Polygyny, Gender Relations, and Reproduction in Ghana

Article excerpt




Most demographic research on the institution of polygyny in sub-Saharan Africa has focused on reasons for the continued persistence or for the transformation of this institution and on individual-level effects of polygyny on fertility desires and behavior. Thus several reasons have been advanced to explain the existence, prevalence, and persistence of polygyny: some studies have focused on religion as an explanation (Kelly 1992; Makoteku and OchollaAyayo 1988), whereas others argue that polygyny is determined by women's role in agricultural production (Cronk, 1991; Jacoby, 1995; Rubin, 1990). Approaching the institution of polygyny from the perspective of the evolutionary behavioral theory some researchers have attempted to apply the "polygyny threshold" model postulating that polygynous marriage is a marital strategy to maximize reproductive success and that it results both from men's differential capacity to support their offspring and from women's (and their families') ability to freely choose their futur e marital partners (Borgerhoff Mulder, 1988; Josephson, 1993). This approach, however, has been challenged by what appears to be a very complex and multidimensional relationship between polygyny and fertility outcomes. Indeed individual-level analyses of the effects of polygyny on fertility have not produced consistent results (e.g., Adewuyi, 1988; Borgerhoff Mulder, 1989; Pebley and Mbugua, 1989; Pison, 1986; Sichona, 1993; Timaeus and Reynar, 1998). Attempts by researchers to examine the aggregate effects of polygyny on reproductive processes are often frustrated by lack of appropriate data and by the endogeneity of polygyny in relation to aggregate measures of socioeconomic development. However, in a recent study Ezeh (1997) did find that areas 'where polygyny levels are high have unique reproductive patterns consistent with the attainment of high fertility.

Generally missing in these studies is an examination of the link between polygyny levels and the pattern of gender relations observed in an area. Understanding the effect of aggregate polygyny levels on the nature and pattern of gender relations may be central to our understanding of the effects of polygyny on reproductive processes. In this study we attempt to examine these effects by using primarily qualitative data from Ghana. We argue that the high prevalence of polygyny in an area, being itself a product of social inequality between sexes; acts to reinforce and perpetuate this inequality through cultural norms and prescriptions that govern gender relations, and these norms and prescriptions apply equally to both monogamous and polygynous couples. We propose that in areas with higher prevalence of polygyny, married women-both in monogamous and polygynous unions--are more marginalized in social, economic, and reproductive decision-making, compared to areas with lower polygyny levels. In a higher-polygyny environment women's position and roles in the household are seen as easily replaceable, which undermines their relative bargaining position vis-a-vis their husbands in major household decisions. Thus marginalized, such women lack the ability to question or challenge the dominant gender hierarchy and the institution of polygyny, are more likely to accept the husband's lordship and authority with resignation and acquiescence, and therefore are less able to articulate, express, and fulfill individual desires even in very private matters such as childbearing.


We investigate these relationships using quantitative data from two Demographic and Health Surveys (GDHS) conducted in Ghana in 1988 and 1993 and the qualitative data collected through focus groups discussions by one of the authors in Ghana in 1991. The GDHS were part of an ongoing large survey series conducted in a number of less developed countries. The GDHS used locally adjusted standard DHS questionnaires that included sections on women's and men's general sociodemographic characteristics, nuptiality, fertility and family planning, child mortality and health, and other related matters. …


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.