|Water Management in Africa and the Middle East: Challenges (IDRC, 1996)|
|Part I - Concepts|
Director, Regional Office for Eastern and Southern Africa, IDRC, Nairobi, Kenya
By the early 1990s, nine African countries had per capita renewable water supplies of less than 1 000 m³/year and were considered water scarce. The nine were Algeria, Botswana, Burundi, Egypt, Kenya, Libya, Mauritania, Rwanda, and Tunisia (Postel 1993). However, most other countries in Africa have large regions that suffer acute water shortages, either periodically or on a permanent basis. Consequently, in most of the continent, effective water-resource management is of critical importance.
In situations of scarcity, decisions about access to and use of water involve actors at the intergovernmental, governmental, regional, community, and household levels and often become highly politicized. The needs and perspectives of large- and small-scale farmers, of small- and medium-sized enterprises, of households, and of fisherfolk and others who earn their livelihood from water can differ significantly. At the same time, level of commitment of the different actors to conservation practices and to protection of water resources from contamination may also vary, and the question of whose interests prevail and receive top priority can create considerable tension.
At a global level, demand for water is increasing steadily, with a general trend toward diversification of use away from agricultural activities. Currently, about 70% of world freshwater resources are used for agricultural purposes, but with rapid global industrialization this is expected to decline to 62% by the year 2 000 and to even less thereafter. Moreover, with increasing populations and improved living standards, domestic demand for water has grown significantly in all parts of the world, including Africa (Biswas 1993). In this context, the past decade has seen massive expansion of water projects in Africa. Although the African continent is still the least-irrigated and least-industrialized region of the world, sustained efforts continue to be made to provide safe and reliable water sources in rural and urban areas throughout the region, both for domestic consumption and for agricultural purposes.
This paper examines some of the concerns that have motivated African governments and donors to become involved with water projects. Although there is general recognition of the needs of communities for reliable water systems, it is argued that the different attitudes, perspectives, and needs of women and men with respect to water access and use have been given little focused attention by environmental planners and water-resource managers in Africa. More specifically, it is suggested that throughout the 1970s and 1980s, although concerted efforts were being made to increase water accessibility, little effort was made to integrate the economic roles of women into water-resource planning.
Drawers of Water (White et al. 1972), a classic American study of East African water issues, presented strong intellectual justification for donors giving attention to the provision of domestic water to rural African communities and becoming aware of the sociopolitical issues related to conflicting needs and competition for water resources at the community level. The study became a standard reference work for water-resource planners in the 1970s and 1980s (M.dhouse, personal communication, 1995), but despite its specific focus on social issues, including water use, health, individual costs, and communities, it did not address the role of women in water-resource management, except to note that in most African societies it is considered womens work to carry water. White et al. (1972, p. 63) also noted that husbands determine the arrangements, such as the site of the house or the investment of money in equipment, which make the job light or heavy. Their research implied that women were not significant decision-makers, even with respect to domestic water use and sourcing. In the ensuing years, this notion has been uncritically accepted, time and again, in the design of water projects around the continent.
Although research in the 1990s has demonstrated that African women are active participants in economic development, there has been relatively little systematic factoring of gender considerations into resource-allocation decisions. Despite substantial evidence of the economic profitability of this approach, traditional assumptions about the domestic roles of women continue to guide policymakers.
This paper suggests that the inherently conservative attitudes of environmental planners and water-resource managers toward mainstreaming gender concerns into their analyses reflect those of the broader society. It also suggests a continued, deeply rooted resistance to power sharing: this resistance finds its origins in a desire to maintain the status quo and to see women as passive bystanders. It provides a means of reducing political controversy around allocation of water resources because, by minimizing or discounting the specific needs of women, water-resource managers are able to remove at least one potential source of conflict over water allocation.
Water-resource planners, representing both national governments and donors, have focused their attention on womens reproductive, primarily domestic, responsibilities. This has been a fundamental error because womens strategic interest in water is closely associated with their productive roles (which, in turn, are often synchronous with their reproductive tasks). In overlooking this reality, planners have seriously undermined the capacity of women to make a substantial contribution to national-development processes, and this has led to significant losses of water resources in Africa.
The paper begins with an examination of the attitudes of donors and governments toward water-resource management. A number of inconsistencies and points of conflict between these attitudes and user needs are identified and discussed. The paper then examines the extent to which women have input into water-resource management, focusing on their needs with respect to both their reproductive and productive roles.
Inconsistencies and points of conflict
Although donors and governments often use the term gender when addressing different needs with respect to water resources, they usually mean women. Male needs and priorities provide a standard against which female (or gender) interests are measured and often minimized. A true gender focus on water-resource management would entail a complete rethinking of the issues of access to water, of the differing and sometimes conflicting interests of men and women, of their relative input into and power over decision-making, and of the role and composition of the community. These are difficult and potentially contentious questions, and it is not surprising that they receive little attention from water engineers, who assume that the provision of water facilities is the primary task at hand and that related social issues will sort themselves out over time.
As noted, governments and donors usually see womens involvement in water-resource management primarily from the perspective of their roles in social reproduction. Green and Baden (1995) cite numerous examples from World Bank documents about womens role in providing, managing and safeguarding water for use by the family. Because household water provision is still a female responsibility in most African societies, especially in the rural areas, donors and governments have tended to assume that womens strategic interest in water is concentrated primarily in having access to convenient, reliable, and safe sources close to the homestead. Further, because women are seen as having first-line responsibility for the maintenance of family health, they are thought to have a special interest in, and responsibility for, water and sanitation. Available evidence suggests that, although both assumptions have some validity, they fail to capture womens equally pressing needs for water to enable them to engage in economic production, whether in agriculture, in microenterprise, or in other income-generating areas.
Government planners rarely factor microanalysis of daily water needs into national water-resource plans. To some extent this is a function of the operational models of most African governments. There are often weak linkages between ministries of economic development and ministries looking after the welfare of women and children. Ministries of economic development (often responsible for water-resource-allocation planning) tend to focus primarily on the attainment of overall water objectives and less on the needs of, or impact on, different sectors of the population (Yoon 1991).
Both governments and donors recognize that water-resource planning must be multisectoral, but water projects are usually designed to meet technical rather than social objectives. Although it is expected that installed water systems will be used by local populations, engineers rarely examine the pattern of needs of different groups of potential users and take this information into account when designing the system. It is usually assumed that social groups will change their habits of interaction to accommodate and take advantage of available water, rather than its being recognized that newly available water will not be used optimally because this does not conform to the existing norms of social groups. Because of this lack of sensitivity to prevailing cultural patterns, it should not be surprising that water projects often fail, especially after donor support is withdrawn.
The Lesotho Highlands Water Project provides a large-scale example. The project includes four large dams, water-transfer works, and a hydroelectric power plant. Funded by the World Bank, the European Economic Community, the European Investment Bank, the United Nations Development Programme (UNDP), several bilateral agencies, and the Government of South Africa, the projects objective is to divert water from the Orange and Senqu rivers in Lesotho to service South African industry (Jayseela 1993). The project will destroy between 1 and 1.3% of Lesothos arable land, which is significant because of the overall shortage of arable land, and substantial grazing land will also be appropriated. It would appear that the needs of South African industry have prevailed over those of local farmers, but because Lesotho has a very high proportion of female-headed households, it is likely that women will be disproportionately affected.
Because womens productive roles usually are not taken into consideration by macrolevel environmental planners, it is not surprising that women were apparently overlooked by the planners of the Lesotho project. But even at a household level, prevailing power relations between men and women usually ensure that the water needs of women receive lower priority than those of men, although the economic contribution of womens work can be of significant importance to the household. As discussed below, womens access to water resources is often strictly controlled, both directly by their husbands and indirectly by existing cultural norms and practices.
Most interventions in the water sector have focused primarily on expanding the supply or availability of water for given populations or on serving specific needs, like irrigation or industrialization. Water-resource planners have assumed that demand exists and that it will adjust equitably to new sources of supply. Disproportionate attention has been focused on issues of delivery and on short-term indicators of achievement (for instance, finishing on schedule and within budget and providing for certain quantities of water flow). Commonly, it has been assumed that populations will be equitably served and that individuals and groups will have access as required.
This has proven to be mistaken, and there is considerable anthropological evidence to suggest that often populations tend neither to be equitably served nor to adjust their water-related practices to take full advantage of new sources of water. To illustrate, it is useful to examine womens water needs in the context of their reproductive and productive roles. It should be stressed, however, that the bifurcation of womens responsibilities into reproductive and productive ones is artificial - womens work tends to encompass both types of responsibilities within a single context. For example, household food processing, which is done primarily for family consumption, often creates a surplus that is sold outside the home and becomes an important source of additional family income. Because of this, the reproductive and productive roles of women routinely become merged.
Womens reproductive roles
The term reproductive is used here in the sense of social rather than biological reproduction. It refers to all of the services provided by women to ensure the healthy maintenance of their families, including cooking, cleaning, and child care. Because reliable and convenient access to potable water is important in helping women fulfill these tasks, donors and governments often assume that womens primary strategic interest in water relates to their domestic roles. Significant research on women and water has been undertaken from this perspective, focusing especially on
· access to water;
· decision-making related to water use; and
· family health and water and sanitation.
Access to water
An average household in developing countries consumes about 40-60 L of water daily for drinking, cooking, cleaning, personal hygiene, etc. Meeting this need usually entails several trips for women and children to water-collection points, sometimes involving several hours. In some mountainous regions of East Africa, women spend up to 27% of their caloric intake in fetching water (Lewis 1994). Many traditional rural water sources have become contaminated as a result of human and animal waste and agricultural runoff. Especially during the dry season, rural households often collect their water from contaminated sources.
In urban areas, the situation is not necessarily better, and when household cash resources are meagre, it may be worse. For example, in Nairobi, slum dwellers buy water from vendors or collect it from communal water points, which are often highly unsanitary, and in Mombasa more than 60% of slum dwellers have no access at all to clean water (Government of Kenya and UNICEF 1992). In Kumasi, Ghana (as in many other African cities), water connections are shared by numerous families, and hygienic conditions are inadequate. The poorest families are forced to purchase water from more prosperous neighbours who have access to water connections (Whittington et al. 1993). Generally, in slum areas in African cities, there is considerable water contamination. Human and other waste is dumped into streams and drainage ditches or dries and becomes airborne, thereby creating a serious health hazard.
Although there have been important exceptions, small-scale water-supply projects aimed primarily at satisfying domestic needs have not been a priority among bilateral donors (Dankelman and Davidson 1988). There are many examples of failed small-scale water-supply projects in Africa. These projects often have been developed in response to specific political agendas, and, consequently, selection criteria for their location have been neither rigorous nor need driven. Projects have commonly been started with inadequate information bases concerning the needs, preferences, and level of commitment of the community (Harnmeijer 1993). Because political capital is at stake, there has often been no real attempt to gauge the interest and commitment of potential beneficiaries. Consequently, when donors have withdrawn, sometimes there has been little interest at the community level in maintaining the system. Moreover, the costs of continuing maintenance and repair have sometimes not been accurately judged by donors, which means that community members may be unable or unwilling to make the necessary financial contributions to keep the new system smoothly functional. These are critical considerations in the sustainability of water projects because specific expectations exist on the sides of both the donors and the potential beneficiaries. For example, a study in Tanzania found that local water systems were not maintained by the communities because they considered this the continuing responsibility of the agency that had installed the system. Villagers had provided labour to install the pipes, but they considered maintenance as being the duty of the agency (Waters 1992). This underscores the fact that, before water-supply projects are implemented, there should be in-depth interaction with local communities, including women, to ensure that expectations and responsibilities on both sides are clearly stated and understood. As discussed below, it is not uncommon for donors to assume that men speak on behalf of the community, because women are hesitant to voice opinions in public meetings or fora.
Research on the use of installed water systems has shown that women consider various factors when deciding where they will obtain water. Water planners often take a somewhat linear approach to water-sourcing, thinking that women (and men) will make choices based exclusively on cost-benefit analyses in terms of time, convenience, and water quality. However, numerous other factors can come into play, and at different times of the year and for different intended purposes, water choice may be driven by a set of unquantifiable perceptions. In some cases, for example, time spent queuing for water at a village pump may be perceived as a welcome opportunity to share gossip and information and to socialize; at other times, use of the village pump may be rejected because of pressure of other obligations that mitigate against the necessary time commitment. Studies in Egypt and Ghana showed that women did not use available public water points or pumps because of the waiting time involved. In both cases, time and convenience influenced decision-making more than water quality did (El-Katsha and White 1989; Kendie 1994).
Sustainability of water and sanitation systems is often problematic in the absence of year-round use of the systems. In some cases, to save time, women will use closer sources of water, even if the water quality is not optimal. In other cases, the financial contributions needed to maintain the system may become too burdensome for communities after donor support has ended, and systems may remain in disrepair for long periods (Narayan-Parker 1988). Lack of local expertise to repair and maintain systems also continues to be a problem, as does unavailability of spare parts.
Decision-making related to water use
Cost recovery and willingness to pay have become considerations for donors in the 1990s, particularly for the World Bank in lending to the water sector. Results with cost recovery in rural areas have been mixed.
Community water projects have appointed committees to collect money from households using the installed water systems, but cash contributions for water-system maintenance have been especially difficult to collect over the long term. It tends to be more problematic to collect money if alternative traditional sources of water are close by or if the water produced by the system is of poor quality and quantity (Narayan-Parker 1988). Gender analysis can lead to some insights because decisions about how household cash resources will be spent are often made by men (who may not have a strong commitment to contributing to water-system maintenance because collection of water is not their responsibility).
In urban areas, householders often pay for a water connection or buy water from informal-sector vendors. Again, in cases where women do not have their own income or control over how it is used, decisions about the purchase of water are commonly made by male heads of households.
In recent years, the World Bank has undertaken considerable research on how much families are willing to pay for individual access to potable water, especially in urban areas. A 1990 study in Morocco found that poor urban households placed a high priority on access to potable water, ranking it in importance immediately after food and clothing as a necessity for which they would be willing to pay (McPhail 1993). A Nigerian study showed that in Onitsha, householders regularly paid 3-5% of their income for water (Whittington et al. 1991). During the dry season, the poorest households paid as much as 18% of their incomes to purchase from private water vendors, and the researchers concluded that on an annual basis, households in Onitsha pay private vendors more than twice what it would cost to operate and maintain a piped water system.
Because most of the World Bank studies did not examine differences in the views of men and women, it is difficult to determine whether gender-based preferences were present. Studies that have desegregated gender attitudes toward payment for water have reported mixed results. Green and Baden (1995) cite evidence from Tanzania and Haiti that suggests that, in general, women were more willing than men to pay for access to public taps, whereas in India and Nigeria, women were not prepared to pay as much. Womens limited decision-making influence over household finances may make them reluctant to suggest greater expenditure on improved access to water, particularly if this expenditure would primarily benefit the women by easing the burden of their household work.
Cultural patterns may also have an impact. Research in Tanzania revealed that even when water was readily available, women hesitated to use it to keep their childrens faces clean because this was not considered a priority and went against the wishes of their husbands (McCauley et al. 1992). A study in Kenya (Government of Kenya and UNICEF 1992, p. 91) noted that there is evidence that in some communities men will give precedence to the building of a corrugated iron roofed house, purchase of a bicycle or marrying of a second wife over the supply of basic house hold necessities such as food and water. It is evident that prevailing cultural role expectations for both men and women can have critical importance in determining attitudes toward water-resource use and management.
African womens capacity to have input into water-resource use and management is further hampered by their lack of exposure to science and technology. To make informed choices and decisions, potential beneficiaries of water systems should have basic knowledge and understanding of the technologies involved. Women are particularly disadvantaged because of their lack of confidence about technological matters and because of negative male attitudes toward female technical knowledge.
Donors have made many attempts to address this issue, particularly in the context of womens reproductive responsibilities. For example, a workshop on water and sanitation, held in Kenya in 1987, emphasized the need for women to be involved in the choice and transfer of water-related technologies. It was stressed that women should have input into decision-making related to the use of individual or collective water technologies; they should be educated on the scope of available technologies so that they can make appropriate choices and recommendations; and they should be consulted about the cultural implications of different technologies (INSTRAW 1987a). It was strongly recommended that women be given appropriate training to allow them to participate in water- and sanitation-related decision-making at all levels. United Nations agencies and bilateral donors sponsored many similar workshops in the mid- to late 1980s in an attempt to erase prejudices about the involvement of women in the repair and maintenance of water and sanitation systems.
In Botswana, the Swedish International Development Agency assisted with the establishment of water points in rural areas throughout the 1980s, and by the early 1990s more than 80% of the countrys rural population had reasonable access to safe water. However, an evaluation study showed that household-use patterns had changed relatively little. Water-related hygiene practices in rural households were still poor, and households still fetched water, on average, seven times daily. In retrospect, it was recognized that the project had failed to take into account the traditional roles assigned to men and women in Botswana (Simpson-Herbert 1992, 1994). First, women were not involved in the original planning of the Botswana water projects. The projects were seen as community-based ones and it was assumed that womens input would come through the community, but in reality the cultural practice was for men to make decisions on behalf of the community. Second, once the water-supply systems were in place, women became involved as operators, rather than as more highly paid managers or professionals (these positions were reserved for men). Third, few women were given the opportunity to seek professional training. They were kept in the clerical pools. The evaluation found that prevailing cultural roles made it highly unlikely that women would seek technical training and that the women themselves had low confidence in their ability to perform well in technical jobs. The overall effect of these combined factors was to marginalize female participation in the water projects, despite the fact that women were considered as having strong interests at stake in the establishment of rural water systems. Women were expected to contribute labour, but they had little real power or input into decision-making.
In contrast, the South Coast Handpump Project, in Kenya, developed by PROWWESS and UNDP in collaboration with the Kenyan Ministry of Water Development, emphasized female participation in pump repair and maintenance from the beginning. This undoubtedly was a critical factor in the projects success (Narayan-Parker 1988). Initially, the project had been intended exclusively to train women for pump repair and maintenance, but the female participants themselves requested male participation, arguing that young women would get married and move to other communities. Interestingly, both men and women preferred to have women serve as treasurers for water committees - both sexes believed that women were more trustworthy with money. The success of this project underscores the importance of flexibility and responsiveness to local cultural patterns as an element of development planning and implementation.
In general, however, although projects and workshops aimed at teaching technical skills to women have addressed important deficiencies and problems, they have also tended to reinforce rather than challenge the notion that women should have primary responsibility for ensuring the availability of water for domestic purposes. As such, they have not questioned the gendered power relations and role expectations of women and men in communities or promoted the idea of joint responsibility for water and sanitation. In this sense, they have not gone far enough in changing attitudes.
Family health and water and sanitation
Health education has been an important entry point for womens participation in water projects. Although measurement of the effects of good water and sanitation on health is methodologically complex, there is some evidence that when there is significant improvement in a communitys level of water and sanitation, the impact on health can be substantial (Government of Kenya and UNICEF 1992).
As a result of the sexual division of labour, women are often at higher risk of exposure to waterborne diseases. Washing clothes, bathing children, drawing water from surface sources, and, in some regions, working in flooded rice fields all increase rural womens risk of exposure to disease-ridden water sources (Lewis 1994). Cultural preferences, shared by men and women, can also have a negative impact. A study in Egypt showed that women (and men) believed that canal water lathered clothes more effectively and made them whiter, so they preferred to wash their clothes in the canal (where risk of exposure to schistosomiasis was high), rather than washing them under a tap or in a washing machine (El-Katsha and White 1989).
African governments and donors have tended to see the maintenance of family health through safe water and sanitation as primarily a female responsibility. For example, in 1987 the United Nations International Research and Training Institute for the Advancement of Women held seminars in Ethiopia and Kenya that focused specifically on women, water supply, and sanitation. These seminars moved beyond a purely functional perspective on womens involvement because they emphasized the need for women to become key actors in water supply and sanitation, with input into decision-making and knowledge about technical maintenance of local water and sanitation systems. At the same time, however, they focused heavily on womens role in cleaning the environment and in looking after family hygiene and on educating women about water and sanitation issues (INSTRAW 1987a, b). To some extent, this reinforces the notion that women have a special responsibility for family sanitation and hygiene. Such efforts do not address the more fundamental issue - that water and sanitation facilities, which are used by both men and women, should be a joint responsibility, not only in terms of sharing of labour input but also in terms of sharing of decision-making and associated power.
Some African research has suggested that, even when the potentially negative health effects of poor water and sanitation systems are known and understood, neither women nor men necessarily give them a lot of consideration when making decisions about water use. For example, a Tanzanian study showed that although women understood the relationship between clean faces and trachoma, they did not wash their childrens faces more regularly when water became easily accessible (McCauley et al. 1992). In Kumasi, Ghana, public latrines were often not used because they were inconvenient (Whittington et al. 1993). A study in Tanzania found that villagers did not consider it their duty to maintain water pipes provided by a donor agency, and they were completely unswayed by the clean-water-good-health argument because it was inconsistent with their own world-view (Waters 1992). This would seem to suggest that public-health messages about the relationship between clean water and good health not only need to be reconceptualized but also must be addressed to both men and women. In other words, maintenance of family health should not be seen as solely a female responsibility.
African womens groups have often been used by donors to implement water and sanitation projects. Frequently, income-generating activities have been added to the projects to make them more attractive to women. However, experience has shown that when such projects become successful they are sometimes called community projects, and men take credit for them (Ongwen 1993). In this way the community label can be potentially damaging for women. Similarly, views that are supposedly held by the community may in fact reflect the opinions of a few of the wealthier, most influential individuals (usually men).
The perspectives of donors and governments and of women and men can be quite divergent on issues of control and decision-making in domestic water access and use. First, it seems clear that, at the village level, decisions are not necessarily made according to either a linear or a Western, rational model. Although it might seem self-evident to a donor representative that maintenance of good health should be a top priority, this may not be so in the mind of a woman burdened with the need to complete a multitude of daily tasks. Her priority might be to save time or to save money; therefore, she might choose to use contaminated water for which she does not have to line up or to pay.
Second, although donors may assume that cultural practices will be changed to take advantage of water that has been provided, this is not true. Cost-benefit analyses (both in terms of economic cost and in terms of time and convenience) do not capture the symbolic importance that people may attach to prevailing norms and practices, and this symbolic importance may well be the decisive factor when choices are made.
Third, the motivations of men and women to pay for potable water may differ, and in an unequal power relationship, it is more likely that the interests of men will prevail. Because provision of water is considered a basic duty of women, the lightening of that task will not necessarily be given a high priority by men, who make the decisions about expenditure of household incomes.
Finally, it would seem that donors have increasingly tended to assign responsibility for water-system maintenance and repair to women, under the assumption that women have the more strategic interest in ensuring operational efficiency to enable them to meet their domestic needs. Consequently, donors have developed womens water projects and sponsored workshops, seminars, and short courses aimed at breaking down stereotypes about women and technology. This has introduced women to aspects of water technology and exposed them to some basic technical skills. However, it has also reinforced the idea that women should continue to take primary responsibility for meeting household water needs and maintaining family health, especially in relation to waterborne disease. In this sense, it reinforces the existing status quo and the notion of socially reproductive labour as being almost entirely a female responsibility.
Womens productive roles
Rural African women need access to water resources to carry out their reproductive responsibilities, but they also need access to water to undertake work with direct economic benefits for themselves and their families. Sometimes their needs are in direct conflict with those of male members of their households (although at other times interests will be shared or complementary).
Interestingly, there is very little literature about womens need for equitable access to water to enhance their capacity to earn livelihoods and contribute to household incomes. The importance of water as a key input into economic production, especially in agriculture, is obvious and has been extensively discussed in the literature on water-resource management in Africa. But the question of access to water is seldom desegregated by gender. This is most apparent when one examines economic production outside the agricultural sector.
Womens participation in irrigation schemes
Irrigation is not as widely practiced in Africa as in the Middle East and other regions. However, existing projects have tended to focus primarily on the development of appropriate irrigation hardware. Progress is measured by the amount of land put under irrigation, rather by than the extent to which farmers make effective use of irrigated land and the consistency of the irrigation system in place with prevailing social practices and patterns (Diemer and Vincent 1992). Not surprisingly, the success rate for large-scale irrigation projects in Africa has been low (Brown and Nooter 1992).
For the most part, women have been omitted from both large- and small-scale irrigation schemes in Africa, often because of land-tenure issues. Zwarteveen (1994) suggested that irrigation settlement schemes in Africa and elsewhere have usually been based on three assumptions:
· Male heads of households control farm resources and labour.
· Improved incomes for male farmers will lead to improved quality of life for the entire household.
· Farm households are composed of nuclear families (man, woman, and children).
It is clear that these assumptions have had inadequate empirical verification.
The establishment of irrigation systems in Africa has been marked by some of the same inadequacies already noted in the design of large-scale water projects. Baseline information about the prevailing gender division of labour in agriculture, farming practices, and land-tenure systems has rarely been collected and carefully analyzed before irrigation systems are established. The three above-mentioned assumptions about family structures and household resource allocation have often guided irrigation planners. One consequence has been the dispossession of female farmers. Research on the Blue Nile Irrigation Scheme, established by the British in Sudan in 1954, found that women traditionally had the right to own land in the region of Wadal Abbas. When the irrigation scheme was established, land was taken away from existing farmers, including women, and reallocated exclusively to men. There is strong evidence of a decline in female farming as the irrigation schemes became widespread (Bernal 1988). Similarly, when irrigated rice farming was introduced in the Jahaly Pacbarr Project, in Gambia, resource and access rights of women declined. Although women benefited from the increased economic prosperity of the area, they became more dependent on senior male heads of households, providing labour for their land, whereas in the past the women had usufructuary rights of their own (Carney 1988). An irrigated rice project in Cameroon was unable to pay for itself because women, who were not assigned land but were expected to work in their husbands fields, withheld their labour in order to grow sorghum for family subsistence outside the irrigation scheme (World Resources Institute et al. 1994). In Kenya, the Mwea Irrigation Scheme appropriated all available land, investing its control in the hands of the scheme managers, who were men. Women lost rights to land they had traditionally used to grow food crops for subsistence. Consequently, women were forced to turn to their husbands for cash to buy food and became more dependent on men than they had ever been in the past (Zwarteveen 1994).
A recent study of smallholder irrigation schemes in Zimbabwe found that irrigating households had higher yields of cash crops but that few female-headed households belonged to irrigation schemes (Chabayanzara 1994). A primary reason for the low representation of female farmers was the prevailing government policy of allocating irrigation plots exclusively to men. In fact, in the Zimbabwean study, those women who headed irrigating households tended to be de jure heads, mostly widows. De facto female heads of households were more prevalent in dryland farm areas because the lower cash incomes earned from such farming practices encouraged men to migrate to urban areas to seek alternative employment. It would appear that traditional, conservative views about the role of women, held in this case by government officials, made it impossible for female farmers to gain entry into the scheme, except by default. Therefore, female farmers had no choice but to continue to practice dryland farming.
Research in Kenya on smallholder rice irrigation in the Kano Plains revealed similar inequities (Hulsebosch 1993). Most women were not active members of the water users associations in the rice schemes, and those who did attend meetings were not allowed to speak before men or to express opinions in opposition to those expressed by men, despite the fact that women performed up to 61% of the requisite labour in their own and their husbands plots. Even when both men and women participated in irrigation schemes, their needs and priorities sometimes differed. Women were less interested in night irrigation because cultural norms made it difficult for them to work after dark. Men were interested in having watering places for cattle; women in having communal areas for washing clothes and dishes. These different perspectives were not effectively represented by the water users associations because women were underrepresented and were not given an equal voice in decision-making. The study also found that the water guards in the rice-irrigation area were exclusively male, and men generally tended to receive more water from the irrigation schemes.
Because there is considerable evidence to suggest that women are not equal partners and beneficiaries in smallholder irrigation schemes, it is not surprising to find that women may sabotage irrigation efforts or withhold labour as a form of protest. In Kenya, Cameroon, Burkina Faso, and Gambia, there is evidence that women have withheld their labour from or reduced their labour input to their husbands irrigated plots if they felt that they were receiving insufficient compensation for their work (Zwarteveen 1994). Similarly, Brown and Nooter (1992) pointed out that women often withdraw from irrigation schemes because the cash benefits from participation go to their husbands (who are the official holders of the land and the official participants in the schemes).
Womens work outside the agricultural sector
Aside from agriculture, the informal sector is Africas biggest single source of employment. In all sub-Saharan countries, both rural and urban women are involved in petty trading, selling of cooked food, and brewing of ale and beer or other traditional drinks. Other informal-sector activities typically undertaken by women include running tea kiosks; processing and selling street foods, like rice balls, roast maize, or groundnuts; producing handicrafts; selling charcoal or firewood; tailoring; and making dresses. In some cases, women prefer to participate in group enterprise because it offers protection against interference or manipulation by husbands or male relatives (Jiggins 1989). However, a Ugandan study found that rural, female small-scale entrepreneurs tended to be isolated and to work independently (Kyomuhendo 1992).
Commonly, the activities undertaken by women are extensions of their domestic roles, and women often operate directly from their homes, sometimes relying on assistance from their children. Most of these businesses require a low initial capital outlay, but access to water is often essential for both production and sanitation. There appears to have been little analysis of the importance of access to water in womens choice of particular informal-sector business activities, in the success or failure of their businesses, or in the capacity to expand their business activities. A study of urban agriculture in Nairobi revealed that urban food production is an important source of family food and additional income for women but that womens access to irrigation was minimal (Freeman 1993).
A study of womens petty-commodity production in rural Uganda revealed that economic necessity, either that of providing basic support for their families or that of supplementing inadequate incomes of their husbands, was the basic motivating factor for participation in informal-sector economic activities (Kyomuhendo 1992). In no case was economic independence or a general desire to improve socioeconomic status a primary motivating factor. More work needs to be done to verify these findings in other countries, but it would seem logical that women engage in informal-sector work primarily because they need the additional income to sustain themselves and their families. If this is indeed the case, then there is a strong argument to be made for incorporating their needs for access to water for economic production into water-resource planning and for assigning such needs the same high priority as assigned to the needs of male small-scale entrepreneurs.
The importance (or lack or importance) of access to water resources for womens informal-sector activities raises a set of interesting research questions, which fall outside the usual concerns of both donors and governments.
Water-resource planners, governments, and donors all seem to share the view that water associated with direct economic production (agriculture or industry) has a higher value than water used for domestic purposes because there is a possibility of greater income accruing to the users as a result of the former. This approach reinforces the tendency to negate or devalue womens work and economic contributions - it assumes implicitly that the time women spend in fetching water for domestic use is not productive time. In other words, the social reproduction activities of women are assigned no economic value, despite the fact that if women did not perform such services, families would have to pay substantially to purchase them in the open market. Of course, this is equally true in other regions of the world, but in the African context it takes on a particular urgency because of the tendency for reproductive and productive work to merge both inside and outside the household.
Development economists and planners have resisted any suggestion that womens unpaid labour for family survival should be factored into national accounting systems, and there seems to be an unspoken agreement that such work is a natural, necessary, and unquantifiable part of the human cycle of reproduction. Economic valuation of womens reproductive work would somehow imply a denigration of core human values. It is beyond the scope of this paper to fully explore this issue; however, it is relevant to point out that most of the tasks performed by men are, in contrast, assigned an economic value. It is unlikely that African economists will begin to factor womens reproductive labour into national accounting systems in the near future, but there is certainly a strong argument to be made, at the microlevel, for sharing reproductive tasks between women and men and, at the macrolevel, for economists giving womens economic aspirations and contributions a priority equal to that of mens.
Areas for further research
A number of areas for further research have emerged. It is clear that there is need for further empirical studies about actual household patterns of decision-making related to water use. To what extent do womens attitudes and perceptions differ from those of men? Are there significant gender differences in attitudes toward paying for water? Do power relations between men and women within the household make it less likely that women will argue strongly for use of cash reserves to pay for water?
A second key area for research concerns water-project decision-making at the macrolevel. To what extent do planners recognize that the perspectives and needs of women and men may differ and that the needs of women go beyond their reproductive tasks and have an economic value? To what extent are microneeds taken into account? There is a need to develop effective methodologies that capture microneeds of various categories of users (including women) in such a way that this information can be effectively used by water-resource planners.
A third important set of issues relates to the cultural biases in gender-role expectations and attitudes toward water access and management. To what extent do tradition and culture circumscribe the capacity of women to make effective use of available water resources for both reproductive and productive needs? What possibilities exist for breaking down some of the existing cultural barriers?
A fourth area for further research relates to an examination of policy and legal instruments that discriminate against womens full access to water resources. How can laws be changed to ensure that women are not excluded from irrigation projects? What are the water needs of informal-sector entrepreneurs?
The analysis presented suggests strongly that development work in the area of water-resource management in Africa has tended to build on traditional views about the roles of women. Although there has been increased emphasis on ensuring that women receive training in water-pump repair and maintenance and in organizing womens groups to manage village water systems, priorities have been set with the assumption that womens strategic interests lie primarily in the fulfillment of their reproductive roles. Donors have played an important part in reinforcing such stereotypes about womens needs, interests, and priorities. There has been continuing emphasis on the water needs of the family and on the maintenance of adequate sanitation to ensure family health.
Few analysts have argued that it makes good economic sense, from an efficiency perspective, to ensure that female farmers and small-scale entrepreneurs have the same access to water as male farmers. By minimizing the importance of women in economic production, water-resource planners have effectively removed a potential source of conflict. Competition over water resources in Africa is already intense, and by viewing the water needs of male and female farmers as essentially homogeneous and by accepting the role of men as spokespeople for the entire community, donors and government planners have reduced the number of actors who have a stake in decision-making related to water-resource management. It is not suggested here that this is being done intentionally but that it is a predictable outcome of the kinds of traditional assumptions about the role of women still held by donors and government planners. Is this significant? Would it actually make a difference if gender were routinely factored into water projects?
The literature suggests that in some cases women have sabotaged irrigation schemes because they felt that their interests were not adequately represented. They felt neglected or disenfranchised when land over which they formerly had control was appropriated for irrigation projects. Inevitably, it is men who have the greatest voice on irrigation-water committees and women who are expected to provide labour to ensure the success of their husbands farms. It would seem that women do employ resistance strategies if they feel that their interests are being overlooked, but clearly this is an area that requires further empirical research.
Again, it would appear to be self-evident that in some cases, water will play a more critical role in the success or failure of a small business than in others. The most significant point to be emphasized here is that womens work must be taken seriously by water-resource managers. Womens work, both inside and outside the household, is essential to family maintenance.
Integrating gender issues into water projects in a systematic way will involve a rethinking of social structures and organizations. Planners will have to go beyond a vaguely defined concept of community and commission baseline studies that reveal actual decision-making patterns and access to resources within social groups. It is the tension between the needs and priorities of these different actors that forms the basis of gender analysis. In essence, this will involve a rethinking of existing social structures and organizations. But to do otherwise would be to perpetuate approaches that fail to take into account the needs of all members of the community. It is not enough to integrate womens concerns. They must become a central part of the water agenda (Yoon 1991), with the same legitimacy and urgency as the concerns of any other major group of users. It is necessary to recognize, of course, that women as a group are far from being homogeneous. Like men, they are divided by social class, by ethnic group, and by differential access to resources. Consequently, it would be a major error to assume that there will be solidarity in womens views and, indeed, experiences concerning water-resource management.
Bernal, V. 1988. Losing ground: women and agriculture in Sudans irrigated schemes: lessons from a Blue Nile village. In Davison, J. ed., Agriculture, women and land. Westview, Boulder, CO.
Biswas, A.K. 1993. Water for agricultural development: opportunities and constraints. Water Resources and Development, 9(1), 3-12.
Brown, E.P.; Nooter, R. 1992. Successful small-scale irrigation in the Sahel. World Bank, Washington, DC. World Bank Technical Paper 171.
Carney, J.A. 1988. Struggles over land and crops in an irrigated rice system in the Gambia. In Davison, J. ed., Agriculture, women and land. Westview, Boulder, CO.
Chabayanzara, E. 1994. Small holder irrigation development: impact on productivity, food production, income and employment. In Steven, A.B., ed., Issues in African rural development 2. Winrock International, Arlington, VA.
Dankelman, I.; Davidson, J. 1988. Women and environment in the Third World: alliance for the future. Earthscan Publications, London, UK.
Diemer, G.; Vincent, L. 1992. Irrigation in Africa: the failure of collective memory and collective understanding. Development Policy Review, 10(2), 131-154.
El-Katsha, S.; White, A.U. 1989. Women, water and sanitation: household behavioral patterns in two Egyptian villages. Water International, 14, 103-111.
Freeman, D.B. 1993. Survival strategy or business training ground? The significance of urban agriculture for the advancement of women in African cities. African Studies Review, 36(3), 1-22.
Government of Kenya; UNICEF (United Nations International Childrens Emergency Fund). 1992. Children and women in Kenya: a situation analysis. Government of Kenya and UNICEF, Nairobi, Kenya.
Green, C.; Baden, S. 1995. Integrated water resources management: a gender perspective. IDS Bulletin, 26(1), 92-100.
Harnmeijer, J. 1993. Selecting villages for impact in water supplies: an expanded role for the baseline study. Health Policy and Planning, 8(3), 224-231.
Hulsebosch, J. 1993. Evaluation of measures to enhance a gender balance in water users associations: experiences in the Southwest Kano Project in the Kano Plains in Kenya. Report to the Small-holder Irrigation and Drainage Project, Ministry of Agriculture, Nairobi, Kenya.
INSTRAW (United Nations International Research and Training Institute for the Advancement of Women). 1987a. Women, water supply and sanitation. A national training seminar, 9-13 Nov. 1987, Nairobi, Kenya.
INSTRAW (United Nations International Research and Training Institute for the Advancement of Women). 1987b. Women, water supply and sanitation. A national training seminar, 23-28 Nov. 1987, Addis Ababa, Ethiopia.
Jayseela, F. 1993. An ethical diversion? The Lesotho Highlands Water Project. Droplets, Newsletter of the African Water Network, 3.
Jiggins, J. 1989. How poor women earn income in sub-Saharan Africa and what works against them. World Development, 17(7), 953-963.
Kendie, S.B. 1994. Survey of water use behaviour in rural north Ghana. In Jacqueline, S. ed., Women, health and the environment: an anthology. WHO, Geneva, Switzerland.
Kyomuhendo, G.B. 1992. The role of women in petty commodity production and commerce: a case study of rural peasant women in Uganda. Eastern Africa Social Science Research Review, 8(2), 1-19.
Lewis, N. 1994. Safe womanhood: a discussion paper. International Federation of Institutes for Advanced Study, Toronto, ON.
McCauley, A.P.; West, S.; Lynch, M. 1992. Household decisions among the Gogo people of Tanzania: determining the roles of men, women and the community in implementing a trachoma prevention program. Social Science and Medicine, 34(7), 817-824.
McPhail, A.A. 1993. The five-percent rule for improved water service: can households afford more? World Development, 21(6), 963-973.
Narayan-Parker, D. 1988. People, pumps and agencies. The South Coast Handpump Project. PROWWESS/UNDP, New York, NY.
Ongwen, O. 1993. Women and community participation. Droplets, Newsletter of the African Water Network, 3.
Postel, S. 1993. Water scarcity spreading. In Brown, L.; Kane, H; Ayres, E., ed., Vital signs: the trends that are shaping the future. The World Watch Institute, New York, NY.
Simpson-Herbert, M. 1992. Women and water. World Health (July-Aug.).
Simpson-Herbert, M. 1994. Gender and management issues in the water sector: the case of rural Botswana. African Water Network Newsletter, 1.
Waters, T. 1992. Lifeworld and system: of water systems and grain mill development in rural Tanzania. African Studies Review, 35(2), 35-54.
White, G.F.; Bradley, D.J.; White, A.U. 1972. Drawers of water: domestic water use in East Africa. University of Chicago, Chicago, IL.
Whittington, D.; Donald, T.L.; Kyeongae, C.; Jeffrey, A.H.; Venkateswarlu, S. 1993. Household sanitation in Kumasi, Ghana: a description of current practices, attitudes, and perceptions. World Development, 21(5), 733-748.
Whittington, D.; Lauria, D.T; Mu, X. 1991. A study of water vending and willingness to pay for water in Onitsha, Nigeria. World Development, 19(2/3), 179-198.
World Resources Institute; UNEP (United Nations Environment Program); UNDP (United Nations Development Programme). 1994. World resources 1994-95. Oxford University Press, New York, NY.
Yoon, S. 1991. Water for life. Paper presented at the Symposium on the Impact of Environmental Degradation and Poverty on Women and Children, 27-30 May 1991, Geneva, Switzerland.
Zwarteveen, M.Z. 1994. Gender issues, water issues: a gender perspective to irrigation management. International Irrigation Management Institute, Colombo, Sri Lanka, Working Paper 32.