle (kg live-weight per ha) Milk (kg ha- 1) Gross return ($ ha'. )
I0
41 53 20 7.2 383 1700 6000 400
419
41 10 II 4.5 15.7 11
The tribe's area of land divided by the number of cattle. Including animals which died and which were eaten. In addition to the milk co isumed by the calves. $0.07 per kg live-weight of cattle; $0.03 per kg milk. There isa market for cattle but not for milk. Thus the calculation of the return on milk is hypothetical. The holding return is supplemented by collecting honey and by other minor activities. *Officially registered sales. As well as these, the herdsmen exchange animal products for the plant products of neighbouring arable farmers. Sources: Parsons (1960), Brown (1963).
264
GRAZING SYSTEMS
losses (5-10 per cent) through disease and accident, which reduce the number
entering the herd to three to five animals a year. The Bahima live chiefly on milk and meat, and animals that die through accident or disease are for the most part eaten. Blood is consumed chiefly when there islittle milk available. For some years the Bahima have been exchanging animal products for sorghum, legumes, and sweet potatoes from neighbouring arable farmers. A growing number of men marry women from neighbouring arable-farining tribes, who are then allowed to lay out gardens and plant crops. The return of cattle and milk in terms of local prices amounts to $419, but only 10 per cent is actually sold. This is therefore most decidedly subsistence farming., 9.3.3. Weaknesses of semi-nonladism Most types of semi-nomadism may be termed economically wasteful. In comparison with large-scale production on ranches, the productivity per hectare, per man-equivalent, and per animal is usually low. Where plant growth has not yet suffered severely from overgrazing, animal densities are often high, since the stock-keepers know how to maximize the number of animals on a given area. But where long-term overgrazing has been practised, the animal numbers per 100 hectares are already smaller than on ranches. In south-west Africa, for example, the animal densities in reservations are on average only 75 per cent of those on ranches operating under the same soil and climate conditions. However, lower productivity per animal is much more serious economically than the low animal densities. Whereas on ranches 20 per cent of the animals are normally sold every year, and these are highquality animals, the semi-nomads and partial nomads of Africa slaughter and sell only 6-10 per cent of their stock, and these are poor-quality animals. L. Brown (1963) goes so far as to suggest that the efficiency in meat and milk production among the semi-nomads and partial nomads of Kenya is only one-third of what ranches achieve in the same conditions. The low productivity of semi-nomadic stock-keeping cannot be attributed to unfavourable natural conditions, which is the case among the total nomads, but is caused primarily by the combination of individual animal ownership with communal grazing and pre-technical veterinary methods. The tendency to overgrazing and to neglect of those measures that could help to increase the efficiency of grazing is rooted in traditional institutions. Everyone tries his best to maximize his number of animals, and no one feels it incumbent upon himself to improve the land, since he has to share it with others. Communal bodies that undertake to organize the use of pastureland
1 The situation of the Bahima, as described by Parsons in the mid-fifties, has changed drastically in recent times. I observed in 1971 that more and more Bahima are turning to settled housing and plantain cropping. A tsetse eradication scheme has extended the grazing area and, together with the introduction of dipping, has led to high rates of increase in cattle numbers without overgrazing. Many herdsmen are no longer at subsistence level, but raise cattle for sale. Here semi-nomads have been changed to commercial producers by a sudden drastic improvement in their situation.
GRAZING SYSTEMS
265
have not been able to survive, or even to establish themselves. Persistent vegetation harbours ticks, flies, and worms, while 3verstocking favours aridity and reduces the incidence of parasitic diseases. Consequently the pastoralists, who are not familiar with modern veterinary practices, consider overgrazing an unavoidable evil. The combination of these factors leads to the permanent overgrazing of a large part of the grassland, and a decline in the quality of the pasture sets in. Perennial grasses with high returns per unit area are displaced by annual grasses and bush. Areas prone to erosion are grazed without any consideration until the grass and the cover are totally removed. L. H. Brown (1963, p. 16) describes a district in Kenya as follows: Baringo District, has reached an 'overgrazing end point' where most of the grass and the topsoil has already gone over larg- stretches of country and the ground isblanketed with thornbush, largely useless to man and beast alike, which cannot be eradicated without the expenditure of large sums of money... Land in a 35-40 inch rainfall area, once capable perhaps of supporting a stock-unit to 4-5 acres, isnow scarcely capable of carrying a stock-unit to 20 acres... In the drier part of the district with a 20-25 inch rainfall, land once capable of supporting a stock-unit to 10 acres is now scarcely capable of supporting one to 30-40 ... Baringo is a case where the human population, in an attempt to maintain enough stock for their needs, have already to a large extent destroyed their own habitat. A consequence of communal grazing and private cattle ownership is the so-called cattle cycle. The first stage is a relatively small number of animals, plenty of fodder, sufficient grazing in the dry season, and a high rate of reproduction. With increasing numbers of animals, the fodder supply becomes less adequate, and there is a particular shortage in the dry season. There is still sufficient grazing to cover the animals' needs in the dry season, and the stock can reproduce, but at more or less regular intervals there is an extremely dry year, when fodder growth and water are insufficient, causing some of the animals to die. In the following year the cycle begins again, with fewer animals, plenty of fodder, and a high rate of reproduction. In some parts of Tanzania, for example in the Masai area, this cycle seems to take about ten to fifteen years to complete. In partial nomadism the damage caused by the inefficient control of cattlekeeping goes beyond the loss of animals during the dry seasons. In arable farming with cash crops, a relatively large amount of money is earned, and it can be assumed that the more successful third of the farmers save considerable sums. Some part of these savings are invested in stock and thus are lost to the nation's economic capital. In addition there are such developments as, for example, the 'cotton cycle'. Cotton cropping brings in considerable amounts of cash, a part of which is saved and invested in stock. This results in a concentration of stock in cotton-growing regions, with the simultaneous
266
GRAZING SYSTEMS
extension of arable farming. The area of pasture per livestock unit decreases and erosion damage increases. The animal-feed situation becomes worse than usual. In extremely bad years the result is a loss of animals, which again stimulates the desire to buy more. Cotton cropping, the way to earn cash, is extended even further, and the process begins all over again. In particular, semi-nomadic stock-keeping is characterized by the following malpractices, which also apply to partial nomadism: (1) Uncontrolled animal densities, overgrazing, and erosion damage in healthy habitats near settlements and at watering places. (2) Insufficient fodder reserves, and thus an inadequate fodder distribution over the year. (3) Hoarding of fully grown or infertile animals. (4) A lack of information on the means of checking animal diseases. (5) Long daily treks between the kraal, the pasture, and the watering place. (6) Low levels of calving and a high mortality rate among calves-the calves compete with people for the milk from the cow and they are not fed adequately in the first months. (7) Daytime grazing, often in the midday heat, instead of grazing at night or in the evening or morning. (8) Theft of stock is regarded by several pastoral tribes as an honourable activity, so that animals have to be driven into a kraal for the night. (9) Lack of systematic breeding-the camel nomads do carry out planned to breeding, but the semi-nomads and partial nomads allow their cows mate with unselected bulls. New areas for cattle-keeping can be opened up by constructing watering can be places, since cattle should be watered twice daily if possible. Grass
FIG. 9.1. The size of the
grazing area, showing its dependence on the distance from th aering place to the kraal. (1)Watering place at the boma: large pasture area. (2)Watering place removed: smaller pasture area. (3) Watering place far removed: small pasture area. From Groeneveld (1968, p. 228).
\,,
grazed for 4-6 kilometres round the watering place, so the available pasture is a function of the distance between the kraal and the watering place (see the Fig. 9.1). Since there are few watering places, overgrazing is the result in
GRAZING SYSTEMS
267
catchment area. Digging new watering places alleviates these conditions only temporarily. With little control or management of grazing, the range deteriorates around the new watering points. Thus there is the danger that the range depletion is being spread by the measures intended to remedy it. Meanwhile, demands for additional water development continue to increase with the steady build-up of livestock numbers and the continuing decline in carrying capacity and range productivity. To stop this vicious cycle, and avoid its predictable consequences, water development obviously must be accompanied by appropriate management to maintain range productivity. The health of cattle must also be maintained, by veterinary means or measures to eradicate the tsetse fly. 9.3.4. Development paths of semi-nomadism In the economic development of regions with semi-nomadism that are not suited to arable farming, the steps promising to be successful involve: (1) A change to a different way of earning a living-there are far more workers than can be sensibly employed in stock-keeping in dry regions, and the relative degree of overpopulation in the thinly settled dry regions is mostly greater than in the densely settled, high-rainfall arable regions. (2) Establishment of organized management of grassland, supported by veterinary practices. Whereas the total nomads usually use areas that are not amenable to commercial methods of farming, the methods of the semi-nomads dissipate the very resources that would make technically modern management worth while. But before grassland can be managed in an organized fashion, institutional controls have to be established that provide incentives to introduce technical advances. In this respect, there are a number of possibilitie,. (a) Individual ranches. On the subject of individual ranches, Webster and Wilson write (1967, p. 280): In view of the great disadvantages of communal grazing, it isoften suggested that a change to individually owned and operated holdings is desirable, in order to obtain good pasture management. However, over a great deal of the pastoral country individual holdings are impracticable for several reasons. First, seasonal migration of stock isoften essential because large areas of pasture can only be used for part of the year. Secondly. provision of water to individual holdings is often impracticable or uneconomic in areas where water is scarce. Thirdly, individual holdinGs would often be unacceptable because few pastoral regions are uniform. Under individual tenure some people would get only poor grazing, whereas under the communal system all have equal access to good and poor areas. Fourthly, the fencing of pastures and water supplies on individual holdings would be expensive, and in many places the absence of fencing materials would make it impracticable. Individual ranches are therefore only sensible where the rainfall and drinking
268
GRAZING SYSTEMS
water are sufficient for an intensive pasture economy, or where there is an abundance of grassland. The Bahima in Uganda, for example, have plenty of grassland with a relatively high rainfall, and smaller ranches of 1300 hectares carrying 530 livestock units are being set up within tile Ankole ranching scheme. But normally division of semi-nomadic and partially nomadic stock-keeping land into individual ranches .s not an economically sound solution. (b) Grazing schemes. Grazing schemes of a type that has been tried in various parts of Africa are a further step towards rationalization of the pasture economy in dry regions. The herdsmen are offered a number of investments and services, on condition that they submit their use of the grassland to controls checked by a grazing guard. Projects like this usually involve a number of measures: (I) a regulation for animal density, which stipulates in most cases culling programmes; (2) veterinary supervision and the payment of veterinary services; (3) digging watering places so that the grassland can be grazed more evenly; (4) the introduction of a grazing rotation; (5) the control of grass-burning in the dry season, clearing the bush, and occasionally sowing more productive grasses; (6) prohibition of grazing on badly eroded stretches; and (7) the organization of marketing. In their early stages, grazing schemes are under the wing of the agricultural administration, but the gradual transfer of management to grazing committees is usually planned. Experience shows that projects of this kind succeed in a few years in promoting animal health, regenerating fodder growth, considerably increasing the grazing capacity, and consequently raising the income of the stock-keepers. But the grassland projects cannot be reconciled with the social function that tradition gives to stock-keeping. In most cases they fail after a few years because of the herdsmen's opposition, and the following are the main reasons for their collapse: (1) Many of the pastoral tribes are already living on a sub-marginal level. The reduction in stock numbers at first reduces the supply of mi!k and meat for the household. The herdsmen and their families apparently do not have the means to bridge the time between the reduction ir: stock and the higher return in milk and meat as a result of it. (2) Culling programmes simply do not fit into the integrated system of human bonds established by the lending, renting, exchanging, and presenting of cattle for different reasons in different situations. (3) The limiting of stock numbers under the rules means that each family herd has to be split up into two-some animals are in the quota on the
269 GRAZING SYSTEMS scheme and the remainder have to find their grazing elsewhere. As a result, herdsmen are separated from their families, and naturally the idea is unpopular. (4) The reduction in stock numbers puts the greatest relative strain on the man with a lot of cattle, who is frequently an open or covert leader of the clan or tribe. Therefore grazing schemes are often opposed by those who have influence in their communities. (c) Group ranches. In the majority of the range areas, economical and ecological considerations may make it acceptable for the pastoralists to come together on the basis of clans, families, or other groupings and establish ranching units, which might comprise two to twenty families and a stretch of land of between 600 and 12 000 hectares. So far, the individual generally exercises grazing rights over the entire area recognized as the grazing ground of his clan. He thus has, along with other members of the clan or tribe, freedom to move with his herds over a considerable area, but has no specific individual rights to a particular portion of it. The new move would mean the division of the area into self-contained units as far as water is concerned. Only people belonging to a certain group of families would then have the right .to graze there. The ownership of the land could go to the group or clan, provided that this could be legalized. Another perhaps more realistic approach would be the establishment of group ranches on the land of several individuals. However, in each case the individuals would retain the ownership of their cattle. Some of the Masai, for instance, are reported to have decided to divide their land into units of over 8000 hectares, to be owned by family groups whose irterests are homogeneous and who have traditional respect for the elders in the group. It is intended that the land of each ranch shall be registered in the participants' names and that a management committee shall be formed whose function will be to run and develop the ranch, taking up ioans when necessary. (d) Cooperc'tive or company ranches. Another possibility is the establishment of cooptrative or company ranches. These should be run by a manager and raise their own funds. The individuals concerned might receive shares according to their contributions. They migit be employed by the ranch management where this is convenient. Actually, the name 'cooperative' can be misleading. The institutional set-up is often much closer to that of a company than of a cooperative. (e) Government ranches. Wherever grassland areas are available that are not claimed by pastoralists and cannot be handed over to private enterprise, the establishment of government ranches is advisable. In order to guarantee efficient management, they should be established and run by autonomous bodies as profit-making enterprises. (f) Nucleus ranches. Another possibility to solve the problems of seminomadism is the combination of private or state nucleus ranches with the
270
GRAZING SYSTEMS
supervision of shifting herding in surrounding schemes. Such an arrangement has the following characteristics: (1) An area newly opened up by watering places is declared a scheme area. (2) Herdsmen are allowed to become scheme participants on condition that they submit their animal-husbandry practices to a clearly defined set of rules, the execution of which is closely supervised. (3) The herdsmen are allowed to join the scheme with a limited number of cattle, the health of which is checked. No other cattle are allowed to graze in the scheme area, aid no scheme cattle are allowed to graze elsewhere. (4) The herdsmen remain the owners of their cattle, but grazing, dipping, and castration are organized by the scheme management. All selling and buying of cattle has to be channelled through the scheme management. (5) The scheme management is backed by a neighbouring nucleus ranch, which provide; the services, the veterinary supervision, and the marketing facilities for the scheme cattle. In group, cooperative, government, and company ranches, individual subsistence farming may well be incorporated into the enterprise. The allotment of land for cropping to members and workers within a system of shifting cultivation provides cleared land that can be planted with better grasses at low costs. Arable farming within a ranch may thus contribute to the reduction and control of bush and to the establishment of better grasses. However, two basic requirements must be fulfilled if money is to be invested in pastoral areas, namely: (1) Limitation of the number of stock that attempt to graze any area to the number that area can carry without deterioration over a long period of time. (2) Control of the movement of stock within any given area through rotational grazing systems or other such measures designed to maintain or improve the quality of the pastureland. Without these two basic practices, no other range-improvement measure-whether additional water supplies, tsetse eradication, reseeding, or bush control-is likely to be of any lasting value. This is also true of veterinary services, which-except with devastating diseases like rinderpest--can be of economic value only within the framework of a rational system of land use. Finally, it must be pointed out that the agricultural problems of most semiarid areas can be solved only by the removal of a great number of people from the pastoral way of life. There are often too many herdsmen who try to live on their stock and who therefore attempt to retain, against advice, as many stock as they can for their own survival, thus continuing the destruction
GRAZING SYSTEMS
271
of their habitat. The relative degree of overpopulation is usually higher than in densely populated, high-potential areas. 9.4. Ranching 9.4.1. General characteristics of ranching Ranching, which is the commercial alternative to the various types of nomadism, is carried out on large stretches of land. Most ranches cover several thousand hectares, sometimes more than 100 000, and carry large, permanent herds of some 1000 or 10 000 animals. The term ranch is generally used to refer to properties with well-defined boundaries (fenced or unfenced), legally owned, or having a long-term lease, and with certain developments present that were effected by the owner or lessee. In ranching, the number of animals and workers, the capital investment, and the return per hectare are very low. However, ranches have large stock numbers and a large fixed capital per man-equivalent invested in, for example, fences and pumps, and there is usually a high income per man-equivalent. Production is limited to stock-keeping, and cropping of arable fodder is worth while only under favourable price conditions, which are rare in the tropics. In most cases, only one type of animal is kept in any considerable numbers, and the aim is a single product, like wool or animals for slaughter, and other products, like the sheep's meat or the cattle's milk, are neglected. These are mostly marginal zones where diversification is not commercially worth while. In particularly dry, stony areas far from markets, sheep-keeping predominates (south-west Africa, Australia, southern Argentina). Cattle do not have the same mobility, and are more demanding with regard to the frequency of watering and the amount of fodder. Whereas sheep are kept in some cases where only 150 millimetres of rain fall annually, cattle-keeping usually requires more than 400 millimetres. We find large-scale cattle ranches chiefly in the tsetse-free savannas of East and southern Africa, in Australia, and in Latin America. The main objectives of ranch management are summarized by Webster and Wilson (1967, p. 256) as follows: (I) To provide, as far as possible, a uniform and year-round supply of herbage for the maximum number of stock. (2) To utilize the herbage at a stage that combines good nutrient quality with high yield. (3) To maintain the pasture in its most productive condition, by encouraging its best grass types and by promoting as full a ground cover as possible. This will protect the soil from direct solar radiation and the beating action of rainfall, thus preventing run-off and erosion.
272
GRAZING SYSTEMS
In evolving methods of management that will attempt to achieve these aims, ranch owners must give consideration to the following factors: (1) the influence of seasonal growth and of grazing on the maintenance of the sward; (2) the variation in the composition and feeding value of the herbage with stages in growth; (3) the value of certain trees and shrubs as browse plants; and (4) the need for bush control. There are considerable differences in the measures used by the ranchers to achieve these objectives. Ranches can be classified according to the following indicators of land-use intensity: (I) Whether pastures are fenced, or to what extent the fencing divides the area. (2) How far the watering points are from one another, or how much capital is invested in pumps, pipes, and troughs. (3) Whether and how the pastures are managed (control of bush, planned burning, weed control, sowing of more productive grasses, rotational grazing, etc.). (4) Whether and to what extent the supply and demand of fodder are balanced 'passively' by: (i) calving or lambing at a time when there is plenty of fodder; (ii) which type of production is chosen-on extensively managed ranches production "spreferred that is less dependent on an even fodder supply than prcduction on intensively managed ranches, so that rearing animals is preferred to fattening them or dairying, and wool is preferred to mutton; (iii) fodder reserves 'on the stalk';' and (iv) balancing of fodder growth and animal numbers by selling when fodder becomes scarce, and buying when it is particularly abundant, or balariced 'actively' by: (i) the establishment of improved pastures; (ii) cropping of arable fodder on irrigated plots; and (iii) purchase of fodder.
Haymaking is rare in ranching systems for the following reasons (Webster and Wilson 1967, p. 269): (1) Much of the land is unsuitable for mowing owing to its topography, uneven surface, or the presence of rocks, ant-hills, trees, and bushes. (2) The low yields per hectare make haymaking uneconomic. (3) The grass reaches an appropriate stage for hay at the height of the At this time the weather makes haymaking either impossible, or at least difficult rains. and liable to heavy losses.
GRAZING SYSTEMS
273
From the point of view of intensity, we distinguish chiefly between the open-grazing system, the field or paddock system, and paddock systems with an actively balanced system of fodder supply. In the open-grazing system, the herds must be continually watched by herdsmen. The productivity per animal and per hectare is relatively low, because rotational grazing and separation of age groups, etc., are difficult to achieve without fencing. In the field or paddock system, the work of guarding the animals is replaced by investment in fencing and watering places. Market productivity is higher, chiefly owing to the better organization of the fodder economy. The third stage in ranch
Stage I Stage II
l4~4
Stage III SI
I I
o 3 :
! Reserve paddocks
FiG. 9.2. Stages in ranch development. Stage 1: no fencing, one water point, long daily migrations. (I) Bare soil, some weeds: (2) overgrazed area, vegetation mainly weeds: (3) overgrazed, some grass, much weed vegetation: (4) good pasture; (5) natural vegetation, scarcely grazed. Stage II: no fencing, several water points. (2)-(5) as in Stage I. Stage IllI: paddock technique. In the dry season the reserve paddocks are grazed. From Andreae (1966, p. 23).
development, which is usually limited to ranches with irrigation facilities or in high-rainfall areas, is characterized by substantial investments in an actively balanced fodder supply and a correspondingly higher productivity per hectare. Fig. 9.2 shows the principal stages in the development from open-grazing systems to paddock systems: (1) Stage I: Where a ranch has only one watering place, concentric grazing rings are formed. Zone (1) is bare ground and secondary bush
19
274
GRAZING SYSTEMS
round the water, with scarcely any fodder in the dry season. Zone (2) is an overgrazed area with predominantly weed growth and little fodder in the dry season. Zone (3) is overgrazed area with weeds and annual grasses. Zone (4) is a reasonably used area with good pasture and fodder reserves available according to the grazing technique. Zone (5) is scarcely used natural vegetation, since it is a long way from the water (4-8 kiiometres), and provides fodder reserves in the dry season. (2) Stage 1I. During the season when fodder is scarce, the fifth zone is a reserve. In so far as the dry season coincides with the cool part of the year, the animals can travel further and make use of the fodder growing in the outer zone. As several watering places are dug, the marginal zones of the ranch can be more easily developed. The distances covered by the animals become shorter. The result is the formation of vegetation zones round each watering place, with a better distribution of grazing and less overgrazing. (3) Stage III. Dividing the pasture into paddocks combined with fencing reduces the damage from overgrazing, facilitates separation of the animals into age groups, and allows reserve paddocks to be formed with hay on the stalk. The reserve paddocks are grazed in the dry season. Webster and Wilson (1967, p. 266) write: A very simple system is that known as the 'three herd-four paddock' system. Each paddock is grazed continuously for three years and in the fourth year grazing is stopped just after the beginning of the rains, the accumulated grass is burnt off at the end of the dry season, and the paddock is then rested for a further one or two months at the beginning of the next rainy season. The capital invested in progressing from Stage I to Stages II and III means that more cattle can be kept at a lower risk and will cause less damage to the pasture. It reduces the input of labour and shortens the time it takes to rear animals for slaughter from four to five years to three to four years. The intensity of grassland use is reflected in the animal densities, which fluctuate between 2 and 50 livestock units per 100 hectares, according to local conditions. The lower limit is where fodder and water are so scarce that permanent stock-keeping becomes impossible, and nomadism or hunting is the most economic activity. Ranches supporting enough fodder to keep 30-40 livestock units per 100 hectares are mostly found in zones where the rainfall is sufficient for arable farming and where, if there were no lack of labour and markets, arable farming would be more competitive than ranching. 9.4.2. Two case studies (a) A Karakulranchin south-west Africa. Sheep ranches differ as to whether they aim to produce wool (as in Australia, Argentina), or skins (as in southwest Africa). Mutton is a by-product, if it is valued at all, on the large sheep
GRAZING SYSTEMS
275
ranches. The production of Karakul skins is suited to particularly dry areas. The majority of Karakul lambs are slaughtered immediately after birth, with the consequence that the ewe is less burdened and in a better state to survive the dry season. Schafer (1963) provides an example of how a Karakul sheep ranch in south-west Africa is organized (see also Gundert 1962). The farm 'Haribes', whose data are contained in Table 9.4, comprises 77 500 hectares of partly stony and partly sandy bush savanna, whose average rainfall over many years is 200 millimetres, although there are fluctuations from 90 to 340 millimetres. This area supports 10 000 Karakul sheep, 800 goats belonging to the workers, and a number of cows to provide milk for human consumption. When the grazing is ample, about 100 head of cattle are bought for fattening, but when fodder is scarce this does not occur. The main production aim is the skin of the Karakul lamb and the sale oi breeding animals. The total area is divided up by fences enclosing between 5000 and 19 000 hectares, and is then subdivided into twenty-four sections according to the position of the watering places. Eighteen sections are grazed in turn for two weeks at a time. Six sections, a quarter of the total area, are not grazed at all, but constitute a fodder reserve in case there is insufficient rainfall. Shepherding has been replaced by jackal-proof fences. We can distinguish a long and a short lambing season. Two-thirds of the births occur between May and August, and one-third in January and February, and at both these times of year fodder is relatively plentiful. The sheep are shorn after lambing. The lambs are taken immediately after birth to the main yard and sorted to see whether they are suitable for breeding or whether they have good skins. Slaughtering takes place in the yard. Karakul skins contribute up to 50 per cent of the return from the ranch, which amounts to $72 500, and of which 29 per cent is obtained from selling breeding animals. The income from cattle-keeping varies according to the amount of fodder. Wool and mutton make no contribution. The local farm price of a sheep is $2.80, which is lower than for a good lamb's skin. The traditional practice iGto keep ewes as long as they bear lambs, which in effect means until they die of old age. In recent years it has been usual to slaughter and sell the ewes at the age of about seven years. The salaries of the ranch managers are the main expense. They are higher than elsewhere on sheep ranches, because of the high skills required in breeding and producing Karakul. Wages are the next major item of expenditure. Material costs consist mainly of vehicles, maintenance, and the depreciation of fixed assets, particularly watering places. The productivity per hectare, at $0.93 per hectare, of which $0.13 is profit, is less than in all other forms of commercial land use, but the gross return per man-equivalent amounts to $942. (b) A cattle ranch in Ken ya. Cattle ranches, which are economically much more important than sheep ranches, may aim at different products. They can be differentiated as: (1) ranches whose production aim is to raise and sell cattle for fattening; (2) Ranches for raising and fattening, whose end product is animals fattened for slaughter;
276
GRAZING SYSTEMS
TABLE 9.4
2400 5580 60 -
n.a. 100
-
72 500
GRAZING SYSTEMS
277
(3) fattening ranches, where the animals from native or commercial ranches are fattened; or (4) dairy ranches. This division of labour bei'ween ranches for breeding and ranches for fattening is characteristic of countries with advanced economies based on labour division, like the United States of America or Argentina. It is rare in the tropics. Fattening ranchc.s are only occasionally found, where nomads, seminomads, or arable farmers sell their unfattened animals, as for example in East Africa or Madagascar. They are of some importance in those areas of Latin America (northern Colombia, Sac Paulo, Brazil) wiLere markets with substantial purchasing power are close to foddo'r-producing areas, where fodder growth at the various altitudes is conducive to the division of labour, and where immatures can be obtained in sufficient numbers at a relatively low price. Dairy ranching is also tied to the proximity of urban consumption centres. It is of increasing importance in, for example, Latin America, South Africa, and Kenya. The data in Table 9.4 apply to a cattle ranch in Kenya, in an area with 600 millimetres of rain and enough fodder to support 22 livestock units per 100 hectares. The production aim is animals for slaughter, for the urban market, and for export. The greater part of the ranch, which colsists of 32 000 hectares, is divided by fences into paddocks of 2000-3000 hectares, and in each there is watering place. The animals, grouped according to age, occupy the paddocks in a four-year rotation: year I, breeding herd; year 2, heifers, year 3, oxen; and year 4, no grazing, regulated burn. In the dry season the grazing rotation is interrupted, and all three paddocks are open to all the animals. The normal animal density is 80 per cent of the fodder capacity, so that even in dry years there is sufficient grass. The chief maintenance expense is control of bush regeneration; as well as burning at fouryearly intervals, heavy equipment has to be used on some sections. The calves are born in the rainy season, and eighty calves are reared per one hundred cows. The animals grow up alternately increasing in weight at times of abundant fodder and maintaining their weight in the two to three driest months, when they are not given any additional fodder. The increase in live-weight amounts to 19 kilograms per hectare per year. The cattle are an imr roved Zebu variety: the Boran. They are sold for slaughter when they are 3.1-4 years old and at a weight of 350 kilograms. With the price of cattle at $15.50 per quintal, the return from the ranch comes to $127 500, which is $3.98 per hectare. The major expenditure consists of wages and the salaries of ranch managers. Next in importance comes the depreciation of fences, pumps, and other fixed capital. (c) Ranches in higher-rainfall areas of South America. Table 9.5 gives the farm-management data of ranching in various South American countries. Although operating on land with much higher potential for plant pro luction than that in the Kenya case study, they show the general characteristics of extensive stock-keeping: large size of holdings, low labour input per hectare,
278
GRAZING SYSTEMS
TABLE
20000 20000 Size (ha g"zed) 14 14 Workers per ranch 0'07 0.07 Workers (ME per 100 ha) 6700 4500 Livestock (LU er ranch)' 592000 459000 Ranch c pital ($) 34% 43., Land 34 % 46% Cattle 32% 1, Others On-farm prices 0'14 Cattle (S per kR live-weight) 40 Labour (0 per - ME and year) 10 Land ( ha ) Yield 20 17 Take-off %' 30 18 Production (kg live-weight per ha) 90 80 Production (kg live-weight per LU) 83800 50400 Gross return (S per ranch) 26800 18900 Costs of production' 57 000 31500 Net return Productivity 33 22 LU per 100 ha 4-18 2'52 Gross return (1 ha- 1) 1 5980 3600 Gross return ( MEt ) 317 1.89 Value added ( ha- ) 1 4530 2720 Value added ( ME- ) 2.85 1.57 1) Net return ($ ha9'7% 6'8% Return on capital
held. Since the 1 LU = I adult cow held for one year. Other stock is computed Proportionally to its live-weight an :ine not necessarily average weight of adult cows generally differs between areas and levels of management, the figures are Proportional to live-weight carried. in quality but prices beef of level Average for all animals sold. Thus. Prices presented reflect differences not only In general as well. for the compulsory expenses Unskilled labour. Includes wages in cash. wages in kind at on-farm prices, and social security
amployer.
GRAZING SYSTEMS
279
9.5
Colombia
Llanos 1500 Case study Average 3000 3'4 011 820 96000 495% 44-5% 6!1 0-17 350 16 23 35 105 18500 8510 9990 33 616 3860 3-89 2430 333 75'. 1500 Model Top 3000 4'8 0'16 1000 134000 36% 47% 17%
Apure (western Llanos) 1700 1700 Case study Model Good Average 20 000 21 0.Il 5330 520000 26% 68% 6%. 0-30 740 610 8 8 29 47800 19 ISO 28 650 27 239 2280 2-18 2080 143 53u. 14 13 48 70500 32500 38000 27 352 2820 282 2280 1.90 6-6'. 20 000 25 0'13 5330 573000 24% 62G, 14'.
Venezuela
Eastern Lianos 1300 Case study Average 7000 7 0.1 860 297 000 51% 38% II1% 0-35 890 22 21 10 78 29 200 10 300 18900 12 417 4170 3'59 3585 2-70 63% 19 22 106 62800 28400 34400 21 898 3690 707 2920 491 9-1 1300 Case study Good 7000 17 0.24 1450 396000 39% 48% 13%
Venezuela
Includes wages for family labour and depreciation: excludes interests paid. In the case of family-sized units (4 or fewer ME). no allowance has been made for an administrator or man. -'r. t Milking is common practice. The flaures for beef production (10 i, milk - I ki beef live-weight), yield in terms of production. and income take milk into account. Source: Owen (1969).
280
GRAZING SYSTEMS
high investments in relation to labour costs and to output, a low productivity per hectare, but high returns per man-equivalent. 9.4.3. Weaknesses of ranching Ranching is far from being a system that can te run particularly easily or lucratively, since it is characterized by: (1) high fixed costs and very high initial investments; (2) a long-term commitment to a particular type of production; (3) high-risk production because of the constant threat of drought; and (4) high costs for personnel in relation to the return from the ranch. The high fixed costs stem less from the value of the land than from the investments necessary to develop and utilize natural fodder. The costs of animals are the primary outlay. To build up a ranch requires either a high level of buying, or slow stocking with animals born on the ranch, and at least ten years of low return have to be bridged. Another costly item is the procuring of water. Grassland is of scarcely any value for cattle-keeping unless there are watering places within 4-5 kilometres; rotational grazing depends on watering places in every paddock. The watering places should have a reserve capacity of 50 per cent, so that sufficient water is left even in dry years to maintain the rotation. There is also the investment in fences and the cost of theit maintenance, for ranching without fencing is normally not a viable activity. Fences reduce labour, facilitate the control of animal densities, permit the separation of sexes and age groups, simplify veterinary care and breeding, and make it possible for cattle to graze at night. On the other hand, they are expensive in relation to the return from the fenced-off areas, especially as wide use is normally made of long-lasting iron or wooden posts, which have to be transported from far away and impregnated against termites. The result of investment is a long-term commitment to a particular type of production. The selected area is usually suitable only for either sheep or cattle, and it is virtually impossible to adapt production if economic conditions change. The three related factors of high initial investment, specialization in a single type of production, and operation in marginal areas inevitably entail a high level of market risk. Ranches must also consider the unreliability of the rainfall. There is a danger in planning the animal density on the fodder capacity of average years, when a few consecutive dry years can result in loss of animals or hurried sales. Furthermore, ranches with a single product are at the mercy of price fluctuations. The high expenditure on personnel is a further feature of large ranches. Admittedly these costs are extremely low per hectare, but they are high in relation to the return of the ranch. A large part of the expenditure pays for
GRAZING SYSTEMS 281 the services of the ranch manager, especially where diseases are particularly hazardous, as in sheep-keeping in the tropics generally, or in cattle-keeping in East Africa, where east coast fever, the tsetse fly, and other dangers are present. Ranch managers with some veterinary knowledge are needed, and they can demand a correspondingly high salary, which can be covered only by a large herd. Since the fodder capacity of pastures is small, it is necessary to set up ranches on a very large scale. The management problems of large-scale stock-keeping in the tropics and subtropics are therefore not dissimilar to those of plantations. High initial costs and the long-term commitment to one type of production make the early stages particularly difficult, and only if there are stable economic and social conditions is there any incentive to invest in undertakings that will show a market profit only after one or two decades. 9.4.4. Development paths of ranching In few areas of agriculture is large-scale production so superior to smallscale production as in the case of stock-keepirg in dry tropical regions. Not only the net return on labour and land, but also the gross return is far above what can be obtained in traditional semi-nomadic stock-keeping. The economies of scale are not mainly due to mechanization. Much more important are benefits arising from access to agricultural and veterinary progress, from proper stocking policies, which justify the high fixed costs of investment in improvements like watering places, and from high levels of management. Large-scale ranching is thus in a far better position to reap the benefits of technical progress that become available in grazing and animal husbandry. The general tendencies that technical progress in ranching seems to favour may be summarized as follows: (1) The main prospect for increasing the output of livestock products lies in improved management of existing breeds, in particular in better grassland management, rotational grazing, improved disease control, etc. Innovations are much more easily applied in large ranches than in partially or semi-nomadic herdings. (2) The improvement of the basic genetic types of local cattle is bound to be a long and complex process. However, upgrading is of increasing importance. Breeding of other types that mature earlier is the usual objective. Measured in terms of the quantity and quality of the production per hectare, per animal, and per worker, and of the degree of competitiveness on the market, the large ranch proves to be more advantageous in every respect. However, large-scale ranching is not apprcpriate in all conditions, being unsuitable in those most marginal that are occupied by the total nomads.
the main types of farming that CHAPTERS 3 to 9 have been conc.rned with features and characteristics ofeach have been
occur in the tropics. The salient to be drawn to the problems outlined and exemplified so as to allow attention organization. For this purand development paths of each pattern of farm each being significantly pose, seven distinct farming systems were identified, different from the others. to imply that very clear This method of treatment should not be taken classified under the various differences are to be found between farms that are that can be drawn between types. There are often no clear-cut divisions class. In some cases, holdings classed as one type and those coming in the next arable systems, permanent and as for example for shifting, semi-permanent, Major differR. parameter the the cut-off point is taken at given values of R values, high and middle, ences exist between farming systems with low, types. different basically these but there are wide areas of transition between of types between differences The treatment used here has allowed the principal oftropical types all that farming to be highlighted. It is not, however, suggested or other of these seven types. farming can be classified precisely into one letween particular Despite the frequent absence of clear distinctions of parts that have systems of farm operation that arise from the combination generalizations comprehensive different operational features, some fairly of farming can be made. types about the development paths of different that are reflected in any combinations The range and balance of enterprise of constraining factors. In a actual farming system is limited by a number have suggested that five main recent work, Duckham and Masefield (1970) and importance of different types of factor limit the possible occurrence farm system. types of crop and livestock found in any actual types of production are (1) Ecologicalfactors determine what different of existing climatic, soil, possible and which species can thrive in the light in which different species and biological conditions, including the way interact within a single system.
2F2
283
concerning such things as land tenure and the (2) Infra-structuralfeatures, like water, power, expertise and advice, inputs supply of various off-farm kind, further restrict the choice of this of material inputs, and things ecologically suitable enterprises. from arise possible combinations that can (3) External economic constraints will also profoundly influence the choice of attractive possibilities, through such features as location, the quality of of communications for the movement of produce, the organization and structure of marketing systems, the existence of alternative sources of food supply that can remove the need for subsistence production, the availability of credit, produce subsidies, and a whole host of other features, that effect the attractiveness of different systems of farming. (4) Internal operationalfactors like farm size, the available labour force, owned resources that can be invested, and fixed improvements are also obvious important determinants. (5) Finally personal acceptance of the adopted system can have some influence in those types of farming where a range of possibilities permits personal choices to be made. Within the framework of successive limitations of this kind, development paths can be observed for different major types of tropical ecological situation. These indicate the directions along which farmers may adapt their systems of operations as circumstances change to alter one or other of the constraining factors. Fig. 10.1 shows in diagrammatic form the evolutionary paths that may be followed in the four major climate zones of the tropics. Starting from the basic, undifferentiated system for a particular situation, each line in the diagram shows a possible evolutionary path, its relative importance and feasibility being indicated by the thickness of the arrow. In humid areas we find above all the change from shifting systems to irrigation farming, mainly with rice, and to the growing of perennial crops. These changes occur either directly from shifting systems or via the stage of semi-permanent cultivation. The pre-conditions for the spread of permanent rain-fed farming are usuall- not yet present, while ley systems for these areas are still in the experimental stage. In the humid savannas the change is mainly from shifting systems to scmi-permanent systems and-with increasing population densities--to irrigation farming. Unregulated ley systems are found in areas that are free from the tsetse fly. Regulated ley farming, which is widely practised in the subtropics, has been introduced into the humid savannas of the tropics only in exceptional cases. In the dry savanna climate the tendency is to change from shifting cultivation to unregulated ley farming and finally to permanent dry farming, with irrigation spreading wherever water isavailable. In grazing areas, the pattern of change is from open grazing to paddock systems. In high-altitude areas the evolutionary pattern is for the intensification of arable farming, developing rapidly from shifting cultivation
g I
2. Humid Savannas
At
stvanna Dry FSemi-permanent cultivtion 3.ifin
Shifting cultivation
t
"
de
"
te
farmng ;
3. Dry savannas
[ Perennial crops
- farming
en
oSemi
permanentof farming
p I F Perennialcp.
1F
/~
ofrmiihg systems. ''hew Fj. 0.. General tendencies in the development of tropical tendencies re valid for indigenous stolholders only. The establishment of large farms and estates is usually connected with the ad hoc adoption of an advanced system of farming.
via unregulated ley farming to permanent farming, with the planting of perennial crops wherever feasible. These tendencies are accompanied by a number of major changes: (1) from long-fallow to short-fallow systems;
(2) from short-fallow systems to permanent land use; (3) (4) from natural grazing to cultivated fodder;
285
(5) from rain-fed farming to irrigation farming; (6) from arable farming to the planting of perennial crops; (7) from single cropping to multiple cropping;
manuring and fertilizing; (9) from hoe cultivation to animal traction or tractors. Changes in adopted systems of farming, whether in the form of improvement, or intensification, or changes to another type of agricultural production, occur as the result of changes in some or all of the constraining features considered above. These change: may arise either within the farm or outside it, and may be caused by new techni-ai possibilities (new seed, pest control, irrigation, etc.) or be the result of a wide and complex range of social and economic factors. The foi'owing are amongst the most important operating in tropical areas today. (1) Population density, and the correlated subsistence-food demand, is a vitally important variable causing changes in farm organization. While there are certainly instances where population growth is mainly dependent on agricultural output, output is more often a function of population growth.' It should not be overlooked, however, that this increased production, with pre-machine techniques, usually requires increasing labour inputs per unit of output. (2) Technical progress has already drastically changed farming patterns in most tropical countries. Farming systems are likely to be increasingly influenced by innovations and there seems to be no end to them. In contrast with the situation in pre-technical systems, changes of this kind may allow increases in output to arise with decreasing units of input. (3) Tile development of urban purchasing power and export markets has proved to be a decisive variable in many places, because changes in price relations are very important to tropical fariners. (4) An important role is also played by changes in human aspirations. There are many firm indications that tropical smallholders change rapidly from target workers to modified target workers and finally to commercial Boserup (1965).
286
DEVELOPMENT OF FARM SYSTEMS farmers who try to maximize income, albeit within the limitation of other objectives, security in particular.
To promote economic development and to meet other objectives, governments in most tropical countries adopt a variety of policy measures designed to modify agricultural production. These changes will be reflected at the farm level in the systems of farming that are followed. Generally these measures are designed to assist or accelerate evolutionary developments of the kind outlined in Fig. 10.1. In some instances, official intervention, through Zhe government policy, is merely incidental in nature, as in the establishment of proper conditions for orderly marketing. which can lead to different levels of adjustment in production on individual farms. However, at the other extreme, government action can take the form of creation and control of all aspects of complete farming situations, as iii settlement schemes of different kinds, especially irrigation settlements. To consider such policy measures is beyond the scope of this book. Nevertheless, it is important to stress that changes in government policy are very important factors influencing the development and evaluation of farming systems. Every policy measure directed towards agriculture will be reflected in changes in farm production systems. It is hoped that this book, by its analysis of the characteristics, weaknesses, and possibilities of each of the major types of farming in tropical areas, will help to provide a better background to the appreciation of different farming situations and their adjustment through policy actions of different kinds.
Bibliography
ADAMANTIDIS, D. (1956). Monographie pastorale du Ruanda Urundi. Bull. agric.
Congo beige 47, 3585ff. AEREBOE, F. (1919). Allgemeine landwirtschaftliche Betriebslehre. Parey, Berlin. AGMON, R. (1966). Agricultural training and demonstration project of BoboDioulasso, Upper Volta, in relation to savanna development, pp. 100-8. Khartoum (Rome, 1967). AGRAWAL, G. D. (1957). Studies in the economics offarm management in Uttar Pradesh. Government of India Press, New Delhi. ALKALI, M. M. (1967). Mixed farming in northern Nigeria. Ministry of Agriculture Kaduna. ALLAN, W., GLUCKMAN, M., et al. (1948). Land holding and land usage among the plateau Tonga of Mzabuke District. Rhodes-Livingstone Papers, no. 14, Cape Town. (1949). Studies in African land usage in northern Rhodesia. Rhodes-Livingstone Papers, no. 15, Cape Town. (1967). The African husbandman. Oliver and Boyd, London. ALWAN, A. S. (1968). Socio-economic issues of nomads' settlement in the Western Desert of the U.A.R. Land Reform 1, 28-35. der ANDREAE, B. (1963). Die extensive Weidewirtschaft in den Tropengebieten Kontinente. Ber. Landw. 41, 11ff. (1964). Betriebsformen in der Landwirtschaft. Ulmer, Stuttgart. ( 1966). Weidewirtschaft im sudlichen Afrika. Geogr. Z. 15 (special issue). ANGLADETrE, A. (1966). Le riz. Maisonneuve and Larosse, Paris. ATTEMS, M. (1967). Bauernbetriebe in tropischen Hohenlagen Ostafrikas. (AfrikaStudien, no. 25). IFO Institut, Muinchen. Australia, Bureau of Agricultural Economics (1961). The coffee industry in PapuaNew Guinea (mimeo). Canberra.
BARKER,
R. and SOOTHIPAN, A. (1968). Economic efficiency and profitability of rice farming in the Philippines and Thailand. Fmn. Mgmt. Notes Asia FarEast 4, no. 1. BARRAU, J. (1961). Subsistence agriculture in Polynesia and Micronesia. Honolulu. In BArTLE-rT, H. H. (1962). Fire, primitive agriculture and grazing in the tropics. Man's role in changing the face ofearth, Chicago. pp. 692-720. BARTON, R. F. (1922). Ifugao economy. Am. Archeol. Ethnol. 15, no. 5. BASAK, K. C. and CHANDA, R. C. (1964). Studies in the economics offarm management in Sambalpur District, Orissa. Government of India Press, New Delhi.
BIBLIOGRAPHY P. T. (1948). The rubber industry. Longmans, Green, London. BAUM, E. (1968). Land use in the Kilombero Valley. In Smallholder farming and smallholder development in Tanzania (ed. H. Ruthenberg), pp. 23-50. (AfrikaStudien, no. 24). IFO Institut, Minchen. BECK, R. S. (1963). An economic study of coffee-banana farms in the central Machame area, 1961 (mimeo). Ministry of Agriculture, Dar-es-Salaam. BECKETT, W. H. (1944). Akokasa-A survey ofa Gold Coast village. Monograph on Social Anthropology, no. 10, London School of Economics and Political Science, London. BEINAERT, A. (1941). La technique culturelle sous l'iquateur. Bull. Inf. LN.E.A.C. 4, 12ff. BEUKERING, J. A. VAN (1952). Types of farming. Carrib. econ. Rev. 4, 140-63. BINNS, B. 0. (1955). Plantations and other centrally operated estates (Food and Agriculture Organization Studies no. 28) Rome. BLANCKENBURG, P. VON (1963). Rubber farming in Benin area, preliminary report (mimeo). Nigerian Institute of Social and Economic Research, Ibadan. (1964). Afrikanische Bauernwirtschaft auf dem Wege in eine moderne Landwirtschaft. Z. ausl. Landw., Sonderh. (special issue). BOSERUP, E. (1965). The conditions ofagriculturalgrowth. Aldine, London. BOUCHER, L. J. (1967). Surface irrigation (mimeo). Food and Agriculture Organization, Rome. BOUCHET, P. (1955). Le secteur expdrimental de modernisation agricole des terres neuves (Scn~gal). Agron. trop. 10, no. 2, 174-216. BOUTILLIER, J. L. (1960). Bongouanou, C6te d'lvoire. Paris. CANTRELLE, P., CAUSSE, J., LAURENT, C., and DOYE, H. (1962). La moyenne vallde du Snegal. Presse Universitaire de France. Paris. BoYE, P. (1968). Animal husbandry in palm plantations (mimeo). Agricultural Research Priorities for Economic Development In: Conference on in Africa, Abidjan 1968. BRADFIELD, R. (1960). Opportunities for soil scientists in freeing the world from hunger. Proc. 7th Int. Congr. Soil Sci., Madison. (1968). Farming systems in Africa. Conference on Agricultural Research Priorities for Economic Development in Africa, Abidjan 1968 (mimeo). BRArON, C. A. and ROBERTSON, L. S. (1954). Farming methods and returns on 126 rice farms in Roxa Isabela, in 1952. Philipp. Agric. 37. BRtMAUD, 0. and PAGOT, J. (1962). Grazing lands, nomadism and transhumance in the Sahel. Problems of the arid zones. UNESCO, Paris. BRJNKMANN, TH. (1922). Die Okonomik des landwirtschaftlichen Betriebes. In Grundriss der Sozialkonomik, chap. 7, Mohr, pp. 27-124. Tibingen. (1924). Bodennutzungssysteme. In Handw'rterbuch der Staatswissenschaften, vol. 1, Fischer, pp. 959-73. Jena. BROUWERS, M. (1965). Report to the government of Cameroon on land use and agricultural development of Cameroon. Food and Agriculture Organization, Rome. BROWN, D. W. (1960). A reconnaissance study of farming organization in the Kama Keng food production area (mimeo). University of Malaya, 1960.
BAUER,
288
BIBLIOGRAPHY
BROWN,
289
L. H. (1963). The development of the semi-arid areas of Kenya (mimeo). Ministry of Agriculture, Nairobi. BUCHANAN, K. M. and PUGH, J. C. (1958). Land andpeople in Nigeria. University of London Press, London. BUCK, J. L. (1930). Chinesefarm economy. Chicago University Press, Chicago. Bureau pour le d~veloppement de la production agricole (B.D.P.A.) (1967). Techniques rurales en Afrique. Ministire de la Cooperation, Paris. BURLE, L. (1961-2). Le Cacaoyer. Maisonneuve, Paris. BURNETT, J. R. (1954). Crop production. In Agriculture in the Sudan (ed. J. D. Tothill). Oxford University Press, London. BUTrERWORTH, K. (1964). Profit from beef (mimeo). Rhodesia National Farmers' Union, Salisbury and Lusaka. Buys, C. J. and M6LLER, C. A. (1961). Economic survey oftobaccofarming in certain irrigationregionsin Transvaal (Department of Agriculture, Bull. 373). Pretoria. CANTOR, L. M. (1967). A worldgeographyofirrigation.Oliver and Boyd, Edinburgh. CAPOT-REY, R. (1962). The present state of nomadism in the Sahara. In Problemsof the aridzones. UNESCO, Paris. CARRUTHERS, I. D. (1968). Irrigation development planning; aspects of Pakistan experience. Agricultural Development Studies, Wye College Report 2, chap. 6, London. Ceylon Department of Census and Statistics (1965). Statistical abstracts. Colombo. CHAMPION, J. (1963). Le bananier. Maisonneuve and Larosse, CHANG, CH. CH. (1963). Analysis of ricefarming methods in Taiwan. Taipei. CHANG, S. K. and LIN, P. C. (1967). Rice cultural practice in Taiwan (mimeo) Taichung District Agricultural Improvement Station, Taiwan. CHAPELLE, J. (1957). Nomades noires du Sahara. Librairie Plon, Paris. CHEN, CH. S. (1963). Taiwan, an economic andsocialgeography. Taipei. CHEVALIER, A. (1926). Les principaux systimes des cultures du globe. Rev. Bot. appl. Agric. Colon. 55, no. 55. Chia-nan Irrigation Association (h.d.). Standarddiagramof rotationirrigation. CLARK, C. and HASWELL, M. (1964). The economics of subsistence agriculture. McMillan, London. CLAYTON, E. (1963). Economicplanning inpeasantagriculture(Kenya). Wye College, Kent. CLEMM, M. VON (1964). Agricultural productivity in Mt. Kilimanjaro. Econ. Bot. 18, no. 2, 99-121. COLLINSON, M. P. (1962-5). Reports on survey at Bukumbi, Usmao and other locations of Sukumaland, Tanzania (mimeo). Ukiriguru. CoNIN, H. C. (1957). Hanundo agriculture. A report on an integral system of shiting cultivationon the Philippines. Food and Agriculture Organization, Rome. COoK, 0. F. (1921). Milpa agriculture, a primitive tropical system. In Smithsonian Reportfor 1919. Washington.
20-(12 pp.)
290
BIBLIOGRAPHY
COSTER, R. N. (1958). Peasant farming in the Petauke and Katete area of the Eastern Province of northern Rhodesia. Department of Agriculture, Bull. no. 15, Lusaka. COULIBALY, S. (1961). Les paysans Senoufo de Korhogo. Cah. d'outre-mer 14,27-59. COURTENAY, P. P. (1965). Plantation agriculture.Bell and Sons, London. 35, part 2, CUNNISON, 1. G. (1954). The Humr and their lard. Sudan Notes Rec. 50-68.
DANIEL. E. F. (1960). Farm planning and management. Government of India Press,
New Delhi. DAVIES, W. and SKIDMORE, C. L. (1966). Tropicalpastures. Faber and Faber,
London.
J. (1962). Le pasan Cambodgien. Joint Publication Research Service, Washington. DEQUIN. H. (1964). Die Landwirtschqft Saudisch-Arabiens und ihre Entwicklungsmiglichkeiten. DLG, Frankfurt. DION, H. G. (1950). Agriculture in the Altiplano of Bolivia. Food and Agriculture Organization, Rome. DITrMAR, K. (1954). Allgemeine Vdlkerkunde. Viehweg, Braunschweig. DRIVER, P. N. and DEASi, D. K. (1957). Stvdies in the economics offarm management in Bomba' State. Government of India Press, New Delhi. DUCKHAM, A. N. and MASEFIELD, G. B. (1970). Farming systems of the world. Chatto
DELVERT,
11, DUFOURNET, R. (1961). Riziculture de la r6gion Tananarive. Bull. Madagascar no. 180, 378-86. DUMONT, R. (1951). Etude de quelques 6conomies agraires au S6n6gal et au Casamance. Agron. trop. 6, 229ff.
-
(1966). Types of ruraleconomy. Methuen, London. DYSON, R. and HUDSON, N. (1969). Subsistence herding in Uganda. Scient. Am. 220, no. 2, 76-89. EDWARDS, D. (1961). An economic study ofsmall-scale fartring in Jamaica. University
Press, Glasgow.
ENGELBRECHT, H. (1930). Die Landbaczonen der !Ede. In H. Wagner Gedaichtnisschrift. Justus Perthes, Gotha. ENJALBERT, H. (1956). Paysans noirs: les Kabrds du Nord-Togo. Cah. d'outre-mer 9, no. 34, 137-80. in EVANS, L. J. C., PHILLIPS, T. A. and WADNEY, A. D. (1961). Cocoa plantations new areas--cost of establishment and profitability. In Cocoa growing. Cadbury Ltd., Bournville. of FALs-BORDA, 0. (1955). Peasant society in the Colombian Andes. University Florida Press, Gainsville. FARMER, B. H. (1957). Pioneer peasant colonisation in Ceylon. Oxford University Press, London, Toronto, and New York.
BIBLIOGRAPHY
FAUCHER,
291
D. (1949). Gdographieagraire.G6nin, Paris. FINNEGAN, R. H. (1965). Survey of the Limba people of northern Sierra Leone. London. FLOYD, B. N. (1960). Changing pattern of African land use in southern Rhodesia. Dissertation, Syracuse. Food and Agriculture Organization (1958, 1960). Coffee in Latin America. United Nations, New York. (1961). Ministry of Agriculture and Natural Resources, Western Secretariat, Ibadan (mimeo). (1963). The farm settlement scheme in the Western Region, Nigeria, appendix I, 2, no. 1720. Rome. (1965). Progressin land reform,fourth report. Rome. (1966). Coconut as part of a mixedfarming system. Rome. FORDE, D. C. and Scorr, R. (1966). The native economies of Nigeria (ed. M. F. Perham). Faber and Faber, London. FOREMAN, W. J. (1959). Farm resource productivity in west Uttar Pradesh. Indian J. Agric. Econ. 14, no. 4, 115-28. Fox, J. W. and CUMBERLAND, K. B. (1962). Western Samoa. Whitcombe and
Tombs, Christchurch, New Zealand.
FRANK, CH. FRICHOU,
House, Nairobi.
H. (1966). L'devage et le commerce du betail dans le nord du Cameroun. Cah. O.R.S.T.O.M. 3, no. 2. FREEMAN, J. D. (1955). lban agriculture. Stationary Office, London. FREMOND, Y. (1968). Coconut-palms selection. Conference on Agricultural Research Priorities for Economic Development in Africa, Abidjan (mimeo), report 8/10. FRENCH, M. H. (1967). Animal production and savanna areas. Report on Meeting on Savanna Development, Khartoum 25.10-6.1I.1966, pp. 153-6. Food and Agriculture Organization, Rome. FRIEDRICH, K. H. (1968). Coffee-banana holdings at Bukoba. Smallholder farming andsmallholderdevelopment in Tanzania (ed. H. Ruthenberg), pp. 175-212. (AfrikaStudien, no. 24). IFO Institut, MUnchen. FROELICH, J. C., ALEXANDRE, P. and CORNEVIN, R. (1963). Les populations du Nord-Togo. Presse Universitaire de France, Paris. M. (1956). Au Tchad les transformations subies par ragriculture traditionnelle sous l'influence de la culture cotonnires. Agron. Trop. 12, no. 5-6, 597-645. GALErr, R., BALDWIN, K. D. and DINA, J. 0. (1956). Nigerian cocoa-farmers. Oxford University Press, London. GEERTZ, C. (1963). Agricultural involution. The process of ecological change in Indonesia. University of Calif",nia Press, Los Angeles. GEORTAY, G. (1956). Donn6es de bie pour la gestion de paysannats de cultures vivrires en region 6quatoriale foretire. Bull. Inf. I.N.E.A.C. 5, no. 4.
GAIDE,
292
BIBLIOGRAPHY
Integrated Survey, Delft. tenant discipline on an irrigated GIOLIOL, E. G. (1965). Staff organization and 202-5. 3, no. 30, land settlement. E. Afr. agric. For.J. (1966). Cost of production of major crops GILL, CH. M. SH. and NAQvJ, A. A. West Pakistan, Lahore. 1961-2 to 1963-4. The Board of Economic Inquiry, on the farm settlement Nigeria of GILL, N. (1963). Report to the government Organization, Agricultural and Food (mimeo). scheme in the Western Region Rome. Barotse Plain. Rhodes (Livingstone GLUCKMAN, M. (1941). Economy of the Central papers no. 7). Livingstone. Editions d'art et d'histoire, GOUROU, P. (1936). Les paysans du delta Tonkinois. Paris. (1940). La terre et I'homme 'n Extrtne-Orient. Hartmann, Paris. the Far East, (1951a). Indochine, part II: Development of upland areas in vol. 1. Institlite of Pacific Relations, New York. Man's role in changing (1951b). The quality of land use of tropical cultivators. Chicago. 336-49. pp. the face of the earth, d'outre-mer 9, 36-57. (1956). Milieu local et colonisation r6unionaise. Cah. landscape. Impact Sci. (1964). Changes in civilization and their influence on Soc. 14, no. 1. (1966). The t, opical world. Longmans, London. and fertilizers in northern Nigeria. In GREENWOOD, M. (1949). Mixed farming on Tropical and SubtropicalSoils. Conference Proceedingsofthe FirstCommonwealth Harpenden. GRIsT, D. H. (1959). Rice. Longmans, Lonuon. and coconut-cattle schemes in the GROENEVELD, S. (1968). Traditional farming development in Tanzania (ed. Tanga region. Li Sr-allholderfarmingand smallholder Mllnchen. Irmtitut, IFO H. Ruthenberg). (Afrika-Studien, no. 24) conditions de modernisation d'un village GUILLARD, J. (1965). Golonpoui, analyse des du Nord-Cameroun. Mouton, Paris. survey of the sisal industry of Tanganyika. GuILLEBAUtI, C. W. (1958). An economic Nisbet, London. autochthone dans les savannes de GuiLLEmiN, R. (1956). Evolution de l'agriculture l'Oubangui. Agron. trop. 12, nos. 1, 2, 3. de ]a r6gion de Man, C6te d'Ivoire. Agron. GUINARD, A. (1961). Le syst~me cultural trop. 16, no. 2, 148-78. in Africa, part 2. Wld. (1967). Conservation and improvement of soil fertility Crops 19, no. 6, 29-31. Verhiiltnisse in den sildlichen GuNDERT, H. (1962). Die betriebswirtschaftlichen Hohenheim. Distrikten von Slidwestafrika. Dissertation,
--
J. H. (1964). Het economisch aspect der erosie. In Stroomgebiedsontwikkeling Tropische Landinrichting (mimeo). Wageningen.
HAAN,
BIBLIOGRAPHY
HADFIELD,
293
J. (1962). Report on the peasant fanning survey, 1960-1 (mimeo). Northern Rhodesia. HAHN, E. (1892). Die Wirtschaftsformen der Erde. Petermannsgeogr.Mit!. 38, 8ff. (1910). Die Brandwirtschaftin der Bodenicultur. Berlin. HANSEN, S. (1969). An outline of a rubber programme in Ceylon (mimeo). Rubber Research Institute of Ceylon, Dartonfield. HARLER, C. R. (1966). Eight thousand pounds of tea from one acre. Investors' Guardian 208, no. 6450, 784ff. HARRISON, J. E. (n.d.). Dairy management survey (mimeo). Rhodesia National Farmers' Union, Salisbury and Lusaka. HASWELL, M. R. (1953). Economics of agriculture in a savanna village (Gambia). Colon. Res. Stud. no. S. (1963). The changing pattern of economic activity in a Gambian village (mimeo). London. HAVILAND, W. E. (1952). Tobacco farm organisation, costs and land use in Southern Rhodesia. Rhodesia agric. J. 49. (1954). The economic development of the tobacco industry of Northern Rhodesia. S. Afr. J. Econ. 22, no. 4. HEADY, E. 0. and DILLON, J. L. (1961). Agriculturalproductionfunctions. Iowa State University Press, Ames. HERSKovrrs, M. J. (1964). The economic life of primitive people. Routledge, New York. -(1965). The humanfactor in changing Africa. Alfred A. Knopf, New York. HEYER, J. U. (1965). An investigation of the limits of peasant agricultural production (mimeo). University College of Nairobi. (1966). Agricultural development and peasant farming in Kenya. Dissertation University of London. HICKLING, C. F. (1961). Tropical inlandfisheries. Longmans, London. HU.L, P. (1963). The migrant cocoafarmer of Southern Ghana. University Press, Cambridge. Ho, R. (1964). Mixed farming and multiple crop:ng in Malaya. In Land use and mineral deposits in Hong Kong (ed. S. G. Davis). Hong Kong. (1967). Farmersof CentralMalaya. Australian National University, Canberra. Ho, TONG PENG (1964). La plantation d'6tat Preah Sihanouk a Ratanakiri (Cambodge). Revue gdn. Caoutch. 41, no. 9. Hoy, D. R. (1961). Agriculturallanduse of Guadaloupe.Natural Resource Council, Washington. HSIEH, S. C. (1966). The effect of improved cropping systems on farm earning capacity in Taiwan. Philipp. Agric. 49, no. 9, 787-98. HUPPERTZ, J. (1951). Viehhaltung und Stallwirtschaft bei den einheimischen Ackerkulturen in Afrika und Asien. Erdkunde 5, no. 1, 36-51. I.N.E.A.C. (Institut National pour l'tude Agronomique du Congo Belge) (1958). Normes de inain-d'oeuvrepour les travaux agricolesau Congo Beige.
294
BIBLIOGRAPHY
Iso, E. (1964). Rice and crops in its rotation in subtropical zones. Japan Food and Agriculture Organization Association, Tokyo. IZKOWITZ, K. G. (1951). Hill peasant in Indochina. G6teborgs etnografiska mus., Goteburg. K. 0. (1957). A study of the economics of a rice growing village in Central Thailand. Dissertation, Ann Arbor, Michigan. Japan Ministry of Agriculture (1960). The economics of/farm management survey, 1960. (Japanese.) JATZOLD, R. and BAUM, E. (1968). The Kilombero Valley (Afrika-Studien no. 28). IFO Institut, Mijinchen. JOHNSTON, B. E. (1963). The staple food economics of western tropical Africa. Stanford University Press, Stanford. JONES, W. 0. (1959). Manioc in Africa. Stanford University Press, Stanford. (1968). Plantation. Int. Encycl. soc. Sci. 2, 154-9. JOOSTEN, J. H. L. (1962). Wirtschaftliche und agrarpolitische Aspekte tropischer Landbausysteme (mimeo). Institut fuer landwirtschaftliche Betriebslehre, G6ttingen. JORDAN, H. D. (1967). Rice production. In Conference on Agriculture Research Priorities foi Economic Development in Africa, Abidjan 1967 (mimco). JURION, F. and HENRY, J. (1967). De l'agriculture itin~rante h. I'agriculture intensifi~e. I.N.E.A.C. hors s.rie. Bruxelles.
JAMLEKHA,
B. (1960). Economies et socit srurales dans les rdgions tropicales, books I and 2. Centre de documentation universitaire, Paris. Kenya Coffee Growers Association (1968). Coffee estate expenses (mimeo). Nairobi. KNAPP, R. (1956). Weide-Wirtschaft in Trockengebieten. In Giessener Beitri'ge zur Entwicklungsforschung, vol. 1. Fischer, Stuttgart. KROON, A. H. J. DE (1967). Characteristics and development tendencies of permanent arable farming on unirrigated land in the humid tropics with particular emphasis on the uplands in South-east Asia (conference paper). Institut for auslandische Landwirtschaft, Stuttgart-Hohenheim. KuLs, W. (1963). Bevidlkerung, Siedlung und Landwirtschaft im Hochiand von Godjam, Nordiithiopien. Frankfurt. KULTHONGKHAM, S. and ONG, SHAO-ER (1964). Rice economy of Thailand. Bangkok. KUNG, P. (1962). Irrigation farming and multiple cropping in the Ganges-Kobadak project area (Food and Agriculture Organization report 1456). Rome. 1962.
KAYSER,
dans ]a zone arachidire, Hanene-S n6gal (mimeo). Institut de Science Economique Applique, Dakar. LAWSON, R. M. (1966). Change and growth in traditional agriculture in the Lower Volta. In: Studies in rural economic growth (mimeo). University of Ghana. LE, S. C. (1967). An economic study of land use in Taoyuan and Shimen reservoir area, 1964. Chung Hsing University, Taichung.
BIBLIOGRAPHY 295 (1967b). An economic study of land use, Tainan Haien and City, 1966. Chung Hsing University, Taichung. LEMBEZAT, H. (1961). Les populations du Nord-Cameroun et de l'Adamoua. Presse Universitaire de France, Paris.
L.StNK, L. L. (1967). The system of dry-farming in the West Nimar District of
Central India. In Jahrbuch des Siidasien-Institutsder Uniersitit,Heidelberg 1966. Harrassowitz, Wiesbaden. LIEBIG, J. VON (1878). Chemische Briefe. 6th edn., Winter'sche Verlagshandlung, Leipzig and Heidelberg. LiM, Y. (1965). Export industries and pattern of economic growth in Ceylon (microfilm). Dissertation, Michigan. (1968). Impact of the tea industry on the growth of the Ceylonese economy (mimeo). University of Hawaii. LrrrLE, K. L. (1951). Mende rice farm and its costs (Sierra Leone). Routledge, Bruxelles.
LocK, G. W. (1962). Sisal. Longmans, London.
LORD, R. F. (1963). Economic aspects of mechanizedfarming at Nachingwea in the Southern Province of Tanganvika. London. LOWNDs, A. G. (1956). The sugar industry of Ffi. Angus and Robertson, Sidney. LUCAS, P. (1965). Premiers r~sultats d'enqu~te obtenus dans le village pilote de Laptinkaha, C6te d'Ivoire. Agron. trop. 20, no. 2, 147-219. LUDWIG, D. H. (1967). Ukara-EinSonderfall tropischerBodennutzunff im Raum des Victoria-Sees (Afrika-Studien, no. 22). IFO Institut, Miinchen. LUNING, H. A. (1961). An agro-economic survey in Katsina Province (Nigeria). Government Printer, Kaduna. (1969). The economic transformationoffanilv rice-farming in Surinam (Agricultural Research Reports no. 718). Wageningen.
MACARTHUR, J. D. (1964). Some economic studies of African farms in Rift Valley
Province, Nandi District, Elgeyo, and West Pakot District (mimeo). Nakuru, Kenya. (1966). Some economic aspects of agricultural development in Nyeri District, 1963 (mimeo). Nakutu. (1969). The economics of perennial crops (mimeo). Institut fuer auslaendische Landwirtschaft, Berlin. - and ENGLAND, W. T. (1963). A reporton an economic survey offarming in the Molo-Mau-Narok area, 1962/63 (mimeo). Nakuru, Kenya. MCWILLIAM, M. D. (1957). The African tea. industry (mimeo). Nuffield College, Oxford. (1959). The Kenya tea industry. E. Aft. Econ. Rev. no. 6. MALCOLM, D. W. (1953). Sukumaland, an African people and their coutry. Oxford University Press, London, New York, and Toronto. MANSHARD, W. (1961). Die geographische Grundlage der Wirtschaft Ghanas. Steiner, Wiesbaden. (1968). Einfihrung in die Agrargeographie der Tropen. Bibliographisches Institut, Mannheim.
296
MARTIN,
BIBLIOGRAPHY
A. (1956). The oil-palm economy of the Ibiblofarmer. Ibadan University Press, Ibadan. MASSFL, B. F. and JOHNSON, R. W. M. (1968). Economics of smallholder Rhodesia. Studies in agriculturaldevelopment, vol. 8, suppl. Food, 1968.farming in Research Institute, Stanford University. MATrHUR, P. N. (1963). Studies in the economics offarm management in Madhya Pradesh. Government of India Press, Nev., Delhi. Mauritius commission of enquiry (1962). The sugar industry (mimeo). London. MAY, ST. and PLAZA, G. (1958). The United Fruit Company of Latin America. National Planning Association, Washington. MEILLASSOUX, C. (1964). Anthropologie deonomique des Gouro de C6te d'Ivoire. Mouton, Paris. MILLER, J. and MAKANNEN, T. (1965). Organisationandoperation of threeEthiopian casefarms. Imperial Ethiopian College, Dire Dawa. MILLIKEN, M. (1968). Dairying in Mauritius. Revue agric. sucr. Ile Maurice, supple. 47, no, 2. MIRACLE, M. P. (1964). Traditional agricultural methods in the Congo Basin (mimeo). Forc Research Institute, Stanford University. -(1967). Agriculture in the Congo Basin. University of Wisconsin Press, Madison, Milwaukee, and London. MOBEIG, P. (1952). Pionierset planteurs de Sao Paulo. Paris. MOERMAN, M. (1969). Agricultural change and peasant choice in a Thai village. University of California Press, Berkeley and Los Angeles. MOHR, B. (1969). Reiskultur in Westafrika (Afrika-Studien, no. 44). IFO Institut, Miinchen. MOREL, R. and QUANTIN, P. (1964). Les jach6res et ]a r6gin6ration du sol en climat Soudano-Guin6en d'Afrique Central. Agron. trop. no. 2, 105-36. MORGAN-REES, A. M. and HOWARD, R. H. (1955). An econom'c survey of African farming among the Sala of the Mumbwa Districtof Northern Rhodesia. Lusaka. MULLINS, L. (1960). Production practices, costs and returns for major enterprises on rice farms in the delta of Mississipi. Bull. Miss. St. Univ. no. 595. MtrUtKU, I. K. (1966). The economics of coffee production on smallholdings. Kenya Co. Lecture at Coffee Extension Seminar, Sept. 1966, Coffee Research Station, Ruiru. National Council of Applied Economic Research (1961). Economically weak tea gardensin north-east India. Government of India Press, New Delhi. NEWTON, K. (1960). Shifting cultivation and crop rotations in the tropics. Papua New Guin. agric.J. 13, no. 3, 81-118. NG KAY FoNG, TAN CHEE LIAN, and WIKKRAMATLEKE, R. (1966). Three farmers of Singapore. ,'acf. Viewpoint 7, no. 2, 169ff. NICOLAISEN, J. (1963). Ecology andculture of thepastoralTuareg. National Museum of Copenhagen.
NETTING, R. McC. (1968). Hillfarmers of Nigeria. Seattle and London.
BIBLIOGRAPHY
297
Nigeria, Northern Region Department of Agriculture (1955). Farming systems in Kano Province. Sols aft. 3, no. 4, 519-23. NIXDORF, 0. (1968). Arbeitsaufwand der Bewirtschaftung einer Kaffeepflanzung in Nord-Parand, Brasilien. Unpublished. NWAK, E. (1954). Land und Volk der Konso. Geogr. Abh. no. 14. NYE, P. H. and GREENLAND, D. J. (1961). The soil under shifting cultivation. Commonwealth Agricultural Bureau, Reading.
OLSEN,
K. H. (1963). Die geographische Bedingtheit agrarischer Wirtschaftsformen. In: Z. ausi. Landw., 2, no. 1, 1-17. OTREMBA, E. (1953). Ailgemeine Agrar- und Industriegeographie.Frankh, Stuttgart. OWEN, R. VON (1969). Produktionsstruktur und Entwicklungsmibglichkeiten der Rindfleischerzeugung in SUdamerika. Dissertation, Gdttingen University. OZAKI, C. (1960). Farmhousehold economic survey in Japan(Agricultural Development Series 13). Tokyo. OZBILEN, R. (1962-3). An agricultural survey of Phase I of Kushtia Unit, GangesKobadak Project (mimeo). Food and Agriculture Organization, Rome.
PAAuw, D. S. (ed.) (1962). Prospectsfor East Sumatran plantation industries: a
symposium. Yale University. PAWmS, A. VAN (1957). Rotations de plantes vivriires dans la r6gion de Nioka (Haute-Ituri). Landbouwtqdschr. Belg- Kongo 48, no. 6, 1515-44. PARSONS, D. J. (1960). The system of agriculture practised in Uganda (mimeo). Kawanda Research Station, Kampala. PAUWELS, M. F. (1960). La'dhuishoudkundig onde onder'oek bij de Jupaliri. Rijksuniversiteit, Gent. PELISSIER, P. (1966). Les paysans du Sdnegal. Imprimerie Fabregue, Saint-Yrieux. PELZER, K. J. (1948). Pioneersettlement in the Asiatic tropics. American Geographic Society, New York. (1957). Agriculture in the humid tropics. History and outlook of plantation agriculturein south-east Asia and its outlook. Pacific Science Conference, Bangkok. PER AM, M. F. (ed.) (1966). The native economies ofNigeria. University of London Press, London. PETU, D. A. and UPTON, M. (1964). An economic study of farming in two villages in llorin Emirate, Nigeria (mimeo). PHILLIPS, J. (1959). Agriculture and ecology in Africa. Faber and Faber, London. (1961). The development of agriculture and forestry in the tropics. Faber and Faber, London. (1964). Shifting cultivation. In The ecology of man in the tropicalenvironment (1.U.C.N. no. 4). Morges. PIGrGoT, C. J. (1964). Coconutgrowing. Oxford University Press, London. PINTEN, K. (1954). De Inlands Landbouwbedrijven in Belgisch-Kongo en RuandaUrundi, vols. I and 2. Ministerie van Kolonien, Bruxelles.
298
BIBLIOGRAPHY
Planning Commission (1962). Study ofthie problems ofminor irrigation. Government of India Press, New Delhi. PORTtREs, R. (1952). Atninagement de lYconomie agricole et rurale au Sindgal. Bambey. POSSINGER. H. (1967). Plantagenbetriebe oder Bauernbetriebe am BeispitidesSisals in Ostafrika (Afrika-Studien no. 13). IFO institut, Mtinchen. (1968). Landwirtschaftliche Entiicklu g in Angola and Mozambique (AfrikaStudien, no. 31). IFO Institut, Miinchen. Presidents Science Advisory Committee Report (1967). Worldfood problems. The White House, Washington.. PURSEGLOVE, J. W. (1946). Land use in the overpopulated areas of Kabale, Kigezi District, Uganda. E. Afr. agric. J. 12, no. 1, 3-10. REES, A. M. (1958). An economic survey of Plateau Tonga improved farmers. Agric. Bull Dep. Agric. North. Rhod. no. 14. and HOWARD, R. H. (1955). An economic survey of commercial farming among the Sala of the Mumbwa district of Northern Rhodesia. Agric. Bull. Dep. Agric. North. Rhod. no. 10. Rhodesia Ministry of Agriculture (1963). Cost of production offlue-cured Virginia tobacco. Salisbury. (1964). Beet cattleproduction costs. Salisbury. RICHARDS, A. I. (1961). Land, labour and diet in Northern Rhodesia. Oxford University Press, London. ROBERTSON, L. S. and TABLANTE, N. B. (1954). Invc.stment and income on 132 coconut farms in Indary, Cavite. in 1952. Philipp. Agric. 38. ROTENHAN, D. \'ON (1966). Bodennutzut. und Viehhaltung im Sukumaland, Tansania (Afrika-Studien, no. 11). IFO Institut, Miinchen. Cotton farming in Sukumalarid. In Smallholderfarmingand snallholder -(1968). development in Tanzania (ed. H. Ruthenberg), pp. 51-86. (Afrika-Studien no. 24). IFO Institut, Mik.hen. ROUNCE. N. V. (1949). The agriculture of the cultivation steppe of the Lake, Western and Central Provinces of Tangany'ika. Dept. of Agric. Tanganyika, Longmans, Salisbury, 1946. RUTHENBERG, H. (1963) Ansatze und Hindernisse der weiteren landwirtschaftlichen Entwicklung Madagaskars. Z. ausl. Landw., Sonderh. 4, no. 1, 18-59. (1964). Agriculturaldevelopment in Tanganyika (Afrika-Studien no. 2). IFO Institut, Berlin. (1965). Probleme des Uberganges vom Wanderfeldbau und semipermanenten Feldbau zum permanenten Regenfeldbau in Afrika sildlich der Sahara. Agrarvirtschaft 14, no. 1, 25-32. (1966). Agricultural production dejelopment policy in Kenya (Afrika-Studien, no. 10). IFO Institut, Berlin. (1967). Organisationsformen der Bodennutzung und Viehhaltung in den Tropen und Subtropen, dargestellt an ausgewaihlten Beispielen. In Handbuch der Landwirtschaft und Ernihrung in den Entwicklungslindern (ed. P. von Blanckenburg and H.-D. Cremer). Ulmer, Stuttgart.
299 BIBLIOGRAPHY in Tanzania (ed.) (1968). Smallholder farming and smallholder development Milnchcn. Institut, (Afrika-Studien, no. 24). iFO acreage,new methodology RtrrrAN, V. W. (1965). The economic demandfor irrigation Press, Baltimore. Hopkins John 1954-1980. andsnme preliminaryprojections in rice growing in the Changes (1966). C. E. VENEGAS, and SOOTHIPAN, A., 18-33. 1, no. 18, Crops Wid. Philippines and Thailand. Europ6enne. Bull. P. (1950). Agriculture iquatoriale Bantou et agriculture 579-662. agric. Congo Beige 41, no. 3, pays Mossi (Haute-Volta). SATEC: Une operation de d6veloppement rural en 579-97. Agron. trop. 19, no. 7, Afrique tropicale. Centre de DocuSAUtRER, C. (1968). Les structures agrairesen Sorbonne. mentation Universitaire, perfection6e pratiqu6 par les BwabaSAVONNET, C. (1961). Un systime de culture Etud voltaiq. no. 1, 19-52. (Haute-Volta) H1ound6 de Bobo-Ould de la r6gion DLG, Frankfurt. SCHAFER, H. (1963). Tierzucht und Tierhaltungin Steppengebieten. der Erde. Trockengebieten den in SCHICKELE, R. (1931). Die Weidewirtschaft Probleme Weltwirt. 53. des Viehs bei den Nomaden SCHINKEL, H. G. (1970). Haltung, Zucht, und Pflege Berlin. Verlag, Akademie Ost- und Nordostafrikas. Routledge and Kegan, London. SCHLIPPE, P. DE (1956). Shifting cultivation in Africa. (1957). Enqu~te priliminaire du syst~me agricole des Barundi de la r6gion Bututsi. Bull. agric. Congo Beige 48, no. 4, 827-82. et Social) S.E.D.E.S. (Soci&6t d'Etudes pour le Developpement Econornique Paris. Korhogo, de Region (1965). d'Ivoire Ripublique de C6te rigionale du cercle S.E.R.E.S.A., Rdpublique du Dahomey (1959-60): itude (mimeo). d'Allada. Cotonou in the Kushita Unit, SHAO-ER ONG (n.d.) Farming systems and suggested changes no. 55/12/8827. report Organization, Agriculture and Food (mimeo). Pakistan SHAW, D. J. (1965). Managil--eine Erweiterung des Gezira-Projcktes-Beispiel 4, no. 2, eines Bewasserungsprojektes im Sudan. Z. ausl. Landw. Materialsamml. 88-128 Forage resources of the Sudan SHEPHERD, W. 0. and BA'ASHER, M. M. (1967). on Savanna Development, Meeting Organization Agriculture and savanna. Food Khartoum 1966. Institut, SICK, W. D. (1969). Wirtschaftsgeographie von Ecuador. Geographisches Stuttgart. Development of agriculture SIMPSON, I. G., YASSIN, M. M., and DISHONI, S. (1966). of Agriculture, Department area. Gezira main the in and organisationalplanning Khartoum.
SAMUEL,
SING,
of India Press, New economics offarm management in Punjab, 1954-55. Government Delhi. of Agriculture, Accra. SMrrH, R. (1957). A report on Kusasi agriculture.Department du Palmier A 1'Exploitation et S.O.D.E.P.A.L.M. (Sociit6 pour le Dveloppement Abidjan. (manuscript). palmier de ha 60000 base, de 1'Huile) (1965). Itude
ATHWAL,
S. S. and
RANDHAWA,
BIBLIOGRAPHY SoRwLDOr, R. (1968). Rational water utilization. In: Conference on Agricultural Research Priorities for Economic Development in Africa, Abidjan 1968 (mimeo). SPENCER, J. E. (1966). Shifting cultivation in south-eastern Asia. University of California Press, Berkeley and Los Angeles. STANER, P. (1955), Les paysannats indig~nes du Congo Beige et du Ruanda-Urundi. Bull. agr. Congo Beige 46, no. 3, 468-558. SToFFrLs, E. H. J. (1949). Les systimes de cultures et la fertilit6 Bull. agr. Congo Beige 40, no. 2, 1564ff. S'RENGE H. VON (1961). Wild coffee in Kaffa province of Ethiopia. Trop. agric.38, no. 4, 297--301. SULTAN, M. (1954). Survey of milk production in the Jamaicadevelopment circle in 1952-53. Delite Press, New Delhi. Taiwan Department of Agriculture (1966). Recordoffarm record-keepingfamilies in Taiwan, 1965. Taipei. TAKE, CH. A. P. (1963). Socio-economic factors aftcting the productivity of agriculture in Okigwi Division, Eastern Region (mimeo). Nigerian Institute of Social and Economic Research, Ibadan. Tanzania Sisal (1969). Mon. Bull. agric. Econ. Statis. 18, no. 5, pp. 1-8. TAYLOR, W. CH (1956). The Firestone operations in Liberia. National Planning Association, Washington. TAX, S. (1963). Penny capitalism. A Guatemalan Indian economy. University of Chicago Press, Chicago. TEMPANY, H. and GRIST, H. D. (1958). Tropicalagriculture.Longmans, London. TERRA, G. J. A. (1957). Landbouwstelsels en bedrijsstelsels in de tropen. Landbouwk. Tqjdschr., 's-Grar. 6, 430-8. (1958). Farm systems in south-east Asia. Neth. J. agric.Sci. 6, no. 3, 157-82. Thailand Ministry of Agriculture (1959). Report on economic survey ofricefarmersin Nakom Phatom Province, 1955156. Bangkok. THORNTON, D. and ROUNCE, N. V. (1963). Ukara island and the agricultural practices of the Wakara. Tanganyika Notes Rec. no. 1, 25ff. THURNWALD, R. (1965). Economics in primitive communities. Oxford University Press. TIMOSl-ENKO, V. P. and SWERLING, B. C. (1957). The world's sugar. Stanford University Press. TONDEUR, M. G. (1956). L'agriculture nomade au Congo Beige. Food and Agriculture Organization. Agric. Nomade 1, 15-108. Tos, J. A. (1966). Economics and the natural environment (mimeo). Tropical Science Centre, San Josd. Tou'E, R., GAUDEFROY-DEMOMBYNES, P. and FAucIi, J. (1954). Perfectionnement des techniques culturales au Sdn6gal. Ans Centre Rich. agron. Bambey Sdngal. TRAPNELL, C. G. and CoETHEa, J. N. (1957). The soils, vegetations andagricultural systems of north-western Rhodesia. Government Printer, Lusaka. TRIPPNER, J. (1955). Ackerdlingung in der Provinz Ching-hai, China. Anthropos, 50.
300
BIBLIOGRAPHY
TuLIPPE, 0.
101
and WiuTwr, J. (1964). G6ographie do l'agriculture en Afrique Centrale. Essai de synthese. Bull. Soc. beige gtud. giogr. 33, no. 1. UpToN, M. (1966). Tree crops: a long-term investment. J. agric. Econ. 17, no. 1, 82-90. (1967). Agriculture in south western Nigeria. Univ. Reading Dev. Stud. no. 3. URQUARDT, D. H. (1961). Cocoa. Longmans, London. J. D. N. (1953-4). The Gezira scheme in the Sudan and the Russian Kolkhoz: a comparison of two experiments. Econ. Dev. cult. Change, 2, no. 1,32-59, no. 2, 120-35, no. 3, 216-35. Venmann VINK, G. J. (1951). De grondslagen van het indonesische Landbouwbedriff. and Zonen, Wageningen.
VESLUYS, WAGNER,
G (1956). The Bantu of North Kavirondo. Vol. 2. Oxford University Press, London. Hirth, Breslau. WAmL, L. (1933). Probleme der Landwirtschaftsgeographie. the tropics. Longmans, in Agriculture (1967). WEBST R, C. C. and WILSON, P. N. London. West Bengal Government (1968). Study in farm management and costsofproduction of crops, vol. 5. Calcutta. and yield WEsTGARTH, D. R. and NARAYANA, R. (1964). The effect of rubber price per acre on estate production costs. J. Rubb. Res. Inst. Malaya 18. Manchester. WHITE, C. M. N. (1959). A preliminarysurvey of Luvale ruraleconomy. Depart(mimeo). Kenya in ranching beef WHITE, H. D. (1963). Economics of Nakuru. ment of Agriculture. WHITE, ST. (1941). Agricultural economy of the hill pagans of Dikwa Emirate, Camerouns. Emp. J. exp. agric. 9, no. 3. Faber, WHmTE, R. 0. (1967). Milk production in developing countries. Faber and London. Press. WICKIZER, V. D. (1951). Coffee, tea and cocoa. Stanford University in tropical development WILDE, J. C. DE (1967). Experiences with agricultural Baltimore. Press, Hopkins Johns 2. Africa, vols. 1and ausl.Landw., 2, no. 3, WILDE, L. DE (1963). Ober die Ackerkultur der Tutshokwe. Z. 112-16. use in WILLS, J. B. (1962). The general pattern of land use. in Agriculture and land York. New Acera, London, Press, University Oxford 201-25. pp. Ghana, Katanga. Mem. WILMET, J. (1963). Syst~mes agraires et techniques agricoles au Acad. R. Scl d'outre-mer 14, no. 5. of WILSON, T. B. (1958). The West Johore coconut production survey (Department Agriculture, Bull. no. 104). Kuala Lumpur. Social Research, WINTER, E. H. (1955). Bwamba economy. East African Institute of Kampala. Leipzig. WrrrFoGEL, K. (1931). Wirtschaft und Gesellschaft Chinas. Hirschfeld, Sozialwiss. 6, WOERMANN, E. (1959). Landwirtschaftliche Betriebssysteme. Handw. 477ff.
21
BIBLIOGRAPHY F. (1892). Die natiirlichen Faktoren der tropischen Agrikultur. Dunker and Humblot, Leipzig. WoNG, C. T. (1968). Enlarging the size of farm business through intensive land use in Hong Kong. Fm. Mgmt. Notes Asia FarEast 4, no. 2. WOOD, G. A. R. (1963). Costs of cacao production. Cocoagrowing Cadbury Ltd. pp. 36-50. W6RZ J. G. G. (1966). Genossenschaftliche und partnerschaftlicke Produktionsforderung in der sudanischenLandwirtschaft. DLG, Frankfurt. WRIGHT, S. J. (1953). Report on the possibility of mechanising tea cultivation in Ceylon. Tea. 24, 59ff.
WOHLTMANN,
302
YouNG, J. E. DE (1958). Village lfe in modern Thailand. University of California Press, Berkeley and Los Angeles. Yu-cHu Hsu (1962). A case study of afarm at Ching-Shui, Taichung, Taiwan. The Republic of China, 1960. The Council on Economic and Cultural Affairs, Inc., New York. YUDELMAN, M. (1964). Africans on the land. Economic problems of African agriculturaldevelopment in Southern, Centraland East Africa, with special reference to Southern Rhodesia. Harvard University Press, Cambridge, Massachusetts. C. W. B. (1957). Studies in the economics offarm management in Madras (1954-55). Government of India Press, New Delhi. ZEVEN, A. C. (1967). The semi-wild oil-palm and its industry in Africa. (Agricultural Research Report 689). Wageningen.
ZACHARIAS,
Index
Adamawa, 103 Aereboe, F., 2 Africa, 109, 115, 166, 167, 177, 181, 182, 220, 268 Central, 19, 25 East, 48, 60, 62, 85, 100, 127, 130, 137, 194,212,220,222,230,242,252 260 262, 271, 277, 281 South, 207, 209, 252, 271 South-west, 264, 271, 274, 276 south of the Sahara, 60, 82, 142 West,2, 18,36,48, 59,60,85, 128,137, 190, 193,195, 215.222, 232,240, 260,262 Agrawal, G. D., 158, 160 Alade, 233 Algeria, 256 Alkali, M. M., 96, 127 Allada, Cercle d', 239, 241 Allan, W., 19, 25, 35, 66, 260 alpacas, 59 Alwan, A. S., 257 Amazon Basin, 137 America, Central, 22, 85, 195, 206, 214 Latin, 3, 36, 47, 54, 59, 113, 130. 131, 138, 193, 205, 214, 215, 220, 222, 225,226, 252. 271, 277 South, 85, 195, 254 Andes, 86, 100 Andreae, B., 171, 273 Andrha Pradesh, 157, 158 Angladette, A., 152 Angola, 55, 56, 66, 85, 193 animal, density, 253, 262, 264, 274 draught, 66, 70, 97, 110, 113, 122, 123, 157, 162, 163, 188 economy, 88, 262 husbandry, 36, 67, 85, 88,96, 97, 110, 127, 154, 164, 188, 224 mount, 254 pack, 254 power, 133, 155 traction, III, 113, 114 Ankole, cattle, 262 ranching scheme, 268 Apure, 279 303 Arabia, 133, 254 Arabs, 44 Arcot District, South, 152 Argentina. 271, 274, 277 Asia. 101, 167, 195 South, 101 South-east, 130, 131, 136, 138, 142, 163. 166, 182 Assam, 17, 228, 230 Australia, 86, 88, 207, 210, 252, 271 274 Northern, 166 avocado,228 Bahima, 262,263 Bahaya,215,218 Baku,84 Bailundu, 55, 56 Baldwin. K. D., 233,234 bamboo, 46 Barnum, 103 Bana, 103 banana, 42, 43, 61, 105, 131, 161, 163, 189, 192, 197, 198, 208, 214, 215, 217, 223, 227, 233, 243, 247 dwarf, 213 ensete, 213 groves, 117, 131, 206 Barbados, 207 Bari, 103 IBaringo District, 265 Barker, R., 158 barley, ('), 93, 138 Barrau, J., 202 Batta, 103 Bauchi, 102 Baule, 102 Baum, E., 20, 24, 39, 40, 57, 58, 74 beans, 100, 105, 107, 117, 121, 143, 190, 191, 192, 195, 215, 224, 225, 234 soya, 146 spring, 165 Beck, R. S., 223, 224 beef production, 194, 271, 276, 278 Beinaert, A., 33 Belgian Colonial Government, 49, 53 Bemba, 32, 35, 43
304
Bengal, 169 East, 169, 172, 173 West, 169, 171, 173 Bengawan, 187 Berron, 102 berseem, 162, 163 bilharzia, 141 Binafo, 187 Blanckenburg, P.v., 196, 205
I NDEX
catena; 60, 62, 105, 106, 144 cattle, 59, 67, 80, 85, 91, 95, 112, 118, 212, 221, 224, 225, 231, .260, 271, 275, 277 dairy, 85, 129 economy, 67, 68, 94, 218 grade, 88, 95, 250 grazing, 55, 91, 260, 271 hoarding, 110, 261
Bobo, 102
Bolivia, 59 Bombay, 116, 160, 161, 162 Deccan, 117, 121, 123 Boran cattle, 277 Borneo, 37, 40 Boserup, E., 115, 285 Boucher, L. J., 137 Bradfield, R., 9, 147, 187 Brazil, 36, 54, 86, 193, 194, 195, 198, 207, 210, 212, 221, 223, 225, 226, 227, 228, 232, 278 southern, 189, 196, 198, 205, 220, 224, 225 bride-price, 262 Brinkmann, Th., 2, 55 Brouwers, M., 35, 52 Brown, L. H., 255, 264, 265 Buck, J. L., 153, 154, 155 Budde, L., 160 buffalo, 154, 161, 175 Bukoba, 198, 215, 216, 217, 219 Bureau pour le Dbveloppement de ia Production agricole (B.D.P.A.), 71,156 Burma, 48, 173 burning, 22, 47, 260, 277 Burundi, 59, 103 bush control, 270 bush fire, 260 regeneration, 67 cabbage, 107, 138, 165, 175 cacao,190, 192, 197, 198,200, 203,204,206, 214, 232,235 calabashes, 234 Calcutta, 173 Calopogoniun, 53
sitios, 225
cola tree, 233
collecting, 2
Colombia, 59, 195, 226, 227, 279 Colombo, 240 Colonial Belgian Administration, 49 Colonial Governments, 44 commercialization, 5, 34 communal work, 81, 181 compost, 34, 109, 122, 164, 228
INDEX
Congo, -Kinshasa, 27, 33, 37, 39, 41, 42, 46, 49, 51,240 Basin, 23 Central, 17 Conklin, H. C., 25, 40
305
cropping (contd.)multij e, 4, 148, 149, 152, 155, 159, 165, 169, 174, 175, 187 pattern, 124 permanent, 84, 99, 100, 104, 108, 110, 119
consumption chart, 29
Cook, 0. F., 22 copra, 235, 242 kiln, 243 cotton, 36, 44, 60, 71, 75, 98, 107, 114, 128, 135, 138, 143, 147, 152, 161, 162, 178, 179, 180, 186, 194, 212 cycle, 265 couloir system, 50, 52
126 principles, 27, 62, 105, 147, 152 seasonal, 107 share, 183 spatial organization, 24, 60, 61, 104, 142, 145, 201, 203 systems, 24, 133, 138, 192 vegetable, 149, 165
winter, 123
Crotalaria striata, 118 Cuba, 85, 207 cucumber, 143 cultivation, 2 flat, 72 hoe, 71, 116 mound, 34, 108, permanent, 52, 61, 99, 100, 101, 104, 105. 112, 115, 117, 126, 127, 130, 142 . plough, 71, 113, 114, 127 ridge, 34, 70, 72, 75, 108 row, 28, 31, 70, 79, 80, 119, 152 semi-permanent, 4, 40, 52, 55, 61, 66, 68, 77. 88, 100, 105, 108, 110, 112, 142, shifting, 3, 16, 19, 24, 31, 34, 40, 42, 45, 48, 50, 54, 55, 60, 61, 91, 100, 122, 124, 166, 167, 189, 232, 254, 258 steppe, 59, 75 Cumberland. K. B., 243 cut and carry system, 130 Dahomey, 102, 238, 239, 240, 241 dairies, co-operative, 94 dairy herd, 220
tree, 3, 58, 104, 130, 142, 143, 190, 192, winter, 175
cropping annual, 203 continuous, 53, 107, 158 diversified, 159 double, 60, 114, 162, 172, 174 garden, 104 index, 107, 120, 121,123 inter-, 28, 147, 151, 152, 155, 203 irrigated fodder. 154 mixed, 25, 28, 34, 41, 62. 105, 152, 190, 194, 196, 214, 217, 223, 250 mono, 40, 147, 168
306 Dipping, 68, 94, 270 Dittmar, K., 5 Dogon, 102 drainage, 138, 181, 182 Driver, P. N., 123 dromedary, 257 Duckham, A. N., 282 ducks, 154 Dumont, R., 27, 50, 53, 60, 66, 109, 121 dung-land, 61 dung for fuel, 254 Dyson, R., 255, 256, 261
INDE X farm, 6 Chinese, 117 large-farm areas, 84, 85 management, 74, 116, 132 Taiwanese, 176 farming, contract, 210 dry, 4, 55 hoe, 4 irrigation, 4, 54, 59, 99, 131, 132, 133, 136, 139, 142,150, 152, 160,172, 177, 178 large-scale, 85, 88, 178 ley, 53, 75, 83, 92, 95, 110 permanent, 4, 58, 79, 89, 99, 101, 104, 105, 107,110,118, 122, 124,130,131, 139, 250 semi-permanent, 4,16, 22, 51, 55, 57, 58, 61, 67, 73, 74, 99, 105, 119, 124, 250 stationary, 3, 190 subsistence, 5,73, 234 system, 2, 7, 124, 142, 164, 189, 223, 234, 250, 282 tobacco, 85, 163 farmstead, earth, 109 farmyard, 105 Faucher, D., 3, 4, 55 fencing, 89, 104, 273, 274, 280 fertilizer, 104, 105, 107, 108, 109, 118, 165, 168 economy, 31, 34, 65, 87, 100, 108, 110, 152, 176 mineral, 36, 53, 67, 80, 81, 97, 98, 119, 118, 128, 152,162,164,173, 180,196, 232, 250 fertility, soil, 31, 33, 45, 81, 100, 110, 234 Fezzan, 256 field, system, 3, 99 Fiji, 207, 210 Fipa, 103 fish, 109, 163 production, 154 flax, 149 Florida, 207 fodder, 154. 170, 178 arable, 89, 111 bushes, 199 economy, I Il grass, 118 production, 101 tree, 62
Egypt, 257 Eleusine, 38 El Salvador, 193, 206, 226,227, 228 England, W.T., 85, 93 Ensete edula Horan, 213 Equador, 59 erosion, 10, 65, 88, 182, 194, 213 control, 104, 108, 265 estate, 193, 208, 222, 226, 232, 235, 236, 237, 244, 247 banana, 189, 193,208,214 cacao, 232, 235 coconut, 193, 206, 235, 242 coffee, 193, 194, 205, 220, 229, 240, 247 nucleus, 241, 251 oil-palm, 193, 235, 240 production, 210, 237, 248 rubber, 193, 234, 235, 236, 237 sisal, 189, 193, 194, 201,208, 211 sugar-cane, 193, 201,206, 207, 208 tea, 193, 229, 231, 236, 240 Ethiopia, 59, 66, 84, 87, 100, 103, 109, 190, 213, 222 evolution, 48, 79, 124 expansion areas, 124 exploitation system, 138, 193
Faidherbia albida, 62, 111, 121, 128 fallow, block, 65 bush, 206 grass, 20, 59 half-year's, 63 long rest, 65 period, 46, 65, 117, 119, 145 seasonal, 11I systems, 2, 16, 20, 25, 31, 34, 55, 56, 59,
folding systems, 34, 59, 66, 88, 108, 121 food subsistence 119, 215
INDEX Food and Agriculture Organization, 200, 221, 223, 226, 228 Foreman, W. J., 158 forest, Miombo, 91 shifting cultivation in, 19, 48, 49 Fox, J. W., 243 Frank, Ch. S., 207 Freeman, J.D., 37, 40,41 Fremond, Y., 243 French, M. H., 261 Friedrich, K. H., 216, 217, 218, 219 Froelirh, J.C., 60, 66 fruit, tree, 105, 143, 163 fuel, 118 Fulbe, 262 Galetti, R., 233, 234 Gambia, 70, 143 Ganges Delta, 169 garden, kitchen, 117 permanent, 61, 77, 104, 112, 142 gazelle, 257 Geertz, C., 31, 46, 49, 101, 124, 157, 167 Gezira, 136, 154, 160, 180 Board, 178, 181 scheme, 178, 179, 183, 186, 259 Ghana, 17, 34, 102, 195, 234 Giglioli, E. G., 177 goat, 118, 120, 123, 125, 224, 253, 254, 257 Gogo, 103 Golonpoui, 111, 112, 115, 116 gourd, 165, 234' Gourou, P., 115 grass, elephant, 18, 58, 75, 87, 94, 188 Guatemala, 188 Imperata, 17, 46, 206 Kikuyu, 87, 94 natural grass regeneration, 87 Rhodes, 94 tropical, 129, 130 grassland, irrigated, 188 tumble-down, 97 utilization, 5, 97, 106, 267 grazing, 60, 112, 118, 203, 250 bush, 112 committee, 268 communal, 67, 77, 83, 88, 267 fallow, 67, 169, 172 guards, 268 night, 86, 88, 260 permanent, 121
307
grazing, (contd.)rights, 81, 269 rotational, 78, 87, 89, 97, 270, 272, 277 schemes, 268 systems, 129, 130, 252, 273 zero, 127, 129, 130, 250 Greenland, D. 1., 3, 4, 18, 23, 33 Groeneveld, S., 190, 191, 266 groundnut, 44, 70, 98, 101, 118, 119, 120, 121, 123, 138, 143, 152, 160, 161, 191, 192, 194, 232, 234 Guatemala, 17, 33, 37 Guernsey, 93, 94 Gulard, J., 111, 112, 115, 116 Guillebaud, C. W., 211 Gui.llemin, R., 47 Guinard, A., 127, 128 Guinea, 102 Gundert, H., 275 gum arabic, 2 Gurensi, 102 Guyana, 155, 207 Hanun6o, 25 hand drill machine, 122 Hansen, S., 237 Haribes, 275 Haswell, M. R., 70 Haviland, W. E., 90 Hawaii, 207, 209 hay, making, 87, 272, 274 Heady, E. 0., 158 Henry, J., 43 herbicides, 97, 98, 175 herding, subsistence, 253 herdsmen, 260, 261 semi-nomadic, 62, 154 Herskovits, M. J., 7 Hever, J. U., 78 Hickling, C. F., 154 high-altitude areas, tropical, 88, 89, 95 Himachal Pradesh, 60 Ho, R., 144 Hodeida, 139 hoe system, 24, 56, 71, 72, 89 holding, banana, 195, 213 banana-coffee, 215, 216, 219, 223 Burmese rice, 136 cacao, 195, 232 cashew, 195 Chinese type of, 154, 157 coconut-palm, 195, 239, 242
INDEX holding, (contd.)irrigation, (contd.)coffee, 194, 195, 219 administration, 139 cotton, 59, 72, 74, 78, 136, artificial, 132 groundnut, 70 basin, 133, 135, 186 groundnut-millet, 59, 72, 119, 120, 160, controlled, 181 161, 165 furrow, 133, 135, 179, 186 irrigation, 157, 188 intermittent, 185 large, large-scale, 91, 196 large-scale, 135, 176, 182 maize-millet, 117, 120 large-scheme, 177 milk-maize, 92 method, 133, 138 millet, 118 minor, 138, 142, 176, 182, 184 millet-wheat, 121, 123 practices, 182 oil-palm, 195, 239, 240, 241 rotational, 138, 145, 146, 148, 162 rice, 73, 142, 147, 159, 168, 169, 170, 173, schemes, 176, 177 174 seasonal, 130, 136 rice-jute, 169, 172, 173 shifting system of, 145 rubber, 195, 234, 239, 240 sprinkler, 133, 135, 138, 209, 214 semi-commercialized, 196 summer, 149 sisal, 195, 210 supplementary, 136 small, 6, 78, 84, 91, 92, 94, 96, 179, 196, surface, 134 227 system, 132, 137, 155, 188 sugar-cane, 162, 206 Iso, E., 151 tea, 228 Israel, 154 tobacco, 90 Ivory Coast, 18, 102, 128, 191,192, 195, 200, vanilla, 195 208, 215, 240, 241, 243, 251 wheat-cotton, 162 Honduras, 214 British, 33 Jamaica, 207 Hong Kong, 164, 165 Japan, 154,155,156,169,170, 171,174,175, Hopeh Province, 155 181,184, 187, 229, 230 housing, Java, 101, 104, !33, 136, 154, 167, 173, 210, stationary, 57, 77, 99, 105, 108, 110 229 Hudson, N., 255, 256, 261 Joosten, J. H. L., 3 hunting, 257, 274 Jordan, H. D., 168 Jumeisha farm, 139 Jurion, F., 43 Ibadan, 234 jute, 149, 170, 172 lban,40, 41 implements, 4 India, 9,48, 59, 101, 113, 114. 116, 122, 123, Kaduwela, 239 128, 135, 138, 142, 152, 154, 157, 160, Kakamega District, 117, 118, 120 161, 162, 165, 166, 173, 193, 195, 207, Kamuku, 102 213, 230 Kano Plains, 84, 102 Indian Constitution, 122 Kanuri, 102 Indonesia, 54, 59, 193, 207, 242 Kaolack, 71 I.N.E.A.C. (Institut National pour I'ltude Karakul ranch, 274 Agronomique du Congo BeIge), 37 Kavirondo, 206 infillings, 190, 206 Uenya, 3, 59, 78, 84, 85, 87, 88, 89, 92, 94, Inselberg, 105, 106 95, 103, 105, 113, 117, 120, 131, 177, intercropping, 25, 28, 152. 242, 243, 245 193, 195, 206, 213, 220, 222, 225, 226, interculture, 28, 191, 205. 249 228, 229, 247, 250, 251, 253, 260, 264, interplanting, 23, 107, 231 265, 275, 276, 277 involution, 48, 79 Kenya Coffee Growers Association, 229 Iraq, 133 Keyu, 103 Iraque, 103 kharif, 113, Ili, 162, Irrigation, 82, 100, 101, 103, 105, 122, 132, season, 114, 122 136,137,141,159,160,161,173,184,188 Kiberege Strip, 39, 75
308
INDEX Kiga, 103 Kin ga, 103 Kikuyu, 103, 105, 112, 131, 250 Kilimanjaro, Mt., 223, 225 Kilombero Valley, 20, 21, 39, 40, 57, 58, 73,
309
Kung, P., 171 Kuru, 103 Kwimba, 68, 72, 125, 126
74 Kipsigi, 103 Kisil District, 112 Kita, 102 Konso, 103 Kroon, A. H. J., de, 48 Kulthongkham, S., 159
ley, (contd.)system, 2,59. 60. 67, 79, 80, 81, 83, 84, 85 86, 87, 88, 89, 91, 92, 95, 97, 98, 99 100, 108, 125 Liberia, 193
Liebig, Justus v., 153 Lim, Y., 230, 242 livestock, 165, 175, 221, 256, 260, 271 economy, 77, 125 density, 112, 125, 264, 265 property, 69, 262 rearing, 62
labour, capacity, 78, 125 division of, 39, 70 econonry, 36, 69, 78, 89, 112, 155, 15Q efficiency of, 78, 244 input, 39, 40, 71, 109, 119, 156, 159, 200, 226, 231 marginal returns of, 125, 157, 158 productivity of, 13, 126 requirements, 107, 108 seasonality of, 72, 115 Lacombe-Orihac, M., 72 lactation period, 255 Lahore, 161 Lake Victoria, 75, 101, 105, 118, 120, 125, land, consolidation, 81, 127 development, 124 ownership, 84 preparation, 114, 116, 141 reform, 141, 184 rights, 110, 127 tenure, 96, 141 latex, 237 leaching, 185 legumes, 108, 169, 172 fodder, 178, 180 grain, 123 pasture, 96 Leguminosae, 105, 106, 170 Leshnik, L. L., 113 lettuce, -65 Icy, establishment, 87 imp _--ment, 97 intensity of, 87 management, 86, 87 short-term, 81
198
llamas, 59 Llanos, Eastern, Western, 279 Lock, G. W., 211, 212 Lorenz curve, 262 Losso, 102 Louisiana, 207 Luanda, 66 lubia, 180, 186 lucerne, 83, 138, 162 Ludwig, D. H., 76, 103, 106, 121
MacArthur, J. D., 8, 85, 93, 197, 201, 204 247 Machakos District, 78 Machame, 223 Madagascar, 4, 54, 59, 66, 116, 130, 136, 142, 147, 154, 155, 156, 166, 190, 193, 195, 252, 260,277 Madras, 9, 152, 155, 156, 161, 163, 172 Maize, 17, 24, 43, 60, 62, 71, 75, 94, 98, 100, 105, 107, 108, 117, 118, 121, 124, 143, 149, 161, 162, 190, 191, 192, 194, 196, 212, 215, 224, 233, 234, 250 malaria, 141 Malaysia, 33, 193, 195, 234, 235, 237. 238, 240, 242, 251 Mali, 102, 109 Malinke, 102 Makonde, 103 Malthusian process, 100, 117 Mamprusi, 102 Managil Extension, 179, Mandara, 102 Mandji, 103 mango, 62, 75, 243 Manhaua, 29 107, 108, 116, 118, 120, 124, 150, 190 191, 192, 196, 237 234 Manshard, W., 1, 34 manure, 110, 122, 152, 224 animal, 67, 103, 109, 128, 153, 175, 218
manioc, 17, 24, 43, 44, 60, 103, 104, 105, Maghreb, 154
310
INDEX
Nakom-Phatom Province, 173 Nandi, 92, 94, 103 Narayana, R., 235 nematode, 85, 91 New Guinea. 16, 232, 235 New South Wales, 86 Newton, K., 51 N'Gay nc, 61 N; Kay Fong. 164 Nicolaisen, J., 145 Niger Delta, 137 Nigeria, 84, 102, 195, 200, 232, 237 Abcokuta, 18 Alayi, 17 Ilesha, 18 Northern, 96, 98, 101, 127 Umuahia, 17 nigerseed (Quireta abyssinica), 84 Nilgiris Mountains, 100 Nimar District, West, 113, 114 nitrogen fixation, 152, 185, 187 Nixdorf. 0., 227 nomads, 257 camel, 256 nomadism, 5, 252, 254, 258 partial, 5, 252, 260 semi, 5, 252, 253, 260, 264 total, 5, 254, 258, 259, 260, 264 Nowak, E., 109 Nunuma, 102 Nyanza, 120, 121, 206 Nye, P. H., 3, 4, 18, 23, 33 oasis, 145,258 oil-palm, 190, 192, 197, 198, 200, 203, 2C4, 206, 214, 233, 235, 239, 240, 241 onion, 194 overgrazing, 77, 262, 273, 274 ox, 81, 98, 122 plough, 71, 79 Ozaki, C., 156 Ozbilen, R., 171, 172 Pacific, South, 243 paddock, system, 273, 274 Pakistan, 160, 161, 162 East, 48, 166, 169, 170 West, 160, 165 Pakot, 260 Panama disease, 214 papaya, 61,233 Paraguay, 189
manure, (conid.)green, 34, 66, 91, 104, 118, 127, 128, 146, 149, 148, 151, 175 leaf, 118 pig, 164, 232 stable, 176 manuring practices, 153 market gardo.ning, Chinese system of, 163 Masai, 260, 265, 269 Masefield, G. B., 282 Matengo, 103 matrimony vine, 165 Mauritius, 206, 207, 208, 210 May, St., 214 Mbuga, 76 Mbugu, 103 McWilliam, M. D., 228 meat, 110, 254 p-oduction, 154, 264 mechanization, 155, 186 Meillassoux, C., 191, 192 melon, 234 water, 149 oi iental pickling, 151 Meru, 103 Mexico. 207 Mexican Highlands, 100 migration, horizontal, vertical, 254 Milfor, 187 milk, 110, 122, 224, 254 production, 84, 154, 217, 231, millet, 17, 24, 41, 62, 71, 100, 106, 108, 116, 119, 122, 123, 124, 143, 160, 161, 172, 196, 258 bulrush, 118 finger, 35, 43, 117, 152, 161, 163 pearl, 160, 161, 163 Milpa agriculture, 22, 37 Mimosa pudica, 243 Mindoro, 25 Miombo. 91 Miracle, M. P., 20, 22, 23, 26, 32, 35, 59 Mohr, B., 143 Molo-Mau-Narok, 92 monoculture, 194, 203, 209, 214, 222 Mozambique, 29, 85, 195 mulch, 215 Mullins, L., 156 Multar, 161 Mundang, 103 Musgu, 103 mustard,
INDEX 311 ire, 106, 257, 274 Queensland, 209, 210 mmunal, 62, 118, 261 .igated, 184 :rmanent, 98, 128, 130 Rabi, 113, 114, 122, 161, 162 ;annat, 49, 51 radish, Chinese, 165 ,105, 117, 190, 191 ranching, 54, 267, 271,277 hick, 123, 160, 161, 162 ,igeon, 105atl, cattle, 275 nout, 146 7 company, 269, 270 i P,60,6development, 273 issier, P., 60, 61, 66 Izer, K. J., 130 co-operative, 269, 270 government, 269, 270 ne Niolo, 27 group, 269, 270 pper, 234 Karakul, 274, 275, 276 :ru, 59, 207 management, 269 :sticides, 175, 178 eta, 187 nucleus, 269 system, 252, 258, 264, 271 hilippines, 9, 17, 25, 40, 147, 157, 158, 167, rape seed, 149, 152, 162, 175 193, 195, 242 raspador, 212 ,hysiognomy, ratoon, ratooning, 147, 161, 204, 208 two-storey, 202 reforestation, 205 multi-storey, 202, 203 Reunion, 207 pi~tinage, 156 Rhodesia, 32, 43, 85, 89, 90 pig, 164, 165, 170 rice, Piggot, C. J., 243 deep-water, 166 pineapple, 189, 192, 204, 243, 250 hill, 17, 24, 39, 40, 53, 128 Pinxten, K., 42, 50 nurseries, 143, 149, 166, 185 pit system, 34 ratoon, 118, 147 plantain, 215, 224, 264 spring, 149, 175 plantation, 49, 189, 196, 200, 214, 225, 228, summer, 149, 175 230 swamp, 133, 137 planting, terraces, 152 interplanting, 28, 150, 151, 202 varieties, 187 phased, 25, 64 wet, 9, 54, 59, 60, 62, 75, 105, 106, relay, 28, 105, 147, 149, 150, 151, 155, 118, 158 120,130, 131,133, 135, 138, 139, 142, row, 128 145, 146, 148, 149, 152, 157, 158, 160, under, 206 16!, 166, 167, 170, 172, 175, 187,194, Plaza, G., 214 238, 243 plough, Richards, A. ., 31, 32, 35, 43 horse-drawn, 84 ridging, mechanized, 109 ox-, 67, 70, 71, 76, 96, 109, 117 rinderpest, 270 regions, 72 risk, 12 system, 5, 24, 128 rotation, 3, 104, 105, 108, 154 tractor-drawn, 76, 89, 109, 187 annual, 107 deep, 80 biannual, 121 tractor, 94, 180 biennial, 107 Possinger, H., 29, 56, 66, 85, 208, 211, irrigation, 146 213 pseudo, 27, 38, 107 Polynesian Islands, 202 system, 2, 20, 83, 84, 86 Porka System, 34 triennial, 107 Portires, R., 65 Rotenhan, D. von, 60, 63, 64, 68, 69, 71, 72, Portuguese, 44 74, 76, 82, 115, 126, 262 potato, 60, 100, 231, 250 Rounce, N. V., 82 Pue Ruanda, 103 Puerariajavanica, 53, 203 rubber, 40, 142, 143, 192, 198, 200, 203,204, Puerto Rico, 209 206, 214, 2.34, 235, 236, 238, Punjab, 135, 161, 162 243, 245, 249 pyrethrum, 85, 92, 224 Rufiji Delta, 137
312 Rundi, 103 Russia, 231 Ruthenberg, H., 182, 250 Ruttan, V. W., 181, 185
INDEX pour le (Soci6t6 S.O.D.E.P.A.L.M. D~veloppement et 'Exploitation du Palmier A lHuile), 200, 235 soil, conservation, 45, 109, 117, 119, 127 fertility, 16, 11, 77, 101, 104, 117, 121, 122, 145, 230 hardpan, 60 mining, 77, 80, 108, 117 night, 109, 110, 175 Sokoto, 102 Soothipan, A., 158 sorghum, 18, 43, 60, 71, 98, 105, 106, 114, 116, 117, 118, 120, 122, 123, 138, 147, 162, 178, 179, 180 soya-bean, 146 spade systems, 155 Spencer, J. E., 16. 19, 23, 24, 36, 44, 46 stock-keeping, 253, 257, 258, 260 Stylosanthes, 128 Sudan, 2, 33, 103, 135, 136, 138, 154, 178, 179, 180, 183, 259 sugar-cane, 62, 85, 107, 122, 123, 135, 138, 145, 146, 147, 148, 151, 161, 162, 170, 189, 192, 204, 206, 209, 213, 220 Sukumaland, 60, 63, 64, 68, 69, 71, 72, 74, 75, 76, 78, 79, 81, 126, 262 Sumatra. 17, 49, 234, 237 sunflower, 212 Surinam, 166 sweet potatoes, 9, 24, 60, 62, 104, 105, 106, 107, 116, 124, 143, 146, 147, 149, 151, 250 Swynnerton Plan, 250 Taichung District, 149, 174, 175, 167 Taiwan, 31, 104, 116, 131, 136, 145, 146, 147, 148. 149, 151, 156, 157, 159, 160, 166, 169, 170, 174, !76, 181, 184, 207, 209,210, 229, 231,232, 244,249 Tananarive, 116, 147 Tan Chee Lian, 164 Tanga, 191 Tanganyika, 260 Tanzania, 2,20,21,23, 34, 35, 39, 57, 58, 59, 60,63,64,68,69,71,72,73,75,76, 104, 126, 137, 191, 195, 198, 208, 211, 212, 215, 216, 218, 219, 222, 223, 224 -taro, 24, 62, 102 Tax, S., 37 tea, 131, 192, 198, 204, 228, 229, 231, 232, 236, 237 teff (Eragrostisabyssinica), 84, 100 Teita, 103 Terra, G. J. A., 3, 4 terraces, 106, 131
safflower, 123 Sahara, 4,142, 145, 252, 253, 254, 256, 257 Sahiwal, 93 Sakkalaven, 260 salinity, 182 salt, accumulation, 138 Samoa, 195, 243 Samuel, P., 42 Sand Veld, 89 Sao Paulo, 221, 223, 225, 277 Sarawak, 17, 40, 41 savanna, African, 67, 70, 85, 101. 127, 253, 260 dry, 108, 284 humid, 284 Schifer, H., 275, 276 Schickele, R., 255 Schlippe, P. de, 27, 30, 39, 63. 107 seasonality, 14 Senegal, 60, 61, 62, 65, 66, 70, 71, 72, 84, 101,102, 116, 119, 120, 121, 122 S.E.R.E.S.A.. Dahomey, 238 Serere, 84, 117, 121 Setaria splendida, 188 settlement schemes, 82, 85, 88, 124 shade plants, ?03 Shambala, 103 Shao-er Ong, 159 sheep, 59, 92, 118, 120, 123, 125, 253, 254, 275, 276 shifting systems, 16, 24, 44, 104, 105, 109, 124, 284 Shinyanga, 68, 74, 126 Sick, W. D., 59, 205 siege methods, siege situation, 115, 117 Sierra Leone, 17 silage, 87, 129, 261 simsim, 18 Singapore, 143, 144, 160, 163, 164 sisal, 192, 195, 197, 198, 203, 204, 208,210, 212, 219, 231, 247 hedges, 62, 75, 198 hybrid, 212 slaughter of cows, 122 smallholder, 119, 131, 176, 193, 196, 214, 215, 222, 232, 234, 237, 238, 244, 246, 247 production, 248 smallholdings, 124, 130, 135, 195, 196, 206, 207, 210, 212, 222, 226, 232, 244, 250
INDEX terracing, 103, 109 Teso system, 84 Thailand, 4, 156, 159, 160, 166, 169, 170, 173, 174 ticks, 260 Tigre, 103 Tilapia, 155 Tjoremas, 187 tobacco, 44, 49, 89, 90, 143, 148, 149, 151, 161, 209 Togo, 102 tomatoes, 143 Tondeur, M. G., 41, 42, 50 tool systems, 24, 113 Tourte, R., 121 traction power, 104, 110 tractor, 5, 96, 97 transhumance, 5, 253 Travancore, 154 tribes, pastoral, 262 Trinidad, 129, 195, 207 trypanosomiasis, 80 tsetse-fly, 55, 59, 67, 75, 80, 281 eradication, 270 Turkana; 260 Turu, 103 Waibel, L., I Wakara, 103, 115, 118, 119. 121 Wasukuma, 66, 78, 119, 262 water, application, 133, 185 availability, 135, 160 charge, 183 distribution, 133, 142, 154, 182 hyacinth, 163 quality, 138 reliability, 132, 135, 137 supply, 133, 181, 183 watering place, 256, 258, 260, 262, 273 wattle, 250 Webster, C. C., 30, 88, 95, 96, 127, 152, 249, 256, 267, 271,272, 274 weed, 118 control, 98 killer, 166 weeder, mechanical, 98 ox-drawn, 70, 80, 119 weeding, mechanical, 79 Westgarth, D. R., 235 West Indies, 104, 130 wheat, 60, 92, 94, 100, 114, 122, 123, 146, 148, 151, 155, 160, 161, 162, 170, 175 White, H. D., 276 Whytv, R. 0., 129 Wikkramatileke, R., 164 Wilson, T. B., 30, 88, 95, 96, 127, 152, 249, 256, 267, 271, 272, 274 Wissmann, H.v., I Wittfogel, K., 141, 177 work, off-farm, 244 Woermann, E., 2, 6 Wbrz, J. G. G., 179 Wong, C. T., 165 Wood, G. A. R., 235 wool, 254, 274 231 Wright, S. J., yam, 17, 24, 103, 191, 192, 234 Yangambi, 33, 50, 53, 54 Yemen, North, 135, 139 Yoruba country, 232, 233, 234 Zacharias, C. W. B., 156 Zambia, 18, 23, 35, 85 Zande, 39 Zebu cattle, 93, 117, 128 Zeven, A. C., 241
313
266, 153
Uganda, 84, 103, 262, 263 Ukara Island, 101, 105, 106, 117, 118, 119, 120, 121, 125 Ukerewe, 68, 72, 125, 126 Uniagewirtschaft,55 United Arab Republic, 207 United Fruit Company, 214 Upper Volta, 71 Upton, M., 233, 234 U.S.A., 155, 156, 170, 176, 277 U.S.S.R., 8 Uttar Pradesh, 157, 158, 161, 162
138, 169,
153,
vanilla, 195 vegetables, 104, 105, 106, 117, 122, 123, 142, 143, 165, 170,217, 250 vegetation, cycle, 107, 135 system, 19 Venezuela, 59, 195, 279 Versluys, J. D. N., 180