Abstract

This article interrogates the extent to which the population censuses conducted in South Africa between 1921 and 2001 provide evidence of a decline in marriage rates among African males and females. It thus differs from other articles in this collection which discusses findings from household sample surveys. In addition to presenting the trends, the article summarises the different arguments offered in the literature relating to this period for the decline in rates of marriage. The article suggests that most of the arguments have merit, but the strength of the different factors would have differed over time and among different groups of women and men. Further, the article offers a possible methodological and linguistic reason for the perhaps ‘incorrect’ finding that the decline in marriage rates has stalled since 1995, but also argues that those who claim that it is post-apartheid factors that are driving the decline in marriage need to confront the fact that this is not a new trend.