Article Crafting in the context of AIDS and rural poverty: a livelihood strategy with prospects Tessa Marcus In South Africa, development initiatives in rural areas that have been directed at Africans have centred almost exclusively on agriculture. This legacy continues into current policy. The development component of land reform, made more explicit and forceful in the second term of the new democracy, has as its core the notion of successful emergent black farmers. The focus on agriculture persists despite the fact of advanced deagrarianisation (Bryceson 1996, Marcus et al 1996) and persistent, historically structured inequalities in land, labour and capital (Adams at al 1999, Mclntosh and Vaughan 1999, Marcus 1989, May 1998). Moreover the major component of the rural economy, commercial agriculture, makes a small contribution to the gross domestic product and is a declining source of occupation, with farm work accounting for less than ten per cent of those working in South Africa. Together, the general absence of small or micro agriculture, limited agricultural employment and widespread low wages in the sector means that relentless poverty affects about half the households and two-thirds of the rural population in South Africa (May 1998). In such a context, the stimulation of non-agricultural economic activity is essential to rural livelihoods. The inherited problems of rural social inequality and the failure of agriculturally focused development to provide livelihoods for the majority of the rural population are compounded by the AIDS pandemic. Africa, particularly sub-Saharan Africa, is at the epicentre of what is a global epidemic. An estimated 13 million Africans have died in the two decades since the start of the epidemic and 23 million people or 70 per cent of all adults and children presently living with HIV are found in this region (UNAIDS/WHO 1999). In South Africa, HIV and AIDS is a generalised, heterosexual epidemic. The tenth national survey of HIV sero-prevalence TRANSFORMATION 44 (2000) ISSN 0258-7696 Tessa Marcus among women attending government antenatal clinics in South Africa conducted in 1999 found an HIV prevalence rate of 22 A per cent. This suggests that at least four million people were HIV positive in 1999, up from an estimated 3.6 million in 1998 (Crisp and Ntuli 1999). The epidemic is concentrated among young adults, with half of those infected in 1999 being in their twenties. Gender differences in infection result in greater infection rates among young women. In 15 population-based studies of HIV prevalence in nine African countries, on average 12 to 13 African women were found to be infected for every ten African men (UNAIDS/ WHO 1999). A study of HIV prevalence among voluntary blood donors over the period 1990-96 showed HIV to be about ten times more common among Africans than whites and in 1996, for every eight African men infected there were 13 African women who were HIV positive. For whites the prevalence of HIV for men and women was 1.27:0.8 (Taylor 1998:53). Geographically, the available data suggests provincial and urban/rural variations in the distribution of the epidemic in South Africa. With respect to urban and rural differences, available data suggests that, although HIV is somewhat more prevalent in urban areas, the current rate at which the epidemic is progressing is as rapid in rural areas as it is in urban areas (Taylor 1998:52). Providing land grants, be they simply to secure tenure or to stimulate farming, fails to take account of AIDS as a development issue that adversely impacts on capital and labour. Policy formulation has blithely ignored the scale of the epidemic or its social consequences, particularly tne loss and diversion of resources as well as the drain on labour power of terminal illness. Yet evidence shows that these AIDS-derived effects have adversely affected small, medium and even large-scale agricultural producers in multiple ways. In agriculturally-based communities in east, central and west Africa, a review of rural household responses to AIDS finds that the epidemic has contributed to an exodus from the land, especially of young people. It has led to the sale of disposable assets, including implements and livestock essential for agricultural production, as well as an increase in informal borrowing. It has caused interruptions in, as well as a depletion of, labour supply and the increased use of hired or borrowed labour. It has forced reductions in the area under cultivation, encouraged crop substitution and the extension of the working day. AIDS has also meant the pursuit of non-agricultural employment and livelihoods as well as a reduction in household food consumption (Mutangadura et al 1999:19-23). 18 Crafting in the context of AIDS and rural poverty Moreover, the AIDS pandemic has impacted on household size leading to the expansion, contraction or disintegration of households. It has forced a change in household composition, severely weakening and often breaking the young adult nexus between generations. This, in turn, exacerbates an already existing social crisis of care (Harber 1999, Marcus 2000a) which can only worsen as the pandemic progresses from HIV to AIDS. It is a social context that is unlikely to withstand the sheer weight of need that AIDS related deaths generate (Marcus 2000b) and many, especially children and the aged, face economic and social destitution. Given a fatal disease syndrome that particularly affects adults in their most economically active years, the pursuit of agricultural revivalism as a livelihood strategy seems to offer bleak prospects to the majority of rural people who are infected or affected by AIDS. So where does that leave people caught in the pincer of historical legacy and contemporary pandemic? Part of the answer must lie in the stimulation of non-agricultural employment or livelihood activities that can help people meet their economic needs and simultaneously contribute to their social wellbeing. But doing what? In southern Africa, one potential answer lies in the stimulation and promotion of an already existing activity: crafting. It is a source of livelihood for rurally based people in the region, particularly women. Crafting is held to have few barriers to entry, is innovative and adaptive to changing economic and social conditions (DACST 1998). Crafting is also regarded as a route through which the history and culture of individuals, their kin and communities can be kept alive and vibrant. Most people involved in crafting do so as part of the so-called informal sector, defined by StatsSA as 'unrecorded economic activities' (DACST 1998:5). It is a sector in which people work with limited resources, information and reward often in relative isolation from one another. Within new government policies elaborated over the last five years, crafting falls within cross-cutting foci. It is part of a development vision that centres on the stimulation and expansion of tourism as an economic sector with prospects. It also fits with the idea of stimulating economic activity across all sectors by widening the range in enterprise scale through the stimulation of small, medium and micro-enterprises. And it is compatible with the planning notion of local economic development. However, the assumptions about the potential of each of these notions are informed by a weak knowledge base regarding crafting. Little is known about the extent of crafting as a source of occupation and income or how skills are acquired 19 Tessa Marcus and transferred across generations. In turn, this means that the conditions needed to stimulate production and encourage skill and knowledge transfer are often poorly understood. These notions also fail to take into account the persistent impacts of present and past economic and social relations, assuming that changes will flow automatically irrespective of a deeply stratified rural society. To explore the potential of craft as a survival, development and regenerative strategy in the context of 'living with AIDS', this article sets out key findings of a pilot study of crafting in the KwaZulu-Natal Midlands. The study site Defining craft and crafting is a vexed issue. This study uses a functional, working definition of craft as the production of a broad range of utilitarian and decorative items manufactured on a small scale with hand processes being part of the value added content. The production of goods utilises a range of synthetic and natural materials. (DACST 1998:8) The development context of the study site area is one of racial segregation where the economy has historically centred on white-owned and run commercial agriculture and forestry, agro-industrial manufacturing and mono-industrial nodes (textiles and motor parts). The majority African population has endured low wages, landlessness, high rates of unemployment, geophysical segregation and generalised poverty. A conjuncture of forces over the past decade, particularly weather and environmental conditions, global market forces and the withdrawal of state buffer, has seen a decline in production and employment in agriculture and industry, a burgeoning, white-driven tourist industry (the Midlands Meander) and the setting in of HIV, TB and now AIDS. The study was conducted at four rural sites. Tendele is a small settlement located on privately owned land where most households rent from the three landlords. The KwaMkhize and eMahlutshini communities live on traditional authority land in Hlatikulu, which lies close to Giant's Castle Nature Reserve. The farm worker 'community' is drawn from white- owned commercial farms in Lidgetton and the Dargle Valley. Mphophomeni is a rurally located township created in the 1960s to accommodate African people removed from the Midmar Dam basin as well as those subsequently excluded from residence in the town of Howick. These communities, like many others, lie in the shadow of the Midlands Meander which is the largest tourist route of its kind in South Africa, involving 113 enterprises and 20 Crafting in the context of AIDS and rural poverty attracting some R5 million annually into the area (Mathfield 2000:10). The communities reside at the back of buildings and artefacts of commercial agriculture, in displaced urban settlements or on crowded patches of land created by the 1913 and 1936 Black land Acts, forced removals and other segregationist Apartheid policies. (Mathfield 2000:5) Methodology The study findings are based on two structured surveys administered in face to face interviews. An initial screening survey was used to identify households in which there were crafters. The extent of the screening exercise by site and sex of respondents is set out in the table below. Table 1: Screened households by location and sex of respondent Place Male % Female % Total Total 504 (35) 931 (65) 1435 Mphophomeni 380 (37) 622 (61) 1002 Tendele 30 (30) 69 (70) 99 Hlalikhulu 54 (23) 184 (77) 238 Commercial Farms 40 (42) 56 (58) 96 All households in Ward 2 and Ward 3, as well as some households in Ward 4 at Mphophomeni were screened, yielding a total of 1002 household respondents. This is estimated to be approximately a third of the total number of households at Mphophomeni and yields results that can confidently be used as a reflection of population values. At Tendele, all 99 households were screened and the findings reflect the population. In Hlatikhulu (at KwaMkhize and Emahlutshini) approximately SO households on either side of the road were screened at each site - ten across and five deep. As the settlement is fairly dispersed, this process made it possible to include people who were both close to and distant from the road. Given the size of the population in Hlatikhulu, at best the data is indicative of crafting in the area. Ninety six farm worker households were screened on 13 farms in Lidgetton and Dargle Valley. Problems of access to workers in commercial agriculture make it impossible to screen all households in a given locality. The findings should therefore be treated as indicative rather than reflective of farm worker households in the KwaZulu-Natal Midlands. The second stage of data collection involved returning to households where members were identified as making or selling craft and interviewing 21 Tessa Marcus crafters in depth, using structured questionnaires. In all, 284 crafters participated in the survey. The break down by area is set out in the table below. Table 1 Interviews with crafters by site relative to crafter households identified through screening Place No. of crafters No. of crafters Proportion identified interviewed interviewed of through screening in depth those screened Mphophomeni 354 190 54% Hatikhulu 185 53 27% Tendele 34 34 100% Commercial Farms 17 9 53% Total 590 284 48% In terms of the in-depth interviews, a little under half (48 per cent) of the study population was interviewed. Statistically, the findings for Mphophomeni and Tendele reflect the identified crafter population. The findings for the commercial farms and Hlatikhulu should be regarded as indicative of crafters at the two sites of study. Results The findings from the screening set out the incidence of crafting in a context of household size and sources of income. The data is analysed by location, since each site was chosen to reflect different social and economic contexts. The diagram below compares family size across communities. It shows that at all sites, most households are large, with between six and ten members. The important differences across the sites are at the extremes. Generally, between a quarter and a third of the households living on commercial farms and in the rural township have with five or fewer members, while about a quarter of households living in traditional authority rural areas have very large households of 11 or more members. Some of the explanation for large and extremely large households lies in the social practice of sending children born out of marriage to the mother or father's family of origin. In part it also reflects the fact that many are already caring for orphans. Thus, close to two fifths of the households in Hlatikhulu and Tendele report caring for orphans, while 14 per cent of households on commercial farms have orphans. At least half the households with orphans reported caring for more than one parentless child. 22 Crafting in the context of AIDS and rural poverty Comparing Family Size Across Communities Mphophom«ni Farms T«ndel» Hlatikhulu • 6-10 m*mb*rt D 11+m«inb«r« At all sites, most formal sector employment is in the locality, in the Midlands region and in the province, KwaZulu-Natal. Only at Hlatikhulu and Tendele does employment outside the province seem to contribute to household income, and then for only a small number of households. Excluding the commercial farms, most households screened at the three other sites report having one earner. Only approximately one quarter report two or more earners, while an average of 26 per cent of households say they have no family member in employment. On the farms, the number of households with multiple earners is considerably greater than those with only one, while the proportion of households without workers drops to ten per cent. Both these differences reflect the conditional link between work and residence that typifies commercial agriculture. Number of Earners by Household 0 Two plus D One earner • No earners Hlatikhulu Mpltophorneni When respondents talk about people who bring money into the household they are referring to people in employment or who have regular sources of income from pensions or trading etc, without including informal sector contributions. Yet, with the exception of farm worker communities, a 23 Tessa Marcus majority of households at each site has somebody involved in informal sector activity as the chart below shows. Percentage of Households in Informal Sector Activity If households are analysed by their reported informal sector activity, then those who report no earnings or even one or more earners tend to under-report income sources. Thus, for example, in Hlatikhulu 90 per cent of households that report being without a person who brings money home have somebody who is involved in some form of income generating, informal sector activity. This is the case in 61 per cent of such households at Tendele and over half of such households at Mphophomeni (52 per cent). Only on commercial farms does the proportion drop significantly to two out of the nine households without earners. This finding highlights three issues. Firstly, at all sites there are households which have no visible source of income, however small and irregular. Secondly, the failure of households to account for informal sector activity suggests that as a source of income, it may well be unreliable, irregular and often very small - something that can't be counted on. Thirdly, the incomes generated by formal sector employment are insufficient to meet basic household heeds for a range of reasofls, including the type and regularity of formal sector work, skill levels and household dependency ratios. Informal sector activities are common to households at all the sites, except on commercial farms. The chart below sets out the kinds of activities undertaken across the communities. The universal and most common activity is crafting. Making and trading in clothes and selling foodstuffs are also common to all sites, although these are carried out in fewer households. Providing services, particularly hairdressing, photography and repairing appliances or vehicles is done at Mphophomeni, Hlatikhulu and Tendele but not by people in households living on commercial farms. At 24 Crafting in the context of AIDS and rural poverty Mphophomeni and among the farm-worker households, building and construction are also mentioned as informal, micro or small enterprise activities, while they are not reported by household respondents at the two other sites. This is either an oversight or reflects that people mostly construct and maintain their homes themselves and do not need to buy in such skills. Generally, the greatest spectrum of services, including artisanal services, are reported at Mphophomeni, reflecting the urban industrial background and greater range in skills of people living in the township. Comparing Informal Sector Activities • 1 Tendele Farms Hlatikhulu Mphophomeni HI•ttili HWI^IWSi ^ ^ ^ ^ E • Clothe t a Food OS*rvl»s • Building In Short, the screening survey shows the extensiveness of crafting as a practice, even though there are households at all sites, and especially on commercial farms, that make craft but do not actually trade their artefacts. It also shows crafting to be a major component of informal sector activity and integral to rural livelihood strategies in the study area. The underlying craft base is predominantly female. Eighty eight per cent of crafters identified and subsequently interviewed were women. Crafting is largely an occupational activity of middle-aged adults. Notably though, in Mphophomeni half the crafters are young adults, as are fairly substantial minorities at Tendele and on commercial farms. No crafters interviewed on commercial farms were over 60 years of age, reflecting the demographic composition of farm worker households, the small number of crafters in the sector and the small size of the sample. In all the communities most people interviewed have been crafting for ten years or less, with the exception of Hlatikhulu where just under half began crafting in the past decade and a little over half have been crafting for 11 years or more. Other than on the farms, there is also a core of crafters who have been involved in the sector for over 20 years. 25 Tessa Marcus While there is the obviously expected association between age and length of crafting, at three of the sites the majority who have taken up crafting in the past decade are middle-aged or pensioners. The exception is in Mphophomeni, where most new crafters are young. This said, even there middle and old aged adults make up a fairly substantial proportion of 'new entrants' into the sector. In other words, the data suggests that many people in the study communities have only begun to craft in the past decade, and that many of these are middle aged women. This trend is likely to reflect two currents. One is the decline in formal sector employment, low wages earned by women in the formal sector and increasing economic pressures on households. The other is an expansion of the craft market with better opportunities and income potential. Looking at the kinds of crafts produced, how people acquire their skills, their levels of performance and skills transfer, the data shows that the predominant form of crafting at all the study sites involves sewing - including knitting, crocheting, as well as embroidery - and weaving. Bead work, pottery, wood carving, working with metal, painting or thatching also are done at most sites, albeit by a small number of crafters. Within these broad categories there is no evident gender bias, except with respect to art, painting and thatching. Only men seem to be involved with these crafts where they are practised. 1_1LI 1 Si • HalkhJu K^roprcriBni Commeiolal firms | •Swing • Mains OBeod OCtw| TenJele IVpesofQaft 26 Crafting in the context of AIDS and rural poverty The role of family, particularly mothers and grandmothers, in teaching crafters their skills is evident at all the study sites. Kin is particularly important in the traditional authority areas of Hlatikhulu and Tendele. However, on the farms, friends and neighbours as well as the local school or club are equally as important as kin, while at Mphophomeni, schools and other institutions are as important as family in teaching crafters their skills. Also, at three of the four sites, there are crafters who are self-taught. Where Crafters Learn their Skills Mphophonwnl Commardal Farms • Family • Frtontfs/nalghbours • School/dub a 3*IMaught The majority of crafters at all four sites work on their own. Yet, at all sites there are crafters who club together to make articles, share skills and sometimes share resources. Group work is strongest at Tendele (40 per cent) and weakest at Mphophomeni (14 per cent) with crafters at Hlatikhulu (25 per cent) and on the commercial farms (33 per cent) falling in between. At all sites there are crafters who already teach others their skills. There are also those who would like to teach and many would like further training. Generally, the present extent of transmission of skills varies across the four study sites. While it is largely directed towards children, it is more the exception than the rule. Most crafters (between 60 and 80 per cent) do not teach their children their skills but where they do, they invariably teach girls rather than boys. Crafters give similar sets of reasons for choosing to teach their children or not. Those who teach, mostly do so because their children show an interest. They also see it as a way of helping them to earn an income, now and in the future, of keeping them busy and preparing girls for marriage. The reasons respondents give for not teaching children is lack of interest (especially among boys), schooling, not having the time or materials for them to practice with, and not feeling confident enough in their own skills and capacity to teach others. 27 Tessa Marcus These findings suggest that while the family's role in transmission of crafting skills is important, it simply cannot be assumed. On the one hand, there is a need to support and encourage family members to transfer their skills. On the other hand, other institutions, particularly schools, are important sites for learning craft skills and they should be seen as pivotal in teaching the organisation and management of production, distribution and sales. Generally, labour expended on craft largely goes unremarked. It is regarded as a free resource by most crafters who mainly depend on their own labour or that of one or more of their children. No crafters hire labour to make articles although some of them are commissioned to produce and are paid for their labour time. While crafters generally do not incur direct labour costs, there are some indirect costs that affect production. People invariably juggle their craft work around other demands on their labour time, particularly caring for children, housework and sometimes waged work. Similarly, workspace and storage largely go unremarked. Nearly all crafters work from home and store materials or finished products in containers in or outside their houses, the exception being in Mphophomeni where a few crafters have workshops. Crafters generally do not incur any direct costs for workspace or storage. Indirectly, cramped living conditions and a lack of appropriate storage may affect both the volume and the quality of goods that crafters produce. Some crafters at Mphophomeni and Hlatikhulu articulate these problems when they talk about the need for a community centre where they can work, share skills and be trained. Many crafters work with low level technology. Others do not have, or borrow, equipment because they cannot afford to buy the tools or machines they need for themselves. Both these aspects affect the range, volume and quality of articles they produce, more often than not negatively rather than positively. Crafters report materials as a cost. The majority of crafters buy the materials they use. A small proportion (16 per cent) collects material from fields or rivers, mostly in the rural areas or from urban or industrial waste sites, specifically at Mphophomeni. Many incur transport costs to buy or collect materials. A few crafters travel as far afield as Durban or the Free State to get the materials they use. At all sites, however, most purchases are made from farms and at stores in the main towns and villages of the KwaZulu-Natal Midlands. Some people 28 Crafting in the context of AIDS and rural poverty buy from traders where they live. They, and some of those who harvest materials, invariably walk carrying loads on their heads or by wheelbarrow. The vast majority of crafters, however, uses motorised transport to carry purchases. With the exception of the few who have private cars or who hire vehicles, they rely almost entirely on the mini-bus taxi industry. While crafters report a wide range of expenditure on transport and materials (from nothing to thousands of rands), on average crafters spend a little under R150 a purchase trip. Output, strictly speaking, does not lend itself to comparison, given the variety of activities and types of goods produced that fall under the rubric of craft in the study area. This said, most crafters report a relatively low volume of output and the majority would like to be able to produce more so as to increase their income. Markets and sales Critical aspects that underlie the real and potential benefits of crafting to poor people and their families include where crafters sell their goods, the costs entailed in sales and the actual income generated by crafting. At all sites there are people who make but do not sell the craft items they make. Essentially craft is undertaken to produce things of use or beautification for domestic consumption and is a way of keeping occupied. The majority both makes and sells craft, except in Mphophomeni where a handful of people sell, usually on behalf of their family members, but do not make. Marketing is mostly undertaken by crafters themselves, as the chart below shows. However, at all sites there are crafters who make for buyers, ie people who bring materials to their homes and fetch finished products, and a few make while family members or non-kin sell on their behalf. Who Markets Craft Commercial Farms D 3 Mphophomeni •1 Tendele J3|•9B I B 2 1 Hlalikhulu SI12 ( 20 40 60 ao 100 • Crafter D Kin or Others • Buyers | 29 Tessa Marcus The primary marketing site for most crafters is within their communities. Just under half of all the crafters who make and sell (49 per cent) also market their artefacts outside their communities. Mostly they sell their wares in the KwaZulu-Natal Midland villages and towns, with a few selling outside the province. Importantly, crafters in the study sites generally do not supply, or have access to, the established tourist and hospitality outlets on the Midlands Meander, with the possible exception of those who supply buyers. Selling outside the community usually entails further costs except where point of sale is within walking distance and the goods can be carried. Here, once again, the problems crafters face are transport related. Crafters have to pay for their passage as well as the carrying of their goods. In addition, bad weather, poor road and vehicle conditions, as well as the reluctance of mini-bus services to carry what they make all impact negatively on their efforts to find markets. People recognise the limits of demand within communities. They are also influenced in what they produce for sale by what is feasible to carry under existing transport conditions and to sell outdoors. Both these factors shape crafter decisions about production and marketing. Given existing constraints in skills, resources and markets, the question is what kinds of gross incomes are people able to generate? Income generated by craft ranges from a few rand to thousands of rands a month, with average earnings at three of the four sites ranging from R278 to R.380 a month. Comparing Hlatikhulu, Tendele and Mphophomeni, 63 per cent of respondents at all the study sites generate gross incomes of R250 or less a month through craft sales. A fifth or more are able to earn between R250 and R500, while nine per cent report earning over R1000 a month. In absolute terms under present conditions most crafters earn relatively small amounts from craft sales. Relatively speaking, however, the value of crafting as a livelihood strategy cannot be underestimated for several reasons. Firstly, as small as the majority of these incomes are, they compare quite favourably with income that most women can generate in formal employment in the existing labour market. Although contributions to crafter households from full-time work tend towards the middle range (R300 - R600) or above, the number of households with such contributors are few. Such incomes are rarely the case for part-time employment or even full-time employment in commercial agriculture or domestic service. Secondly, while state grants, especially pensions, also fall within this middle range category, many crafter households (75 per cent at 30 Crafting in the context of AIDS and rural poverty Mphophomeni, 62 per cent at Tendele and 64 per cent at Hlatikhulu) do not have pensioners or people getting grants. Those that do, invariably support numerous dependants, especially in the rural areas. For them crafting brings in much needed additional income. Thirdly, for between a fifth and a third of crafters at the three main sites of the study, craft sales are their main source of income. For a further 21 to 34 per cent it is regarded as an important source of money that adds to household income. Finally, although most craft-derived income is concentrated at the lower end of the income range, even under present conditions crafting provides some people with incomes that equal or even exceed the middle and top end of contributions made to households by formal sector employment, other informal sector activities and state grants. The very fact that some respondents earn R1000 or more a month from craft suggests that, if production and market conditions were improved, access to level earnings in the middle to upper range of existing household incomes could be increased substantially. The money craft brings into poor households is seen by many respondents as a way of caring for their children both now and in the future, as well as of keeping occupied. Several respondents at all sites see craft as a way of teaching children about their culture and history, the better to equip them when they are older. In terms of their desire to produce more and better craft, most respondents mention the need for resources - money, materials, equipment and the creation of physical sites at which to share skills and knowledge, train, produce and trade. They also call for training and skill development both in production and marketing. Conclusions The screening survey reveals the role of craft in the livelihood strategies of the study populations. It is seen to be an important, often primary, non- formal sector income generating activity in many poor households. This study shows that crafting encompasses a range of production conditions. The majority of people in crafting are producer-sellers: women who work from home using their own labour and capital to produce for a restricted and limited market. This said, with the exception of households living on the commercial farms, there are some crafters at all the sites whose capital resource base is such that they are able to produce and sell on a scale that generates incomes comparable to those commanded from 31 Tessa Marcus formal sector employment. This is possible because of a number of factors. These include multiple household income sources or external inputs that allow for greater investment in craft production, a pursuit of markets beyond the immediate locality, as well as reduced domestic and other demands on their labour time. Changes in production conditions can significantly influence craft- derived incomes. The fact that there are few barriers to entry into craft production should be treated as a positive starting point. But it is not enough. Efforts need to be made to support, improve and enhance production and income. Access to small amounts of capital that allow people to buy more materials, or improve the technology they use, can have a significant impact on the income individual crafters are able to generate. Low risk micro credit that is interest capped can contribute fundamentally to sustained horizontal and vertical expansion in production and income for many poor households. At the same time, protracted and repeated ill-health that is typical of AIDS is likely to impact negatively on income levels, as illness and the demands of care-giving eat into the available labour capacity and time of crafters - be they ill themselves or caring for sick dependants. They are likely to have to reduce craft production and they may be forced to withdraw from any form of expanded marketing of the craft they make. Under these conditions, the need to stimulate some form of group purchasing and marketing seems to be imperative. Not only is it more likely to meet the needs of those presently in production, but it will create the conditions within communities for children to enter crafting, when they need to take up where their adult caregivers have left off. This is crucial to the realisation of any strategy around home-based and community-based care for orphans. Two critical issues for home-based workers are their levels of pay for unit-artefacts produced, as well as the need to ensure continuity of access to this source of livelihood where they have dependants, in the event of their own illness and death. Crafters themselves, have to be made aware of the income generating opportunities that their skills can command. They also need actively to work at transferring skills to other household members - especially their children - as a replacement, should conditions dictate. One of the constraints on present conditions of craft production and marketing is skills and training. Presently, craft production in poor households is beset by a lack of variety, variability in quality of output, 32 Crafting in the context of AIDS and rural poverty narrow markets that are spatially and materially limited, and limited production flexibility. These limitations need to be overcome to enhance the income generating capacity of the sector. They can be overcome through a systematic training initiative, which both enhances and elaborates existing skills and creates new ones where these are needed. This is critical for the new entrants into the sector, most of whom are middle-aged or older women. Three conditions are likely to typify their circumstances - limited or no access to formal-sector employment, except in low-income manual labour, responsibility for grandchildren, dependent adult children as well as the ill and dying, and limited or no access to state supports. Their crafting is a critical strategy on which family survival often hinges. Investing training and skills in them, as well as in the new young adult entrants into craft, is likely to have immediate and long-term social and economic benefits that are essential to development at community level. The heart of this training, particularly in craft production, can be resourced from the core of long established crafter capacity that exists within all the communities, with the exception of those on commercial farms. It is important to recognise that this skill is not always family based and is often not transferred from adults to children where it exists. Under these circumstances, a multi-pronged strategy is needed that encourages crafters to teach their children where this is possible and children are willing and interested. At the same time, learning and skill-transfer opportunities outside the family - in clubs, schools and other community- based sites - are already important in some communities and need to be created more generally. These can meet production and training needs where they are not adequately met by kin and they can address the more general need for procurement and marketing skills. Clubs, schools, and development centres are more likely to depend on resource people and institutions external to the specific communities concerned creating an opportunity for better placed crafters and marketing specialists to share their knowledge and Skills. Generally, the importance of familial and formal training and education is likely to grow in the context of AIDS. This study presents strong evidence of two inherent and related ironies. The first is the marginalisation of crafters who live in the shadow of what is a thriving tourist and hospitality initiative: the Midlands Meander. They are invisible as a skills and resource base that is integral to the locality, they have little or no access to existing outlets and establishments and with a few 33 Tessa Marcus notable exceptions, nothing has been put into their development. The second is the extent to which African crafters at the study sites contribute to the Midlands economy, through their purchase of materials and use of transport services especially. There is no escaping the fact that a strategy to improve their income generating capacity will positively affect the area's economy. There is also no doubt of how formal sector production and services are linked to, and rely on, informal sector economic activity at the local level. This study challenges development orthodoxies for rural areas, which for too long have hinged on expanded access to agricultural production, either as direct producers or as workers. A rurally located town like Mphophomeni has long cut most of its connections to agricultural labour. The poverty of communities at other sites is due, in large measure, to their dependence on agro-industrial employment and the impossibility of eking much out of the land they access. The flaw in assumptions, given existent, historically established production relations, is made more fatal by AIDS where the premium on both labour and capital is significantly raised by young adult death, loss and decline in income and skills, and often, of a hold on the land. Craft, by contrast, is a transferable skill that is relatively easy and cheap to acquire. It can be developed in a way that is responsive to changing conditions within the production unit as well as in the larger economy. Also it is not contingent on nature or on access to a large capital asset such as land. Crafting is also compatible, within limits, with the demands of household reproduction, including care giving. Probably, the most important feature of craft is its earning potential and the relatively limited investment that is required to achieve social and economic outcomes for poor rural households that are far greater than they will ever achieve in agriculture. The value of craft as a source of income to poor households is clearly evident from this study. Craft has the potential to stimulate development in rural areas, even under conditions of AIDS. To realise such a goal, minimal conditions of expanded production need to be created. Particularly, state regulated micro credit needs to be made available to crafters and the rural poor for non-agricultural purposes. There is a need for investment in training and skill development, that enhances already existing skill capacity but which targets new entrants, including young people, and that develops new skills especially in areas of identified weakness. Partnerships of learning and opportunity across sectors and localities need to be stimulated and encouraged, beginning with the Midlands Meander initiative. Crafting in the context of AIDS and rural poverty References Adams, M, B Cousins and S Manona (1999)' Land tenure and economic development in rural South Africa: constraints and opportunities'. Paper presented at the 'National Conference on Land and Agrarian Reform in South Africa', Cape Town. Bryceson, C (1996) 'Deagrarianisation and rural employment in Sub-Saharan Africa', World Development 24(1). Crisp, N and A Ntuli (1999) South African Health Review. Durban: Health Systems Trust. DACST (1998) The South African Craft Industry Report. Cultural industries growth strategy report to the Department of Arts, Culture, Science and Technology. Harber, M (1998) 'Social policy implications for the care and welfare of children affected by HIV/AIDS in KwaZulu-Natal'. Unpublished MA dissertation, University of Natal. Marcus, T (1989) Modernising Super-Exploitation: restructuring South African agriculture. London: Zed. (2000a) 'Communities and care giving: a qualitative case study of poor community responses to children in need', Research in Community Sociology X, Fall. living with (2000b) Wo! Zaphela Aids. Pretoria: UNICEF.Izingane: death and dying in the context of Marcus, T, K Eales and W Wildschut (1996) Down to Earth: land demand in the new South Africa. Durban: Indicator Press/LAPC. Mathfield, D (2000) 'Impacts of accommodation and craft based tourism on local economic development: the case of the Midlands Meander'. Unpublished MA dissertation, University of Natal. May, J (1998) Poverty and Inequality: summary report, http://www.gcis.gov.za/ pubserv/1998/pirsum.htm. May 19. Mclntosh, A and A Vaughan (1999) 'Experiences of agrarian reform in South Africa: the limits of intervention'. Paper presented at the 'National Conference on Land and Agrarian7Reform in South Africa'. Mutangadura, G, D Mukurazita and H Jackson (1999) A review of household and community responses to the HIV/AIDS epidemic in the rural areas of sub- Saharan Africa UNAIDS/99.39E (English original) http://www.unaids.org. April 20. Taylor, V (1998) HIV/AIDS and Human Development South Africa 1998. Pretoria: UNAIDS/UNDP. UN AIDS/WHO (1999) AIDS Epidemic Update December 1999. http:// www.unaids.org. May 5. 35