Spousal Interdependence, Female Power, and Divorce: A Cross-Cultural Examination

Abstract

A recurrent speculation in the social science literature is that divorce rates climb when women gain more independence in marriage or more equality with men. This idea is present in many accounts of the secular rise in divorce accompanying industrialization (Goode, 1965). These accounts point to a decline in patriarchal authority and an increase in women's employment in urban, industrial society as important factors. A similar assumption is lodged in treatments of lower-class black family life and kinship relations (Stack, 1974), where men's unemployment and women's independence are said to make relationships between the sexes more fragile. Armed with quantitative data, researchers are debating on the applicability of this general speculation to America today. Earlier research Ross and Sawhill), 1975; Hannon et al., 1977) reported that women's income or employment increases chances of subsequent divorce. Much of the more recent research on women's employment and divorce has attempted to untangle the differing effects of women's employment and income on divorce (Huber and Spitze, 1980; D'Amico, 1983; Philliber and Hiller, 1983; Booth et al., 1984; Greenstein, 1990; Cain and Wisoker, 1990a; Hannon and Tuma, 1990; 1990b; Starkey, 1991 ). All studies find some connection of women's working outside of the home to divorce, either as a zero-order coefficient or as a multivariate coefficient.This paper follows up on previous cross-cultural research testing this speculation by attempting to separate the effects of power from the effects of functional independence. Rather than dealing with the American scene, we attempt to answer the question of whether this speculative account of the link of divorce to women's equality and independence is broadly generalizable in a cross-cultural perspective.PREVIOUS RESEARCHCross-Cultural Research. Cross-cultural researchers have shown more interest in the process of marrying than in the process of divorcing. The considerable literature on this topic examines such aspects as marriage transactions (Murdock, 1949; Schlegel and Eloul 1988), marriage arrangement and marriage motives (Rosenblatt and Cozby, 1972; Coppinger and Rosenblatt, 1968), and cousin marriage restrictions (Levi-Strauss, 1969; Homans and Schneider, 1955). While divorce may not be as universally claimed as marriage, it is deserving of more study. Previous studies may be categorized in three kinds of investigations: (1) qualitative comparisons, (2) quantitative studies of normative aspects of divorce, and (3) quantitative studies of divorce frequency.A flurry of qualitative anthropological essays on divorce followed Gluckman's (1950) case study on the fragility of marriage in matrilineal societies compared to patrilineal groups (Leach, 1957; Mitchell, 1961). The subsequent literature raises questions on variation in divorce within both patrilineal and matrilineal groups, and on the exact mechanism that might make a difference in divorce rates between matrilineal and patrilineal societies. This interest in descent as a determinant of divorce rates has not been sustained within the cross-cultural research community, perhaps because studies using larger samples of societies do not bear out most of the ideas based on small scale comparisons. In a later comment, Gluckman (1971) backs away from his earlier view, saying that he had tried to make an excessively forceful argument in order to stimulate anthropological research on divorce frequency.Several quantitative cross-cultural studies examine normative aspects of divorce rather than its frequency. Minturn et al. (1969) find that ease of divorce for men is correlated with matrilocal residence. Frayser (1985) adds that ease of divorce varies by world region. It is less accessible to women in the Circum-Mediterranean area, and men in Africa have more obstacles than elsewhere. Betzig (1989) looks at the expressed reasons and motives for divorce in non-industrial societies. …

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 17 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.