The clothing and textile sector met the same fate in Hammarsdale, the town adjacent to Mpumalanga. The founders of Hammarsdale, the Industrial Development Corporation (IDC), had argued that ‘a well-established textile industry would have tremendous employment potential for semiskilled operatives, which meant that it could raise the standard of living of the Bantu’ (IDC, 1971:57, cited in Young, 1972). The first clothing factory, Hammarsdale Clothing, was established in the area in 1957. Within three years of its establishment, Hammarsdale had three textile factories and a factory that manufactured sewing machines which employed 2 135 people. By the end of 1971, there were 13 factories employing 8 500 workers in the area (Young, 1972). Nearly every second person in Mpumalanga worked in one of the factories in Hammarsdale in the 1980s. But from 1990 onwards, an estimated 60 factories closed down, and 3 500 Mpumalanga residents lost their jobs. The majority of those who worked in the factories remain unemployed today. Potential new entrants to the labour market join their ranks every day.
There are many similarities between Enhlalakahle and Mpumalanga. They share a similar economic and political history. Both townships were part of the KwaZulu homeland. Both were established as labour reserves for clothing, textile and footwear industrial nodes. Both places have been hit hard by mass unemployment, poverty and HIV and AIDS. They both experienced political violence in the late 1980s and early 1990s. The shared legacies of apartheid, political violence and post-apartheid policies have played an important role in shaping the livelihood options that are available to the two communities.
There are also significant differences between the two communities. The size, physical location and political dynamics of the two townships are different. Enhlalakahle is adjacent to the medium-sized town of Greytown, while the nearest sizeable town to Mpumalanga is the city of Durban, almost 30 kilometres away. Agriculture also provides more employment possibilities for the people of Enhlalakahle than it does for those of Mpumalanga. Enhlalakahle is one of only three ANC-led wards among eleven IFP-led wards in the IFP-led municipality of Umvoti, a state of affairs which produces a considerable degree of tension between ward councillors and the local ANC branch. Mpumalanga, on the other hand, has ANC-led wards within the ANC-led municipality of eThekwini.
The economic and political trends that have emerged in the two townships will be described in the chapters that follow. Despite difficulties, the two communities are showing embryonic signs of political recovery. While the years between 1984 and 1994 can be described as a ‘decade of devastation’, ‘proliferation of war zones’, and ‘the destruction of pillars of the home’ (Sitas, 2002), the post-1994 period represents the beginning of an era of reconstruction and rebuilding in the two communities. However, while there is hope politically, the economic situation is bleak. Both townships have been sinking deeper into poverty and unemployment despite a plethora of ‘progressive’ post-apartheid economic policies, and poverty alleviation programmes. The evidence presented in this book, and confirmed by similar studies, suggests that South Africa’s post-apartheid economic, social and labour policies have failed to adequately address the systemic causes of poverty, inequality and unemployment, even though they represent a radical departure from the racially-based economic policies which preceded them. The macro-economic policies that have aimed at accelerating the integration of the South Africa’s economy into the global economy have had the adverse effect of deepening inequalities within the country itself (Webster and Bezuidenhout 2000; Ruiters, 2002; Webster and von Holdt, 2005; Seekings and Nattrass, 2005).
Conducting research in Mpumalanga and Enhlalakahle
The emphasis of this book is on the stories of poor households rather than on analysis. Drawing on over 100 intensive interviews, the book provides a space in which multiple voices can speak. Chapters one, two and three provide a vivid picture of individual lives, perceptions and experiences. Chapter four uses evidence from the interviews in order to present a deeper analysis of the socio-economic context of contemporary South Africa. It also demonstrates the limitations of the sustainable livelihoods approach and specifically the ‘asset vulnerability framework’, which is widely accepted in the development literature (Rakodi, 1992, 1999; Beall, 2002; Francis, 2000; De Haan and Zoomers, 2005). Contrary to the conventional view, the study shows that households are not just places of consumption and leisure, but significant places of production and services, a significance that has increased considerably with the growing levels of unemployment, poverty and HIV and AIDS infections in the country. The book concludes with a summary of its key findings, arguing that poor households have become crucial but fragile sites of survival.
The book is based on in-depth qualitative research that was conducted in the two communities of Enhlalakahle and Mpumalanga at selected intervals in 2002, 2003 and 2004. Face-to-face interviews formed the core of the research. Pilot interviews, complemented by two focus group interviews, were undertaken initially in each of the research sites. Sixteen households in Enhlalakahle and thirteen households in Mpumalanga were selected for further study on the basis of these initial interviews. Forty-four semi-structured interviews were then conducted. This was followed by 29 oral history interviews with older women from the 29 selected households in the two townships. These were augmented by oral history interviews and fifteen key informant interviews with members of local community organisations, local leaders and government officials.
Participant observation was the key to collecting data. It was often the only research technique that was appropriate given the sensitivity of the household issues I was exploring and the need to gain people’s trust. Many of the respondents were reluctant to accept the use of a tape recorder. As a result, the household visits generally involved conversation and observation rather than note-taking and recording. In addition, all the household informants were guaranteed anonymity. Accordingly, pseudonyms have been used for them as well as for those informants from the community who requested not to be named. Community members are usually referred to in terms of their affiliation and/or role in the community.
Archival or documentary research and analysis also formed an important component of the research methodology. I examined the correspondence and documentation that related to land ownership, the renting and buying of houses, and work permits in the two areas. Newspaper articles from the archives of the Greytown Gazette and the Natal Witness proved especially valuable in enabling me to understand the broader political economy of the two townships. They were also important sources of information about the political violence and resistance that took place in the townships in the 1980s and 90s and the economic development of the two areas.
I also kept a fieldwork journal in which I recorded direct observations during household visits, interviews and informal discussions with community members. The journal also included my personal research process notes. These notes served three main purposes: they offered a way to evaluate direct observation or make inferences when I went over the notes later, they served as a source of data, and they helped me to relieve the stress that accompanied the process of participant observation. I also included analytical notes in the journal, which I used to record methodological issues and to plan the way forward.
The dynamics of the research process were often complicated. Entering people’s personal spaces and asking difficult questions such as, ‘How do you survive?’ required great sensitivity. I was overwhelmed by the way that people opened their homes to a stranger. Most of the people interviewed had experienced suffering and hardship. I found that they were willing to share their problems and frustrations with me. They also tested me by asking me questions. The interviews were conducted in the informants’ home language, isiZulu. This facilitated communication although differences in accent sometimes had to be overcome. We often laughed about how Setswana speakers (my home language), and particularly those from Gauteng (my home province), speak isiZulu differently to isiZulu speakers from KwaZulu-Natal.
Although the dichotomy between researcher and informant, between ‘us’ and ‘them’, often fell away during my interactions with the respondents, my level of education, my perceived class status and the fact that I came from Gauteng (a big urban and metropolitan area) were