Journal Issue: Children and Divorce Volume 4 Number 1 Spring/Summer 1994
Historical Changes in Divorce and Remarriage
Until the latter part of the nineteenth century, divorce was largely proscribed by law and shunned in practice much as still happens today in many nations including some European countries such as Italy and Ireland.6 Most marital disruptions occurred not as a result of divorce but from desertion or informal separation. Because population surveys were not available prior to the middle part of the twentieth century, it is difficult to know how often de facto divorce took place in the United States. But, it seems likely that all but a small minority of marriages survived until the death of one or another partner, an event that typically occurred much earlier than it does today.7 Some have argued that the rise of divorce was partly prompted by increasing survival rates, which placed a greater strain on the ability of couples to manage marital stress or maintain marital contentment.8 However, there is no firm evidence to support this conjecture.
Divorce rates in the United States began to rise shortly after the Civil War and continued on a steady upward course for more than a century. Over this time rates have fluctuated, often falling in poor economic times and generally surging after major wars. But these short-term variations have been far less consequential to the long-term pattern of constant growth.9 Nearly two decades ago, Preston and McDonald calculated the likelihood of divorce for each marriage cohort beginning in 1867 and continuing until the mid-1960s.10 Their results showed a continuous trend of dissolution among successive marriage cohorts. Roughly 5% of marriages ended in divorce just after the Civil War compared with an estimated 36% in 1964. Thus, the pattern of prevalent divorce was firmly in place in this country even before the divorce revolution of the 1960s.
Nonetheless, there was a sharp increase in the incidence of divorce from the mid-1960s to the late 1970s. During a span of a decade and a half, divorce rates for married women more than doubled (from 10.6 per 1,000 in 1965 to 22.8 in 1979), pushing the risk of divorce much higher for all marriage cohorts, especially those who wed after the mid-1960s.11 Some researchers speculated that a majority of all marriages contracted in the 1970s and after would end, especially when both informal separations and formal divorces were counted.12 Other researchers reached more conservative estimates but still projected that more than two in every five marriages would end in divorce when divorce rates reached their peaks in the middle 1970s.9
Divorce rates began to level off in the late 1970s and actually declined by about 10% during the 1980s.13 As mentioned earlier, fluctuations of this sort are common historically and do not necessarily signal a reversal in divorce trends. Nonetheless, most demographers think that divorce is not likely to continue its upward pattern, at least in the near term. There are several demographic explanations for the failure of divorce rates to increase after the 1970s which do not necessarily imply that Americans today are becoming more committed to staying married than they were in the previous two decades.
The huge cohort of baby boomers, reacting to changing economic opportunities, postponed marriage.9,14 A larger proportion opted to obtain more schooling and wait to form a family.15 Marriage age for women rose from just above 20 in the mid-1950s to 24.4 in 1992, an increase of more than four years.16 It has long been known that early marriage and lower education are associated with marital instability.17 Thus, the pattern of delayed marriage might have had a role in curbing the rates of divorce.
Another potent source of marital disruption, associated with early marriage, is premarital pregnancy. Fewer marriages today occur as a result of a premarital pregnancy.18 It also seems plausible that the greater availability of contraception and abortion in the 1970s may have discouraged the formation of early unions, reducing the number of ill-considered marriages, though evidence to support this hypothesis is not available.
Furthermore, the population has been getting older as the baby boomers mature. Older couples in long-standing marriages have a lower propensity to divorce.19 Thus, as the baby boomers reach middle age, a larger proportion of those married have passed through the high-risk years, when their marriages are young and relatively more fragile.
Finally, growing rates of cohabitation before marriage may have brought down the rate of divorce. As more and more couples elect to live together prior to marrying, it seems likely that many unions that would have ended in divorce end before marriage occurs. That is, a growing number of Americans are divorcing without marrying, making the official divorce statistics a less reliable barometer of union stability.20
For all these reasons, it is probable that the modest drop in divorce rates does not indicate a higher propensity toward marital stability. Instead, the composition of those marrying has changed in ways that only make it appear that marriages are becoming more stable.Remarriage
Not so many years ago, it was common for family experts to reassure those who were alarmed at the steady increase in divorce rates by pointing out that divorce typically is not a terminal event but a transition from one marriage to the next. So it was said that couples who separated lost faith in a particular marriage but not in the institution of matrimony.21 In 1975, close to three-fourths of all women in their fifties who had experienced a divorce had remarried. For formerly married men, the occurrence of remarriage was even higher, about four in five eventually remarried, owing to the greater pool of eligible partners. (It is easier for men to attract younger partners than it is for women.) But recently, the rate of remarriage has been declining.22
In part, the trend toward lower remarriage rates may reflect the greater tendency to postpone second unions as both men and women may be more willing and able to live as single persons. But recent evidence from the National Survey of Families and Households (NSFH) suggests the rate of recoupling has not declined notably.23 Many divorced persons have become more cautious about reentering matrimony, preferring instead to cohabit in informal and more fluid unions. This pattern, discussed below, poses particular problems for children who are, to an increasing extent, being raised by quasi-stepparents who are often transitional figures in their households.
The lower rates of remarriage may reflect a growing reluctance to formalize unions after a failed first marriage. Couples who remarry are known to have a higher risk of divorce than couples entering first marriages. And divorces from second marriages occur more quickly than from first unions. Cherlin has shown that the proportion of couples who will marry, divorce, remarry, and redivorce has risen eightfold during the course of this century, climbing from barely 2% of those who were born in the first decade of the twentieth century to 16% of those born after 1970.9
Cherlin described the changing patterns of marriage, divorce, and remarriage for four birth cohorts of women (see Figure 2 in the article by Shiono and Quinn in this journal issue). For all but the most recent cohort, the proportion ever marrying remained relatively stable while the prevalence of divorce, remarriage, and redivorce progressively increased. In the youngest cohort, women born after 1970, Cherlin projects that marriage (and remarriage) will decline significantly and divorce will remain high among women who elect to marry or remarry.Racial/Ethnic Differences in Patterns of Divorce and Remarriage
Rising rates of marital instability have been experienced by all Americans regardless of socioeconomic status, race, religious affiliation, or region of the country. However, the extent of marital instability differs enormously among various social groups. It is beyond the scope of this article to explore in detail the patterns described above for different social classes, religious groups, or regions of the country. It is hard to ignore, however, racial/ethnic differences in patterns of marriage, divorce, and remarriage because the experiences for whites, African Americans, and some Hispanic groups are so very disparate.
African Americans have long exhibited different patterns of family formation.24 As far back as the nineteenth century, blacks were more likely to marry earlier, had a higher incidence of premarital pregnancy and nonmarital childbearing, formed less stable unions, and were less likely to remarry when disruption occurred. Scholars disagree on the origin of these patterns.25 Some believe that they are rooted in different notions of kinship brought to America; others argue that distinctive patterns of family formation emerged in slavery; and still others contend that these family differences did not really take hold until after Emancipation, when black Americans were exposed to economic discrimination and racism. Still others argue that the differences are more recent in origin.26
Whatever the particular origin or combination of origins, there is convincing evidence that African Americans are much less likely to marry, more likely to divorce, and less likely to remarry when divorce occurs.27 More than 90% of whites will marry compared with about 75% of African Americans; of those who do wed, African Americans have a substantially higher risk of divorce.13 Ten years after marriage, 47% of blacks have separated or divorced compared with 28% of non-Hispanic whites. Blacks are also far less likely to remarry after separating. As a result, African Americans spend far less time in marriage than do whites.28
Much less information exists on the marriage patterns of other racial and ethnic groups. Census data on Hispanics suggest that their levels of marriage, divorce, and remarriage fall somewhere between those of whites and those of blacks.13 However, official statistics actually conceal as much as they reveal about the behavior of different Latino groups. There is reason to suspect that as much difference exists between Cubans or Mexican Americans and Puerto Ricans as between whites and blacks in rates of marriage and marital stability.29 Still, such as it is, the evidence on Hispanic subgroups reveals similar trends to those described for blacks and whites in the United States.
In sum, virtually all population subgroups have experienced a postponement of marriage, a steady increase in divorce, and a decrease in remarriage after divorce. Cohabitation as a prelude, aftermath, and perhaps alternative to marriage has become more common. These patterns are more evident among African Americans.Childbearing
The declining institution of marriage has important ramifications for patterns of childbearing. Typically, now, marriage no longer regulates the timing of sex, and to an increasing degree, it no longer regulates the timing of first birth.30 Nonmarital childbearing has become more prominent over the past several decades as rates of marital childbearing have declined and rates of nonmarital childbearing have held steady or increased. In 1960, only 5% of all births occurred to unmarried women; in 1990, this proportion had risen to 28%.31 The increase for whites has been tenfold, from 2% to 20% in this 30-year period.
Figure 1 depicts the remarkable rise in the number of first births among women between the ages of 15 and 34 which have occurred before marriage for whites, blacks, and Hispanics. Among each of the racial/ethnic subgroups, the increase has been remarkable over the past 30 years. For whites this number rose from 8.5% for births occurring in the early 1960s to 21.6% for those that took place in the late 1980s. The rise for blacks was even more spectacular, going from 42.4% in the early 1960s to 70.3% in the late 1980s. The proportion for Hispanics doubled during the same period, going from 19.2% to 37.5%. Clearly, out-of-wedlock childbearing has become a far more important source of single parenthood for all Americans and especially so for African Americans, who now have a sizable majority of first births before marriage.18 (See Figure 1.)International Comparisons
The weakening of marriage as a social institution is not unique to the United States. Most developed countries are witnessing similar demographic trends.32 In some instances, the retreat from marriage is even more pronounced. For example, in Scandinavia cohabitation has become a widely accepted alternative to marriage.33 France and England have higher proportions of out-of-wedlock births than occur in the United States, though a higher proportion of these births occur to parents who are cohabiting than in this country.34 Divorce rates have also risen sharply in a number of European nations, though none equals this country in the prevalence of divorce. Still, about a third of marriages in Northern Europe will end in divorce; in England and Scandinavia, as many as two in five marriages may dissolve.35 Thus, explanations for the de-institutionalization of marriage cannot reside solely in the special features of American culture or society.