GeoJourr~l 14.1 77-86

77

9 1987 by D. Reidel Publishing Company

Inequality and Development in Rural West Africa* Binns, J.A., Dr., University of Sussex, School of African and Asian Studies, Falmer, Brighton, BN1 9QN, England

Abstract: Inequality seems to be a feature of all societies, but some are probably more equal

than others. It is in Third World countries however where inequalities between particular areas, groups and individuals are most marked. The paper first traces the growing interest in both the study of inequality and the rural household as a focus of analysis. Some of the problems of measuring inequality are then considered and various indicators of inequality are examined with reference to empirical evidence from rural W Africa. The paper concludes that recent development strategies have often directly or indirectly led to greater rural inequality and makes suggestions as to how future projects might achieve greater impact in reducing the gap between rich and poor.

I ntroduction "The pursuit of equality is a mirage. What is more practical than the pursuit of equality is the pursuit of equality of opportunity. And opportunity means nothing unless it includes the right to be unequal." Margaret Thatcher (quoted in CIS

(1980) p. 1) Inequality seems to be a fact of life in all societies, yet degrees of inequality vary considerably and some societies are probably more equal than others. Even in supposedly free and democratic countries such as Britain, inequalities are often quite astounding. For example, in the mid-1970's the Diamond Commission discovered that 10% of the population owned 60.6% of the nation's wealth (CIS 1980). It is in Third World countries, however, that inequalities between particular areas, groups and individuals are most marked. In recent years this has become a topic of much interest, since m a n y regard a reduction in inequality as a major c o m p o n e n t o f the development process and a step towards raising living standards and eradicating poverty.

* Paper presented at Commonwealth Geographical Bureau, Workshop on Spatial Inequalities in the Developing World, Kano, Nigeria 16-20 September 1985

This paper will examine some of the issues relating to inequalities amongst rural people in W Africa. Recent work has increasingly focussed on the rural household, which is invariably the basic production and consumption unit and the main decision making unit affecting the character of agriculture and many other aspects of rural life. After a brief consideration of the importance of recognising inequality, we will then examine some problems of appraisal and measurement. A number of the most important 'bases' of inequality will then be analysed with reference to empirical evidence from rural W Africa. Finally, we will examine the record of some recent development strategies in contributing to a reduction of rural inequality and look towards the possible shape of future development policies.

Interest in Inequality "A more equal distribution of the gains from economic growth has emerged as an increasingly prominent development objective during the 1970's" (Matlon 1979). Two decades ago Third World rural communities were regarded as the home of an "amorphous peasantry", uniformly poor, its poverty largely of its own making. Writers referred to 'simple societies', 'subsistence communities' populated with 'irrational peasants' who were

78

seemingly oblivious to economic life and market forces. An attempt to dispel some of these myths was made in Polly Hill's important paper on N Nigeria, "the myth of the amorphous peasantry", together with Michael Lipton's equally influential work on rationality and decision-making in Indian villages (Hill 1968; Lipton 1968). These marked a major breakthrough in social scientists' understanding of rural people in the Third World and a large number of other micro-level studies followed. But the late 1960's was also a time for reflecting about the meaning of development, for critically evaluating western-based, export crop/industry-led models, and for assessing whether economic growth necessarily did result in development. Some writers used terms like 'growth without development', suggesting that far from growth 'trickling down' the transport routes and through the urban hierarchy to the poorest people in the remotest areas, national wealth was in fact becoming increasingly concentrated in a few places (such as the capital cities) and in relatively few people's hands (such as the powerful urban elites) (Clower 1966; Lipton 1977). The early 1970's represented a time when the focus of interest in development studies and development planning was switching from wealth to poverty, with increasing attention devoted to issues of inequality and distribution. A prime mover in this was Dudley Seers, who forcefully stated that development was very different from economic growth and suggested that the reduction of poverty, unemployment and inequality must be at the core of any definition of development and the main thrust in development planning (Seers 1969). In 1970 the ILO launched the first of its series of missions under the World Employment Programme, in which the general theme was "Redistribution with Growth". The poverty focus of development efforts became crystallised in the Basic Needs concept, where development was viewed as not only about raising productivity, but also, and perhaps more importantly, about satisfying fundamental human requirements of nutrition, water, health, education, transport etc. Whilst Seers and others were usefully analyzing what development should be, others were drawing attention to the need for a more detailed understanding at the local level. Terms like 'development from below' came into parlance, with its emphasis on grass-roots understanding and motivation in formulating more appropriate development strategies (St6hr 1981). Attention has recently been focussed on villages and households and there has been an increasing recognition of the wealth of knowledge about their physical and economic environment possessed by Third World farmers (Brokensha 1980). Although economists still perhaps dominate the field of development studies and development planning, other social science disciplines (including Geography) are, rather belatedly, being recognised for their valuable work at the microlevel.

GeoJournal 14.1/1987

The last two decades, therefore, have been full of debate about the nature of development and the current focus remains on aspects of poverty and inequality, particularly at the micro-level. Increasing recognition has been given to the importance of a wide range of non-economic variables and their effect on decision-making and development. It is a great tragedy, however, that such a fertile period in the (early) life of development studies, has not been reflected in policies which have effectively reduced poverty, unemployment and inequality. The gaps between north and south and rich and poor within Third World countries have widened at an alarming rate and there is much basic groundwork still to be done.

Appraising Inequality Two major problems make studies of inequality difficult; the absence of reliable statistics on various indicators of inequality and the sheer complexity of the phenomenon. Many Third World countries lack reliable series of statistics on a range of indicators of inequality. Many W African countries, for example, have not had accurate and regular population censusessince independence. Household surveys, if available, invariably relate to a small proportion of the country's area and/or population and data for rural areas are particularly fragmentary. Castro et al. (1981) identify ten groups of factors which would indicate differences of wealth in any community. These are: land, capital equipment, income, livestock, non-productive property, fuel, ceremonial expenditure, diet, nutrition and health, education, and household size and composition. However, national data on these indicators would probably only be available in a handful of developing countries. Even in countries with a small land area and small population such as Sierra Leone, it would be difficult to examine inequality in any detail from published statistics. It is surprising how, given the limited availability of data in Sierra Leone, Lisk and Van der Hoeven (1979) feel confident in concluding that about 66% of the rural population are below the poverty line. The absence of national data has prompted many researchers to draw their information from the relatively few reliable and detailed micro-level studies conducted in W Africa during the past 20 years or so. These studies, though variable, usually provide in-depth information relating to some or all of Castro's indicators (see for example Atlas des Structures, Haswell (1953, 1963), Dey (1981), Hunter (1967a, 1967b), Hill (1972), Norman (1979), Matlon (1979), Longhurst (1984), Watts (1983), Mortimore (1967, 1971 ), Swift (I 984), Binns (I 981 ), Little (I 967)). A second problem facing the student of inequality is basically, where to start? Inequality may be considered at many different levels; international, national, regional, local, and household. It is perhaps only at the village and

eeoJournal 14.1/1987

household level, through detailed sample surveys, that information on the causes of inequality or variations in inequality can be obtained. Inequality may vary spatially one village has water, another has not; temporally - the preharvest period may reveal tensions and marked inequalities not observable at other times; socially - in terms of the life-cycle stage of particular households or customs relating to religion, status or gender. These features can really only be understood in a meaningful way through detailed micro-level studies, hence the greater interest in this level of research.

Causes of Rural I n e q u a l i t y An accurate understanding of present day inequality is dependent upon careful examination of historical evidence, whether written or oral. Although inequality has always been a feature of rural communities, many W African rural societies had elaborate redistributive mechanisms to cushion the poorest members of society from the effects of severe drought and famine. Shenton and Watts (1979) have shown how food shortages were a regular feature of life in the SaheI-Sudan region and such periods of stress were ameliorated by effective long term food storage, communal farm work and complex systems of redistributive and reciprocal gifts. Dry season migration of young males (known as cin rani) in search of work helped to conserve village food supplies, whilst in Hausa society the sarkin noma (king of farmers) was elected because of his ability to produce large amounts of food some of which were redistributed at ceremonies and times of shortage (Shenton; Watts 1979). Such redistributive mechanisms are found in pastoral societies too. The Twareg of the inland Niger Delta in Mall maintain a strong pattern of social stratification, yet there is scope for households with few animals to join the others for production and consumption purposes and to remain economically viable (Swift 1984). Other transactions assist households whose herds have been decimated by an epidemic. It has been suggested that degrees of inequality have increased since the onset of colonialism and these traditional redistributive strategies have been undermined. Military conquest, the introduction of strict taxation and the encouragement of export crops undoubtedly created tensions in many rural communities. Shenton and Watts suggest that the babban yunwa or great hunger in Nigerian Hausaland in 1914 occured at a time when the population was vulnerable due to the introduction of colonial policies. Furthermore, the colonial regime refused to establish a comprehensive policy of grain reserves to lessen the impact of famine years (Shenton; Watts 1979). Franke and Chasin are equally critical of French colonial policies in the Sahel

79

and associate periods of famine with adverse environmental factors and specific taxation measures, cash crop cultivation, forced labour etc. (Franke; Chasin 1980). An appreciation of such historical detail is essential to an accurate understanding of present day inequality. We will now consider some of the most important indicators of inequality with reference to empirical evidence. Access to land

Africa was traditionally an underpopulated continent with plentiful land owned communally and guarded by the chiefs. Community members could farm where they wanted, there was little competition and buying and selling of land was rare. The situation is now changing rapidly, particularly in areas of high population density often in close proximity to growing towns. Land competition has increased, fallow periods are shorter, land is used more intensively, there is individual ownership and buying and selling of land is common place. Landholding is therefore becoming an increasingly important indicator of inequality in W Africa. In Batagarawa, N Nigeria, ownership of manured farmland proved to be a key indicator of inequality. Hill (1972) classified the 171 farming units into 4 economic groups. The 17 farming units in Group 1 owned 28% of the mapped acreage, whilst the 41 farming units in Group 4 only owned 11% of the land. All except one of the farmers with no manured farmland were in Groups 3 and 4 whose holdings were frequently less than 2 acres (0.8 ha), whereas most of those with holdings of 10 acres (4 ha) or more were in Groups 1 and 2. Richer farmers with insufficient inherited farmland often bought manured farmland adjacent to their own plots to enable consolidation. Most farm sellers were poor, forced to sell to buy food, repay loans, marriage expenses or other reasons (Hill 1972). Other N Nigerian studies also reveal inequality in ownership of farmland, but other factors are also important. Matlon, for example, found income from off-farm activities accounted for the major proportion of income variation between households (Matlon 1979). In Dan Mahawayi some 30 km from Zaria, Norman found a high degree of inequality in the distribution of farmland with 50% of the farmers owning only 22% of the farmland. Rather surprisingly, well-watered fadama land was not cultivated as intensively as it might have been due to seasonal labour shortages, flooding and the low value per unit weight and expensive transport costs of cash crops grown on this land (Norman 1973; Norman et al. 1979). Collier reported considerable inequality in land holding in the cocoa belt of SW Nigeria, with traders and professionals having the largest holdings with their access to loans, extension facilities and their political power (Collier 1983). Compared with the densely settled regions of N Nigeria, there was relatively little competition for farmland

80

in E Sierra Leone. Amongst the Kono people, land continued to be communally owned and households could make a farm wherever they wanted, provided someone else did not have a prior claim. Those wishing to make larger farms usually cleared areas at some distance from the settlement. Cultivation was for one or two years and then the land reverted to the communal pool. In Mendeland further S, land was owned by extended families whose members could clear land and make a farm whereever they wanted if nobody had prior claim. Competition for land was greater amongst the Mende than the Kono, due to greater population density, more land under permanent swamp rice cultivation, more economic trees and land devastated by diamond mining activities. Labour rather than land was the major constraint in both areas and the idea of land sales was viewed with some revulsion (Binns 1981). In some parts of W Africa inequalities in land ownership are considerable and increasing, but in other areas factors other than land are a more crucial constraint on farm size and income. Access to labour In a region where a low level of technology generally prevails, human labour continues to be a key input in farming systems. The main types of farm labour in W Africa are family, hired and communal labour, but it is the size of the family labour force which seems particularly crucial. In E Sierra Leone 67 out of 100 farmers interviewed identified labour as a major constraint on farm size. Households with large family labour reserves were most able to expand their holdings, given the plentiful supply of land. Those with limited family labour often subdivided their farms and staggered the various operations to reduce possible labour bottlenecks. The composition of the farm family was also important. Those with a predominance of very young or old members could only cultivate small areas without outside assistance. Since these families were often the poorest they also had difficulty paying for labour and few of them participated in the limited communal labour activities because of their many dependents (Binns 1981). The question of the demographic cycle of family formation and the ratio of consumers to workers was considered important by Chayanov (1966) and more recently by Matlon in the context of N Nigeria. Amongst the 30% poorest households in Matlon's sample, three sets of households were disproportionately represented - they were: (i) households headed by persons aged 60 years or older (ii) households headed by persons less than 25 years old (iii) nuclear households consisting of 7 or more residents (the average household size) (Matlon 1979). Matlon noted, interestingly, an inverse relationship between household size and income per consumer, except for a

GeoJournal 14.1/1987

small elite group in which active farmers were supported by a work force of several adult sons, and household size was three times the random sample average. He concludes that, "systematic changes in demographic factors and access to land, both of which are associated with household growth and development, contribute to a life cycle income pattern" (Marion 1979, p. 62). Many W African rural societies have communal work groups of two broad types: (i) groups of farmworkers who work in turn on each other's farms, usually with an obligation to reciprocate for work done on one's own farm (ii) groups of farmworkers working for chiefs or other local leaders to maintain important patron-client relationships. Both of these types were evident in E Sierra Leone, though communal labour was relatively unimportant compared with family and hired labour. Over 70% of farmers interviewed in Sierra Leone hired some labour, principally to overcome major labour bottlenecks such as clearing the farm or swamp and hoeing. Between 1974 and 1978 there was an increasing use of hired labour, though some farmers complained that it was either difficult to obtain or very expensive. During this period there was a greater use of hired labour for clearing and harvesting plantations of economic trees. Remuneration was either on a daily basis often with food provided, or alternatively on a contract basis in which case labourers provided their own food (Binns 1981; 1982). The variability of labour inputs on W African farms requires further investigation, since labour is obviously a major factor limiting farm size and income and could be an important cause of inequality between households. Class and access to inherited wealth Domestic slavery was widespread in W Africa until well into the 20th century and some individuals and households are poorer than others because they have traditionally been subservient. Elsewhere, strong patron-client relationships place some households in an inferior position relative to others. Swift has identified amongst Twareg of the inland Niger delta, four status groups of "free" Twareg (H/elan) and an ex-slave group (Eklan), who are descended from captives seized by Twareg warriors or purchased by free Twareg in the pre-colonial era. Present-day Twareg society exhibits marked social stratification, though slavery no longer legally exists. The Eklan were always the poorest group and today their poverty has forced them into dependence on their former masters (Swift 1984). Marriage between the free Twareg and Eklan is rare, the latter have very low incomes and are excluded from the redistributive exchanges of livestock and loans which are an important feature of this pastoralist society. Poverty has forced some

GeoJoumal 14.1 / 1987

Eklan to migrate to towns for work, whilst others have remained in the pastoral economy using any income from non-pastoral work to build up their flocks and herds, notably goats, which are less demanding and a more flexible resource base in the harsh environment (Swift 1984). Hesse's study of a group of Fulani agro-pastoralists between Mopti and Gao in Mall reveals similar stratification. The traditional nobility, the Fulbe, once entirely cattle herders, have been forced to grow crops since the liberation of their slaves, the Rimaib~, who were bought, inherited or captured to perform agricultural tasks. Though official liberation came with colonial penetration, the Rimaib~ were only effectively free from Malian independence in 1960. As slaves, the Rirnaib~ were denied access to livestock, but since emancipation have converted surplus millet and cash into livestock which are entrusted to the care of the Fulbe. The seasonal and unreliable rainfall causes periods of stress, particularly before the harvest, when Rirnaib~ purchases of cereals increase considerably, whilst the Fulbe, with their greater livestock resources, are able to live off their milk products. The Rirnaib~ are forced to purchase cereals at the time when prices are highest, work on the farm is hardest and nutritional status may be lowest. Some alleviation of stress is afforded by out-migration of young males after harvest, loaning livestock to poor relatives, and collecting bush produce (Hesse et al. 1985). Other W African rural societies have traditional ruling classes and elites who are frequently richer and have better access to education and government resources and may still exact tolls on some political and economic transactions. In Hausaland, the households of district heads, village heads and sarkin noma (head farmer) are much larger than others, they farm much more land and appropriate and allocate resources which government supplies to the village, such as fertiliser, tractors and even seed provided for famine relief (Marion 1979). Another type of patron-client relationship is exemplified by money lenders and borrowers. In Marmara, S Katsina, at least half the households borrowed money from traders and other farmers to buy grain in the preharvest period. Cash loans were repaid at harvest with 50 % interest (Clough 1982). In N Sierra Leone, Karimu found rural elites playing a prominent part in controlling credit, with interest rates between 50 and 100% per annum. Loans in kind, such as rice advanced in the "hungry season" and repaid at harvest, attracted higher rates with farmers repaying 2 or 3 bushels for every bushel of rice loaned (Karimu 1980). In E Sierra Leone, Binns found traders lending money to farmers to pay for hired labour to tend economic trees in return for a guarantee of purchase at an agreed price after harvest (Binns 1981 ). In many rural communities, class, access to wealth and subsequent patron-client relationships lead to the reinforcement of patterns of inequality, a vicious circle which poorer households find it very difficult to avoid.

81

Gender and/nequaflty In many W African rural communities men control the production and income from cash crops, whilst women cultivate less profitable food crops. As Katona-Apte comments, "the more the women contribute toward earning the cash, the more say they have on how to spend it" (Katona-Apte 1983). In nutritional terms, women and children may be much worse off than the men of the household. In SE Sierra Leone, Longhurst found a very poor nutritional situation with infant mortality rates as high as 294 per 1000. He was concerned that increasing production of coffee and cocoa under an agricultural project would raise male incomes, but this would have little effect on female incomes and child nutrition (Longhurst 1983). Amongst Mandinka farmers in The Gambia, there was a marked division of labour, with women traditionally growing rice in tidal swamps whilst men cultivated sorghum, millet and groundnuts on the free-draining upland. With the expansion of groundnut production from the late 19th century, men have concentrated increasingly on this crop, often at the expense of sorghum and millet, while women continue to grow the subsistence rice crop. Men's control of cash crop revenues brought them an increase in power within the family, which together with the spread of Islam into the area, has reinforced male supremacy in the community and household (Dey 1981). In E Sierra Leone, women were expected to help with weeding and harvesting on the upland rice farm, in addition to collecting firewood, preparing and cooking food, looking after children and undertaking most of the trading. There was a fair degree of independence in their food sales, and some women had their own vegetable gardens. Amongst the Kono, some women had a separate small farm or gbagba on which they grew rice and vegetables, the proceeds of which remained their own. However, major cash crops such as rice, coffee and cocoa remained under male control (Binns 1981 ). Men also tend to receive a larger share of the household's food supply at the expense of women and children. The rationale amongst poorer families seems to be to distribute food in relation to the earning potential of the members (Schofield 1979). In Mandinka villages, pregnant and lactating women and young children had calorie intakes only 75-80% of WHO/FAO standard requirements whilst men usually had a roughly adequate intake. Dey comments, "the women serving the food feel obliged to give more of the nutritious sauces, meat and fish to the men, while they satisfy their hunger on the bulky but lower " calorie staple" (Dey 1981, p. 117). A similar picture was found in E Sierra Leone where the men always ate first and women and children were left with the less nutritious food (Binns 1981). If men control the bulk of the cash income and are already relatively well fed, they may see no need to spend more on food for their wives and child-

82

ren. Such findings need further investigation in a wider selection of rural communities, since they could explain variations in nutritional status and other health characteristics.

Off-farm employment and income The importance of off-farm employment and income as a cause of village and household inequality should not be underestimated. In E Sierra Leone wealth generated by the development of diamond mining since the 1930's has played an important role in the expansion and increased marketorientation of agriculture and has resulted in some farming households being much better off than others (Binns 1982b). Those with adequate labour supplies have been able to make larger farms and sell more produce in the markets of the rapidly growing mining towns. Wealth generated from these sales has been invested in hired labour, larger farms, coffee, cocoa and citrus plantations, and better living standards, notably in terms of home improvements and payment of school fees and medical bills. Furthermore, since farming is mainly a wet season occupation and mining a dry season activity, the two economies dovetail and a significant number of farmers are also regularly involved in diamond mining for part of the year. Three-quarters of all farmers interviewed in two settlements, Panguma in Kenema District and Kayima in Kono District, reported increasing sizes of upland farms, vegetable gardens and numbers of economic trees. Investment in economic trees was particularly great in Kayima between 1974 and 1978, an area outside the traditional cash-crop belt, but where these items were seen as important investments. Much of this expansion was undertaken at a time when official extension services in the area were non-existent and farmers bought seedlings and learnt husbandry techniques from the few who had carried cocoa and coffee seedlings from nurseries further S in Mendeland. Local markets in both Panguma and Kayima had expanded considerably, serving as bulking points for the large scale transhipment by lorry of fruits and vegetables to the mining area markets. Much of this trade was controlled by women, some of whom moved from village to village purchasing crops, in some cases before they were harvested (Binns 1982 a & b). 53 % of farmers interviewed admitted to doing some diamond mining, but it is likely the figure was very much higher. One wealthy farmer, the paramount chief's younger brother, had the only concrete, glass-windowed house in Kayima and in 1978 was awarded a Master Farmer's certificate from the newly appointed extension officer. He had three large upland farms and citrus, coffee and cocoa plantations. Further enquiries revealed that he owned several

GeoJournal 14.1/1987

houses in Koidu, the nearest large town, and also had fulltime workers employed in diamond mining. Although he was reluctant to disclose his mining interests, it was obvious that his prosperity and living standards were to a large extent financed from his mining earnings (Binns 1981). In N Sierra Leone, Karimu discovered that many farmers had secondary sources of income from part-time trading and craft activities (Karimu 1980). Lisk and van der Hoeven suggest that about 20 % of the income of Sierra Leone rural households is derived from off-farm activities, but in some cases the figure must be very much higher (Lisk; v.d. Hoeven 1979). Amongst the Mossi of Burkina Faso, migration to Ghana and Ivory Coast is important to supply the growing cash needs of families. Roche found that migrants' remittances made up 26 % of the total incomes of "rich" households and 32% of "poor" households, though "middle" households only gained 10% of their incomes from this source. In absolute terms, however, the average cash income from migrants received by "middle" and "poor" groups was about the same. Craft work, collection of bush produce and paid farm work were the main sources of local employment (Roche 1985). In Matlon's study of three villages in Kano State, Nigeria, the increasing proportion of income generated off-farm was a major factor in village level inequality. Higher income households invested revenue earned through surplus farm production in capital intensive off-farm enterprises, whereas poorer farmers were restricted to low paid labour-intensive enterprises (Matlon 1979). As Williams puts it, "The richest people in Hausa villages derive their wealth from their position within an urban-based political hierarchy or their commercial links to urban traders" (Williams 1981, p. 33). In his study of Kankara and Malumfashi, Nigeria, Clough concludes, "Every rich farmer is, or has been, a trader, specialising in cotton-buying, grains, cattle, cloth or kola - the main exports and imports of the village community ... it is difficult to find a rich rural trader who is not also a big landowner" (Clough 1979). Alternative sources of employment and income are equally important in pastoral societies. The impoverished ex-slaves in the inland Niger delta are often forced to leave Twareg communities and undertake a range of jobs in the delta villages and Mopti. Porterage and casual labouring jobs seem to be most common, whilst women set up stalls to sell coffee, tea and other items (Swift 1984). The contribution of off-farm income to rural households may be an important component of their wealth generation and amongst poorer families it may be the only way they can stay above the poverty level. The marked seasonality in both farming and pastoral activities in W Africa creates many problems, but also provides an opportunity for other work during the dry season which may be up to eight months long. Earnings from off-farm era-

GeoJoumal 14.1/1987

ployment may be used to purchase food in the hungry season, pay for taxes, education, clothing and medical costs or may be reinvested in agriculture or livestock. Other causes of inequality

It would be impossible to catalogue all other causes of household inequality since they are many and highly complex. Three indicators which have received much attention are education, cash crops, and health, all of which may be closely interrelated. Fluctuations in cash crop prices may have a considerable effect on household income as Williams has shown in the case of the cocoa belt of SW Nigeria. In the colonial period, cocoa farmers experienced a marked increase in prosperity, yet during the 1970's they lost ground to the food producers who were well placed to take advantage of price rises and who experienced a considerable increase in living standards (Williams 1981). Investment in crops such as coffee and cocoa is always risky, since by the time the crop reaches maturity world prices may have moved adversely. Furthermore, there has been much interest in the effects of cash crops on village nutrition levels, and Schofield discovered many examples of cash crop villages with poorer nutrition status than more subsistence-oriented communities in similar areas (Schofield 1979). There is scope for more detailed work on household nutrition and health and how these features vary in relation to income and composition of the household (see for example, Pines 1983; Longhurst 1983; Katona-Apte 1983). Cash and export crops may well be first adopted by those farmers who have access to extension officers and advice. Five of the 6 village leaders in Matlon's elite group had contact with the extension agent in the previous 5 years. The elite group and their children were also very much better educated than the rest of the sampled population (Matlon 1979). In E Sierra Leone 78% of farmers interviewed in Kayima and Panguma had not attended school and were illiterate. However, many were concerned to send their children to school, provided they could afford the fees. Many farmers with no formal education were actually heavily involved in cash crop cultivation, in an area where there was little, if any, official extension provision (Binns 1981; 1982). Inequality is a complex phenomenon and frequently there are many factors responsible for one household being poorer than another. We will now consider the extent to which rural inequality has been reduced or increased by a variety of recent rural development projects. D e v e l o p m e n t S c h e m e s and Rural I n e q u a l i t y Many writers have criticised development schemes in W Africa and some would even suggest that they have direct-

83

ly contributed to greater rural inequality. "There appears to be little foundation for the assumption that the activities of rural development programmes lead to the improvement of the welfare of the rural population, let alone the rural poor" (Heyer 1981). One of the main criticisms relates to the question of inappropriate targeting and the exclusion of areas and people which are most in need of help. For example, the Funtua Integrated Rural Development Project, established in Nigeria in 1974 with assistance from the World Bank, chose to concentrate its extension facilities on socalled "progressive farmers" with the expectation that improved farming methods would then 'trickle-down' to the remainder of the rural population. This is likely to increase rather than reduce rural inequality, as Mabogunje and Gana have observed; "This minority had received a disproportionate number of visits from extension agents, of farm inputs from the farm service centres and of credit from the relevant institutions recognised by the Project. No doubt incipient class differentiation within the rural community was reinforced if not initiated" (Mabogunje; Gana 1981, p. 57). A large proportion of farmers interviewed could see no beneficial social changes associated with the Project. In Sierra Leone, where integrated projects form the major thrust of rural development efforts, areas, communities and households have been excluded. The Eastern Area Project excluded farmers if they had less than three acres (1.2 ha) of swamp that could be developed, and in so doing prevented many of the poorest households from participating (Binns 1977, 1983). In The Gambia, it was the women who were excluded from swamp rice development projects, although they had a wealth of accumulated knowledge from their traditional cultivation of swamp rice. Technical teams contacted male household heads and invited their participation. "Women were effectively excluded from owning irrigated land and receiving the credits necessary for cultivating irrigated rice on their own account, their labour, particularly for transplanting and weeding, was nonetheless crucial for the success of the projects" (Dey 1981, p. 118). In some of the irrigated agricultural schemes in N Nigeria rural communities and pastoralists have received little consideration. A survey undertaken in February and March 1984 amongst 127 Fulani in Kano State, revealed that the Tiga Dam, constructed in 1976 to provide drinking water and irrigate a vast area to the S of Kano, had caused a number of problems for the pastoralists (Binns 1984). Although water was now more plentiful, grazing and mobility had been restricted by areas of perennial cultivation. Trees, which once provided valuable shade and fodder, were removed when the scheme was constructed. Tension between pastoralists and cultivators had increased and project officials complained about animals destabilizing the banks of irrigation channels.

84

Downstream of the project, to the E of Kano, the river channel is incised and narrower than before, and no longer receives an annual flood. Once valuable fadama land, is now left high and dry above the channel and cultivation is only possible where wealthier farmers have bought diesel pumps to lift water up to their fields. As one Fulani commented, "the rivers are no longer flowing as before, they dry out sometimes even in the middle of the rains". Restrictions on grazing and mobility, together with severe drought and outbreaks of rinderpest, are causing major difficulties for pastoralists and many Fulani are abandoning their migratory existence in favour of sedentary cultivation or work in cities such as Kano (Binns 1984). Wallace is particularly critical about the resettlement associated with the Bakalori irrigation scheme in Sokoto State, Nigeria. Farmers received inadequate compensation and no compensation for their valuable economic trees. The new settlement of New Maradun had rows of corrugated-roofed houses in straight lines, no shade trees to protect the houses, no fadama land and it took over three years to allocate new farmland. People "have sold their livestock to buy grain, many have migrated to Sokoto, Kaduna, Minna to be casual labourers, a few lucky ones work for the contractors on the scheme" (Wallace 1980, p. 61). The nutritional impact of development projects is another area of considerable interest. In the Bakel District of E Senegal a 'Range/Livestock Development Perimeter' has been establishedto utilise 80,000 to 130,000ha of bush and grasslands for the market raising of beef calves. The project shows little understandingof the Fulani pastoralists' way of life and it has been suggestedthat an improvement in the nutritiona] condition of livestock will be achieved only at the expense'of human nutritional standards (Teitelbaum 1977). Cattle will be segregated from settlements and farms, thus depriving Fulani of valuable manure and ignoring the mystical attachmentof the pastoraliststo their livestock. Human use of milk will be curtailed to enable young calvesto benefit and a new system of hired herders and milkers will be introduced. This could havetremendous nutritional implications for vulnerable groups such as pregnant and lactating women, children and old people in an area with already inadequate health facilities. Teitelbaum comments, "The project conflict pits Fulani subsistence dairying against Senegalese government policy which designatesthe Sahelian region for commercial beef production intended for urban consumers" (Teitelbaum 1977, p. 137). Lunven has studied a number of developmentprojects in the Third World and concludesthat improved nutrition is not intrinsically a benefit of development. FAO has recently developed a methodology to integrate nutrition considerations into rural and agricultural development projects, but it remains to be seen whether this will have

GeoJoumal 14.1/1987

any positive effect at village and household levels (Lunven 1982). Both Lunven (1982) and Pines (1983) are concerned about the problems of excluding from projects the lowest socio-economic groups who may have malnourished members and with less accessto land and capital they face special difficulties in obtaining and using effectively the credit, services and inputs made available in projects. Another concern about the effect of development projects relates to the creation of difficulties in labour organisation and inputs, particularly in areas where labour is already in short supply. In the Northern Area Project in Sierra Leone the early development phase of establishing a swamp rice farm with associated land levelling, bund construction and digging of irrigation channels, proved to be very time-consuming in an area where shortage of labour was an important constraint. Karimu estimated that whereas 180 man/days/ha were needed for traditional upland farming, some 300 man days would be needed to develop one hectare of swamp and then 400 man days for subsequent annual cultivation requirements. The annual labour requirement is double that for traditional swamp cultivation and upland rice farming, and necessitated the employment of non-family labour to cope with seasonal bottlenecks. A major peak occurred in the transplanting period and at harvest time which frequently coincided with important festivals and ceremonies in December and January (Karimu 1980). Farmers in E Sierra Leone were reluctant to engage in swamp rice farming because of heavy initial labour inputs, the preferred taste of upland rice varieties, the coldness of the water and associated diseases,and the wide range of other crops which could be produced on an upland farm. Some of the first farmers to join the Eastern Area Integrated Project were wealthy farmers who were not so concerned about the risk involved and could perhaps maintain an upland farm at the same time to ensure subsistence food requirements (Binns 1981). Projects may alter daily routines which may involve women working more in the fields, resulting in them having less time to do other household-related activities such as attention and care of the children, with a possible decline in the extent and duration of breast feeding having important implications for nutrition (Katona-Apte 1983). Agricultural and rural development projects have had wide-ranging effects on communities and households, some of which have resulted in greater inequality. Unless the people who most need help are put in the forefront of development plans, there is a strong possibility that the gap between rich and poor will increase even further. The wealthy will benefit from a wide range of project inputs, whilst the poor are excluded because of insufficient land or because they are not regarded as sufficiently innovative and are therefore seen as a high risk in terms of fulfilling the aims of the project.

GeoJoumal 1 4 . 1 / 1 9 8 7

Future Priorities

If we accept Seers' definition of development outlined earlier, then in many so-called 'development' projects, development has not actually occurred because inequality persists and may actually have increased. It would probably be impossible to eradicate inequality altogether as it is a well-established fact of life, but future development plans must achieve greater progress in reducing inequality at a time when Africa's rural areas are experiencing more stress than perhaps ever before. Two major reasons help to explain why development plans have failed to reduce inequality. Firstly, rural societies, communities and households have been inadequately understood. Secondly, and following on from the first point, development plans have failed to target and direct assistance towards the most needy areas and people. Detailed studies of existing food production systems in the context of more general social and economic characteristics of rural communities, must precede the formulation and implementation of development projects. Great care should be taken in conducting such studies to avoid the many biases that Chambers (1983) suggests. He believes that 'rural development tourists' rarely see or talk to the poorer members of rural societies because they live 'off the beaten track', are uneducated and perhaps rather reticent. Educated people, invariably males and perhaps members of elite groups, are the main suppliers of information, whilst women, the poor, the old and the sick rarely feature in such data-gathering exercises. The picture of poverty is therefore a distorted one and a most unsatisfactory foundation for development planning. Detailed household surveys are vital, with information on labour inputs, land ownership, sexual roles and nutritional status. The importance of non-farming activities, employment and income must also be appreciated. Peopleenvironment relationships must be understood, since recent research has shown how many rural communities have a wealth of indigenous technical knowledge about factors

85

such as soil types, pests, storage problems, etc. (Brokensha 1980; Richards 1985). People-people relationships must also be appreciated, links between pastoralists and cultivators and important patron-client relationships, for example. Such micro-level research must understand indigenous aspirations for the future and incorporate these into a development strategy. Too often in the past, a transformation approach has been used and many participants have been alienated at an early stage. Information obtained from such detailed local-level research could then be used to target the most needy groups and the most backward areas, to improve nutrition, reduce labour bottlenecks, provide credit and introduce appropriate technological inputs. Too often the poorest and most vulnerable groups have failed to benefit from development efforts and in some cases they have become relatively worse off. The dynamics of inequality must also be understood. Do certain problems recur at particular times of the year? Are more long-term cycles, fluctuations and trends perceivable? Agricultural and rural development must receive more positive help from governments in W Africa - carefullytimed rhetoric is not enough. A major concern is that such poverty-focussed programmes, designed to reduce inequality, may not be seen as meeting the interests of ruling elites who are concerned to maintain their power and control both nationally and in their local base. The production of cheap foodstuffs for a more organised, vociferous and potentially threatening urban population may appear to be much more politically expedient. "It is clear that it is not enough just to 'hand out the hoes'. We have to make sure that we are giving the hoe to the right people, that there is enough land to hoe, that the hands we are trying to put the hoe into are not too busy and overworked to use it, and that the people with the hoes know enough about how to feed themselves and their families that they can actually do so, given changes in agriculture and increases in income" (Lunven 1982, p. 22).

References Atlas Des Structures Agraires Au Sud Du Sahara. (various dates) a series of detailed regional studies in West and Central Africa, Orstom, Paris. Binns, J.A.: The marketing of food crops in the diamond mining areas of the Eastern Province of Sierra Leone. Mimeo paper at Rural Geography Study Group Symposium, Institute of British Geographers, Birmingham September 1975. Binns, J.A.: Integrated Agricultural Development: a case study from Sierra Leone. Oxford Polytechnic, Disc. Papers in Geography, No. 6, Oxford (1977)

Binns, J.A.: The dynamics of Third World Food Production Systems - an evaluation of change and development in the rural economy of Sierra Leone. Unpublished Ph.D. thesis, Centre of West African Studies, University of Birmingham, UK 1981. Binns, J.A.: Agricultural change in Sierra Leone. Geography 67, 2, 113-125 (1982a) Binns, J.A.: The changing impact of diamond mining in Sierra Leone. Univ. of Sussex, Research Papers in Geography, No. 9 (1982b) Binns, J.A.; Funnell, D.C.: Geography and integrated rural development. Geografiska Annaler 65B, 1,57-63 (1983)

Geo,lournal 14.1/1987

86

Binns, J.A.: People of the six seasons. Geographical Magazine 640-644 (1984) Brokensha, D.W.; Warren, D.W.; Werner, O. (eds.): Indigenous knowledge systems and development. Univ. Press of America, 1980. Castro, A. P.; Hakansson, N.T.; Brokensha, D.: Indicators of rural inequality. World Devp't 9, 5,401-427 (1981) Chambers, R.: Rural development: putting the last first. Longman, 1983. Chayanov, A.V.: The theory of peasant economics. Richard Irwin, Inc., Homewood 1966. Clough, P.: Inequality in Kankara and Malumfashi. Mimeo paper Queen Elizabeth House, Oxford 1979. Clough, P.: Indebtedness among the rural Hausa.Africa (1982) Clower, R.W. et al.: Growth without development, an economic survey of Liberia. Northwestern Univ. Press, USA, 1966. Collier, P.: Oil and inequality in rural Nigeria. pp. 191-217. In: Ghai, D.; Radwan, S. (eds.), Agrarian policies and rural poverty in Africa. 1983. Counter Information Services (CIS): The Wealthy, Report 25, London (1980) Dey, J.: Gambian women: unequal partners in rice development projects? J. Devp't Stud ies 17, 3, 109-122 ( 1981 ) Franke, R.W.; Chasin, B. H.: Seeds of famine - ecological destruction and the development dilemma in the West African Sahel. Allanheld, Osmun & Co. & Universe Books, 1980. Haswell, M.R.: Economics of agriculture in a savannah village. HMSO, London 1953. Haswell, M.R.: The changing pattern of economic activity in a Gambia village. HMSO, London 1963. Hesse, C.; Thiarn, A.; Fowler, C.; Swift, J.: A Fulani agro-pastoral production system in the Malian Gurma. Draft report to Int. Livestock centre for Africa (I LCA), 198.5. Heyer, J.; Roberts, P.; Williams, G.: Rural Development in Tropical Africa. Macmillan, 1981. Hill, P.: The myth of the amorphous peasantry: a northern Nigerian case study. Nigerian J. Econ & Soc Studies 10, 2, 239-260 (1968) Hill, P.: Rural Hausa: a village and a setting. Cambridge 1972. Hunter, J.M.: Population pressure in part of the West African savanna: a study of Nangodi, north east Ghana. AAAG 57, 101-114 (1967a) Hunter, J.M.: Seasonal hunger in part of the West African savanna: a survey of body weights in Nangodi, north east Ghana.TRANS I.B.G. 41,167-185 (1967) Karimu, J.; Richards, P.: The Northern Area Integrated Agricultural Development Project - the social and economic impact of planning for rural change in northern Sierra Leone. School of Oriental and African Studies, Univ. of London. Occasional Paper 3, New Series (1980) Katona-Apte, J.: A sociocultural perspective on the significance of sex roles in agriculture, pp. 28--43. In: Longhurst, R. (ed.), Nutritional impact of agricultural projects. Int. Fund for Agric. Devp't, 1983. Lipton, M.: A game against nature: theories of peasant decisionmaking. The Listener 79 (28th March and 4th April, 1968) Lipton, M.: Why poor people stay poor: urban bias in World Development. Temple Smith, London 1977. Lisk, F.; van der Hoeven, R.: Measurement and interpretation of poverty in Sierra Leone. Int. Lab. Rev. 118, 6, 713-730 (Nov-Dec 1979)

Little, K.: The Mende of Sierra Leone. 2nd ed. Routledge & Kegan Paul, London 1967. Longhurst, R.: Integrating nutrition into agricultural and rural development projects: an application of the FAO methodology. pp. 73-89. In: Longhurst, R. (ed.), Nutritional impact of agricultural projects. 1983. Longhurst, R.: The energy trap: work, nutrition and child malnutrition in Northern Nigeria. Cornell Int. Nutrition Monograph Series No. 13, Cornell University, Ithaca, USA 1984. Lunven, P.: The nutritional consequences of agricultural and rural development projects. Food & Nutrition Bull. 4, 3, 17-22

(1982) Matlon, P. J.: Income distribution among farmers in northern Nigeria: empirical results and policy implications. African Rural Economy Paper, 18 (1979) Mabogunje, A. L.; Gana, J.: Rural development in Nigeria: a case study of the Funtua integrated rural development project, Kaduna State, Nigeria. UN Centre for Regional Devp't, Project on regional Devp't Alternatives Focus on Rural Devp't, Univ. of Ibadan, Nigeria 1981. Mortimore, M.J.: Land and population pressure in the Kano closesettled zone, northern Nigeria. Advancement of Science 23, 677-686 (1967) Mortimore, M.J.: Population densities and systems of agricultural land use in northern Nigeria. NIG. G.J. 14, 3-15 (1971) Norman, D.: Economic analysis of agricultural production and labour utilization among the Hausa in the north of Nigeria. African rural employment paper 4, Michigan State University, USA 1973. Norman, D.: Technical change and the small farmer in Hausaland, Northern Nigeria. African Rural Economy Paper, 21, Michigan State University, USA 1979. Pines, J.M.: Nutritional consequences of agricultural projects. pp. 44-72. In: Longhurst, R. (ed.) 1983. Richards, P.: Indigenous agricultural revolution. Hutchinson, 1985. Roche, C. J. R.: Households, food production and aid in Burkina Faso: a case study, mimeo paper presented at Institute of British GeographersConference, Leeds, January 1985. Schofield, S.: Development and the problems of village nutrition. IDS/Croom Helm, London 1979. Seers, D.: The meaning of development. Int. Devp't. Rev. 11, 4 (1969). Reprinted in: Lehmann, D. (ed.), Development Theory: four critical studies, Cass, 1979. Shenton, B.; Watts, M.: Capitalism and Hunger in Northern Nigeria. Rev. of African pol. econ. 15/16, 53-62 (1979) St~hr, W. B.; Taylor, D. R.F. (eds.): Development from above or below? Wiley, 1981. Swift, J.; Winter, M.; Fowler, C.: Production systems in central Mall: the pastoral Twareg of the inner Niger delta. ILCA Arid and semi-arid zone programme, 1984. Teitelbaum, J. M.: Human versus animal nutrition: a "development" project among Fulani cattlekeepers of the Sahel of Senegal. pp. 125-140. In: Fitzgerald, T. K. (ed.), Nutrition and anthropology in action, van Gorcum, Assen/Amsterdam 1977. Wallace, T.: Agricultural projects and land in northern Nigeria. Rev. of Af. political econ. 17, 59-70 (1980) Watts, M.: Silent violence: food, famine and peasantry in northern Nigeria. Univ. of California Press, Berkeley 1983. Williams, G.: Inequalities in rural Nigeria. Univ. of E. Anglia Devp't. Studies, Occ. paper 10 (1981)