Dispensing with Marriage: Marital and Partnership Trends in Rural KwaZulu-Natal, South Africa 2000-2006

Article excerpt

Abstract

This paper describes marriage and partnership patterns and trends in rural KwaZulu-Natal, South Africa from 2000-2006. The study is based on longitudinal, population-based data collected by the Africa Centre demographic surveillance system. We consider whether the high rates of non-marriage among Africans in South Africa reported in the 1980s were reversed following the political transformation underway by the 1990s. Our findings show that marriage has continued to decline with a small increase in cohabitation among unmarried couples, particularly in more urbanised areas. Comparing surveillance and census data, we highlight problems with the use of the 'living together' marital status category in a highly mobile population.

1. Introduction

By the 1980s, when retrospective analyses of African censuses and fertility surveys showed that changes in nuptiality were occurring in many sub-Saharan countries, marriage patterns in South Africa were already exceptional (Harwood-Lejeune 2000; Lesthaeghe & Jolly 1995; Van de Walle 1993). The mean age of marriage for men (28.0 years) was higher than all other regions in Africa, and that of women (23.2 years) was one of the highest (Locoh 1988). In seeking to document and explain the early and advanced decline in African marriage in South Africa, authors have paid particular attention to the profound and lasting influences of apartheid policies. In the 1990s, it was unclear if, and how, the anticipated increase in political, social, and economic opportunities for Africans resulting from political change would affect family life. Taken in their entirety, contemporary tribal, religious, and legislative structure and processes are favourable towards marriage and seek to promote it as the preferred family institution. The action by the country's first post-apartheid government to formulate a new marriage act that sought to recognise and legitimise the plural religious and ethnic marriage traditions in the country exemplifies strong social norms about the positive value of marriage. Would political change lead to a reversal in the movement away from universal marriage? However, at the same time as political transformation, South Africa started to experience a rapid and severe HIV epidemic further complicating predictions of future trends in marriage. Any impact of HIV and AIDS on marriage, re-marriage, and widowhood would occur in tandem with other social and economic changes affecting people's decisions about family formation (Heuveline 2004).

For demographers seeking to document trends in marriage and partnering in South Africa, there are surprisingly few sources of data. In a country where marriage is far from universal, the lack of discrimination between marital states and cohabitation arrangements in the most recent South African censuses limits the interest of this data (Budlender, Chobokoane, and Simelane 2005; Ziehl 1999). Furthermore, relating marriage and partnership trends to the complex social and residential arrangements in which many people live in South Africa, needs detailed, population-based data. In this paper, we use demographic surveillance system data on a population of approximately 86,000 people in rural KwaZulu-Natal between 2000-2006. The Africa Centre Demographic Information System (ACDIS) was established with the purpose of collecting empirical data about socio-demographic change and has been well documented (Hosegood, Benzler, and Solarsh 2005; Hosegood and Timæ us 2005; Tanser et al. 2007). Data on marriage, non-marital partnerships and patterns of cohabitation are collected routinely in ACDIS.

This paper describes contemporary patterns and trends in marriage and partnering in rural KwaZulu-Natal. The descriptive findings are a starting point for further research on this subject. They fill a gap in the South Africa marriage and household literature, and provide a platform for exploring causative factors. The paper is structured as follows. …