Levels of Divorce: The divorce rate is found to be inversely related to duration of marriage. This has also been found elsewhere [McCarthy 1978, Menken et al. 1981, Bhuiya et al. 1999]. The least satisfactory marriages end in divorce first thereby reducing the proportion of the high-risk marriages among the surviving unions. When newly wed partners discover that they are incompatible whatever be the cause, they opt for divorce rather early in the marriage. In rural Bangladesh, marriage makes the bride live in the husband's home and with in-laws. The in-laws and relatives expect her to adjust to her new situation and demand a lot from her. Her failure to adjust to in-laws and to satisfy them can cause marital discord and the result of which is ill-treatment to her by her husband or in-laws leading to divorce.
Polygyny and Divorce: Polygyny is a special feature of Muslim society and is allowed on condition of the husband's equal treatment of all his wives. Infertility, particularly inability to give birth of a male child and chronic illness of the wife are two main reasons which enhance the divorce risk. They can lead either to a marriage of the husband to a second wife (and becoming a polygynous marriage) or to a divorce of the previous wife followed by another marriage. In Teknaf, 17 per cent of the current marriages were polygynous as opposed to five per cent in Matlab. Of the polygynous men, 26 per cent were in their twenties, and 40 per cent in their thirties. The high incidence of polygyny at younger ages of husbands is much more due to widely accepted norms and values held by men in this community than to wives' infertility or chronic illness.
Polygynous marriages were found to be divorce-prone; the odds of divorce were 2.5 times higher for polygynous unions than for monogamous unions other things being equal. A marriage in which the husband already had one or more than one wife faces more complex adjustment problems than a marriage with a man having no other wife. Moreover, polygynous men have less to lose from divorcing one of the wives. They also tend to have a lower commitment to marriage as an institution and can more easily disregard the stigma of divorce and remarriage. High tensions among co-wives and deeply ingrained attitudes of polygynous men towards women and marriage in general may raise the divorce rate. What remains unknown is whether the high risk of divorce of polygynous marriages is caused by complex adjustment between co-wives or by attitudes of polygynous men to women and marriage as institution. Future studies should attempt to deal with this issue.
In Nigeria, it has been found that polygynous unions with two wives were the most stable whereas unions with three or more wives were associated with the highest rate of disruption [Brandon 1991]. In our study population, unions with three or more wives simultaneously exhibited higher divorce rate, but occurred less frequently than unions with two wives.
Spouses' History of Marital Disruption: Spouses who already experienced a divorce in their earlier marriage probably faced substantial adjustment problems with which they were unable to cope. We hypothesize that remarriages exposed them to the same problems and for this reason they remain vulnerable to divorce. While expecting high risks of divorce of remarriages, one must take into account that in rural Bangladesh women are aware of the lower prospects of remarriage for them than for men and this may force women to compromise to keep the marriage intact. Reconciliation is more likely for wives than husbands since the cost of divorce is higher for women and alternatives are hardly available (women are economically and socially dependent). One of the underlying causes for this is undoubtedly the unequal position of women compared to men in Bangladesh society. Our data exhibit a gender difference in the risk of divorce of remarriages. While grooms who remarried after a divorce or after becoming a widower did not have a higher risk of divorce than grooms who married for the first time, brides who remarried (after divorce or becoming a widow) had a higher risk of divorce than brides who married for the first time.
The difference in divorce propensity between previously divorced brides and grooms may have several explanations. An important one is related to the weaker position of the wife in the husband's family. She is blamed for every trivial fault; for such faults her husband is less likely to be blamed. Too much blame expressed too frequently puts her marriage in discord leading towards divorce. Of the brides who were divorced before, 42 per cent were still in their teens. Their inability to live up to the expectations of their husbands and in-laws may have brought marital discord leading eventually to divorce as a solution to marital unhappiness. Another possibility is that the presence of children from a previous marriage may make her stay more problematic than that of her peers who had married for the first time. Higher instability of remarriages of women than men may be explained by men's abuse of the right to divorce and women's powerlessness.
Age at Marriage: As found in other studies (Shaikh 1995, Bhuiya et al. 1999, White 1990), the groom's age at marriage was inversely related to the risk of divorce. Young grooms tended to marry young brides. Many grooms were perhaps too young to be able to carry the heavy load of responsibilities involved in marital life. Divorce is likely to be more common among couples who are poorly prepared to undertake such responsibilities. This finding is consistent with the common explanation that young age at marriage may be associated with poor role performance as a spouse.
Groom's Education and Household Socioeconomic Status: Higher levels of education of grooms were associated with lower odds of divorce in the first year of marital life. Literacy in Teknaf was very low (14 per cent of grooms had one to two years of schooling) and nearly all brides were illiterate. The odds of divorce were higher if grooms came from households with fewer assets. Shortage of economic resources which are needed for a successful beginning of marital life, may have raised tensions and caused unhappiness. Moreover, it could be that poor grooms were less tolerant and respectful to their wives (and vice versa) than rich grooms. Inverse relationships of education and socioeconomic status with divorce have also been reported in other studies conducted in Bangladesh, the USA and elsewhere [see e.g., White 1990, Bhuiya et al. 1999].
Infertility and Divorce: In rural Bangladesh, child bearing soon after marriage is desired. Contraceptive use before first birth is in general very low and was in the study population even lower than the national average derived from the Demographic and Health survey conducted in 1993/1994 [Mitra et al. 1994]. The birth of a child after marriage may signify a degree of spousal satisfaction that is conducive to marital stability. Our analysis reveals that the odds of divorce were much lower if women had a live birth in the proceeding six-months than if women had not had a birth. The birth of a child helps to keep couples together at least until they are older. This result is in accordance with findings in the USA and Sweden [White 1990, Waite and Lillard 1991, Andersson 1995].
Limitations of this Study: An anthropological study carried out in rural Bangladesh reported that the most desirable traits in a bride are to be young and to have physical beauty and lineage. For the groom wealth and earning capacity are desired most often [Aziz and Maloney 1985]. However, data especially on these physical, personal and family attributes of grooms and brides were not available in this study. These attributes were prime considerations when parents and relatives of the partners arranged the marriage for them. This point was also made by White  in her review of research on divorce in the 1980s although she focused more on the need to do research on processes of marriage formation and dissolution than on desirable attributes. She also stressed to conduct more studies on the impact of the processes of family formation on the processes which lead to divorce. One study in Maltab revealed that the process of marriage formation and spouses' behaviour after marriage influenced the divorce risk [Bhuiya et al. 1999]. This finding is in keeping with our result that kinship between partners, which facilitate seeing and knowing each other, was associated with a lower risk of divorce. While the process of marriage formation is important, groom's polygamy and bride's divorce in a previous marriage appeared to be the most significant antecedents of divorce in the study population.
Another limitation of our study is that since the follow-up period was five years, the long-term effects of various determinants on risk of divorce could not be examined. One can expect that reasons for divorce change with duration of marriage. Nevertheless, high rates of divorce of polygynous marriages and of brides' remarriages were evident and may be explained by the unequal position of women in the family and society.
Determinants of Divorce in a Traditional Muslim Community in Bangladesh
Nurul Alam, Sajal K. Saha, Jeroen K. van Ginneken
© 2000 Max-Planck-Gesellschaft ISSN 1435-9871