At the Council on Contemporary Families, we are fond of saying that the right research question in today's world is not “What kind of family do we wish people were living in?” but “What do we know about how to help every family identify and build upon its potential strengths and minimize its characteristic vulnerabilities?” In my 75th anniversary address to the National Council on Family Relations (NCFR) (Coontz, 2013), I celebrated the commitment to scientific inquiry and debate that has helped the NCFR's members correct many prejudices and misconceptions about family diversity over the decades. But it also seemed a good time to identify some of the new challenges NCFR researchers and practitioners face on the road to the organization's 100th anniversary.
One of the markers of successful aging—of organizations as well as individuals—is a willingness to acknowledge the mistakes, misconceptions, and errors of youth without feeling humiliated by them. That ability stems in part from understanding that the errors were seldom ours alone but were shaped by the culture in which we were raised or immersed. And it also stems from having had to rethink and expand one's worldview in light of new information and experiences. It follows that one of the markers of successful future aging is a recognition that this process never ends—that lifelong learning, even after 75 years, requires lifelong unlearning and rethinking.
I am not one to underplay the historical weaknesses of the family research and marriage counseling movement—its association with the eugenics movement; its uncritical embrace of functionalism; its blind spots on class; and its perpetuation of racial, ethnic, gender, and sexual stereotypes. But instead of being embarrassed by these errors, NCFR should be energized by the debates and self-criticism that allowed so many new outlooks, values, and research methods to take root. For one thing, NCFR members were willing, right from the beginning, to couple the phrase “marriage and family” with “social sciences.” When I studied history in college—25 years after NCFR was founded—private life and interpersonal relationships were generally dismissed as distractions from the “serious” history of diplomacy, war, politics, and economic structures. It was a revelation to me, early in the 1960s, when I discovered Bertolt Brecht's (1935/1976) poem “A Worker Reads History”:
- Who built the seven gates of Thebes?
- The books are filled with names of kings.
- Was it kings who hauled the heavy blocks of Stone?
- And Babylon, so many times demolished.
- Who raised it up so many times?
- In the evening when the Great Wall of China was finished
- Where did the masons go?
As the daughter of a onetime labor organizer, I loved that poem. Even so, until the 1970s it didn't occur to me to follow those workers home to their families in my own studies. Yet more than 30 years earlier, NCFR was already following those workers home, literally. At NCFR's very first conference, family scholars analyzed the houses workers lived in and how that affected their lives.
Speakers at NCFR's first conference included Paul Popenoe and Ellsworth Huntington (NCFR, 1938), whose eugenicist views on race and gender should have been an embarrassment even then. But just a year later, in the first volume of NCFR's journal Living, Frederick Osborn (1939) argued that anyone truly interested in improving the human race, as was the stated purpose of eugenics, should reject the negative goal of determining who was unfit to be a parent in favor of the positive goal of ensuring that every child was wanted. The best way to ensure the “fitness” of children, said Osborn, was to make contraception available to all couples, whatever their income level, and to reduce the costs of raising children. He advocated government subsidies for family housing, day nurseries, free meals in the public schools, and public recreation programs for children. Compare those recommendations to the views of so many of today's politicians, and I think the NCFR of 1939 comes off pretty well.
In retrospect it is certainly disturbing to realize how many prominent NCFR members subscribed to the functionalist version of “normal” family gender roles, labeling any departure from those roles a threat to marriage and to the proper development of children. Yet in the early 1950s, the NCFR journal also published articles discussing “favoritism” toward men in family systems and noting women's growing desire for both marriage and employment as a trend to be analyzed rather than a threat to be decried. Some NCFR leaders even questioned whether homosexuality was really a disease, as it was then labeled by most professional associations.
The NCFR did not begin to seriously study domestic violence until the 1970s, but as a historian I would take silence any day over the discussion I found in a 1964 article of Archives of General Psychiatry, published by the American Medical Association (Snell, Rosenwald, & Robey, 1964). Analyzing the cases of 37 women married to men who had physically abused them, the authors reported that the wives typically did not call the police until more than a decade after the abuse began, usually following an incident in which a teenage child intervened to stop the violence. However, instead of proposing ways to bring such abuse to light earlier, the psychiatrists concluded that the child's intervention disrupted a family process that was “working more or less satisfactorily” to restore proper gender roles (p. 109). In their view, the real problem was that the wife was “aggressive, efficient, masculine, and sexually frigid.” Domestic violence restored the “marital equilibrium,” allowing the wife “to be punished for her castrating activity” and the husband “to re-establish his masculine identity” (p. 111). By the time this article came out, NCFR founding member Emily Mudd was already pushing marriage counselors to abandon their insistence on rigid gender roles and to encourage the “flexible interdependence” necessary for a healthy relationship (Mudd & Fowler, 1969).
It's easy to find class and racial stereotypes in the writings of researchers in that era, yet as early as 1963 NCFR published an article with a commentary by Lee Rainwater arguing that researchers should stop seeing the problems in low-income families as the result of maladjusted individuals or a dysfunctional lower class culture. Instead, he urged, researchers should view the patterns they see in low-income communities “as an imperfect adaption to a particular structural position in society which makes certain kinds of behavior”—for example, sustaining a stable family life—“very difficult, and other behaviors easier” (n.p.). Rainwater even denied that marriage was a panacea for the problems in such communities, suggesting that “cultures [can] provide other supports for personal stability beside the marriage tie” (Willie, Weinandy, & Rainwater, 1963, p. 447)—another point that has yet to register with many contemporary policy makers.
I don't want to exaggerate the extent to which NCFR rose above the intellectual and political climate of its time. But NCFR's early emphasis on scientific principles and free debate paved the way for researchers to analyze rather than condemn the rapid changes in family life that upended their expectations and predictions. And that's impressive, considering how many of those expectations and predictions have been overturned in the past 50 years.
In 1963, less than 20% of American children younger than age 6 years had mothers who were in the labor force. Today, 65% of children younger than age 6 have a mother who is working or seeking work. In 1960, only one third of women were still employed 5 years after graduating from school. Today women are increasingly permanent, lifelong members of the labor force. In fact, women are the sole or primary earner in 4 out of 10 American families.
Fifty-three years ago, the median age of marriage for women was just over 20 years, 6 years younger than today. Of adults age 25–29, 70% were married (and 90% of women in that age group), compared to only 25% today. Only 13% of households contained just one person, most of them elders who had lost a spouse. In 1960 people simply didn't live alone for any significant amount of time unless they were widows or widowers. Nor did they enter into long-term commitments, personal or financial, outside of marriage.
Unmarried couples accounted for less than 1% of all households in 1960. Just one child in 20 was born to an unmarried woman, and there were only 35 divorced people for every 1,000 married people. Today 40% of children are born out of wedlock, and there are approximately 175 divorced individuals per 1,000 married ones. Divorce rates have fallen since their peak in the period 1979–1981, but this does not apply to elders. In 1980, 4% of individuals age 65 and older were divorced. By 2008 that had almost tripled, transforming the role of marriage in old age as well as in young adulthood.
Fifty years ago, 85% of the US population was White and 11% Black. Less than 4% was Latino and less than 6% foreign born. Today there are as many foreign-born Americans as Black Americans, with each category representing 13% of the population. Latinos now account for 17% of the population and are projected to account for 70% of the growth in the labor force over the coming 3 decades.
In the early 1960s, children younger than 18 represented 36% of Americans. Today they represent less than 25%, and their racial-ethnic makeup is changing rapidly. In 14 states the majority of toddlers are now “minorities,” and in the past 12 years, the number of mixed-race children has grown by 50%. We are seeing a growing racial divide between young and old that has been accompanied by a reversal in socioeconomic status: Unlike the 1950s and 1960s, children, not elders, are now the poorest age group in America.
It's hard to tell the extent of changes in the proportion of gay and lesbian families because so many felt compelled to remain in the closet for so long. But today there is not a county in the United States that does not have same-sex couples who are out of the closet. And more same-sex couples are openly raising children, even in traditionally conservative areas of the country. For example, 26% of same-sex couples in the South have children in the home, 24% in New England, and 21% in the Pacific states. We have seen a stunningly rapid increase in support for same-sex marriage and family rights in the past decade.
All these changes are transforming marriage and family life, here and across the globe, forcing family researchers to play catch-up. Everywhere, despite periodic retrenchments or detours, we are seeing a dramatic reduction in the incentive or ability of parents and society to organize people's lives through mandatory marriage and in the ability of men to exclude women from economic and political citizenship. We are also seeing an eclipse of the male-breadwinner households that researchers 50 years ago expected to be the wave of the future. Everywhere, too, the legal monopoly exerted by one socially sanctioned type of family over the organization of people's lives, the initiation of young people into sex and adulthood, and the distribution of economic and political rights has been eroded—in some cases shattered—for better and for worse.
One way to understand the changes in family life over the 75 years that the NCFR has existed is to see them as similar to what happened following the disestablishment of religion, to borrow and adapt an analogy by Harvard historian Nancy Cott (2000). When the state stopped supporting one official church, and a person no longer had to be a member of that church to access economic benefits, political power, and social respectability, religion did not disappear. But new churches and sects proliferated. And even when people stayed in the old church, or joined it for the first time, they did so for different reasons than when it was the only route to economic and political respectability. As a result, even the traditional established church had to change the way it recruited members and held their loyalty.
.The same is true for marriage and gender relations. Many alternatives to marriage have emerged. Even when people marry, they often do so for different reasons from the ones of the past and organize their marriages in different ways. As a result, as Journal of Marriage and Family so tellingly reveals each month, many of the old “rules” about who marries, who doesn't, what makes for a satisfactory marriage, why people divorce, when and why people do or do not have children, and how those children transition to adulthood are changing, often in contradictory ways for different sections of the population.
Consider just a few of those changes, culled from articles in Journal of Marriage and Family (see Coontz, 2000, 2004) and briefing papers of the Council on Contemporary Families (2014):
- In the 1950s, people who married later than average had a greater chance of divorce. Today, every year a woman delays marriage, up to about age 35, and decreases her chance of divorce. After age 35 the risk doesn't continue to go down, but it doesn't go up again either, even though later marriages are likely to involve unconventional matches.
- Until the 1950s, highly educated women were less likely to marry than less educated women. But for women born since 1960, college graduates and women with higher earnings are now more likely to marry—and much less likely to divorce—than women with less education and lower earnings power.
- A survey of divorcing couples in Britain during the postwar period found that violence, adultery, and “unpleasant personal habits” (e.g., nose picking) were the main reasons women asked for divorce. Today, women are more likely to complain of men's excessive work hours, lack of communication, and failure to participate in household chores. In fact, a recent UK study found that a man's failure to do housework, whether or not his wife was employed, was a strong predictor of divorce.
- For most of the 20th century, couples who lived together before marriage had a greater chance of divorce than did those who entered directly into marriage. But for couples married since the mid-1990s, cohabitation before marriage is not associated with an elevated risk of marital dissolution. Indeed, for some subgroups of the population, cohabitation with the intent to marry is associated with a lower risk of subsequent divorce than is direct entry into marriage.
These sorts of changes are typical in times of major social transformation. Correlations and cause-and-effect relationships that prevailed in earlier eras are often turned on their head. Family arrangements that superficially seem similar to the past turn out to have very different causes and consequences. In both colonial America and 19th century America, for example, many elders lived with their adult children and grandchildren. But that involved one set of family dynamics when the parents owned the farm or had signed it over with specific bequests about what they were owed for the duration of their lives, and quite another when parents moved into an urban household with their children after losing their farm. And both these forms of multigenerational households had quite different causes and consequences from the current increase in the number of young adults moving back into their parents' homes.
In light of these dramatic changes, I'd like to suggest a few challenges facing researchers as we contemplate the future. First is the need to follow the moving target. We should pay special attention to ongoing changes in “the rules of engagement” in family life and gender relationships, making sure we don't assume that the results of a data set from the 1980s or early 1990s still apply.
I also think we need to be much more conscious of the different dynamics and internal variations that exist underneath even seeming continuities in this period. Over the past 40 years, we have developed new analytical categories and theoretical tools for understanding racial, sexual, and gender hierarchies. But we need to reexamine these as we follow the changing nature of those hierarchies. One big challenge is how to understand the continuing discrimination and prejudices against women in the context of the repeal of patriarchal laws and the ongoing revolution in gender relationships and values.
I do not subscribe to the idea that we are seeing “the end of men” or the emergence of women as “the richer sex,” as two recent book titles claim (see Mundy, 2012; Rosin, 2012). But there have been qualitative changes in the relations between men and women that may not be adequately addressed by the tools and vocabulary we developed in the 1960s and 1970s. Certainly, prejudices against women and gendered double standards still abound. Implicit bias tests reveal a stubborn undercoat of old prejudices even in people who genuinely support racial and gender equality. And some men (and women) cling to explicitly patriarchal values. But over the past 30 years, most modern states have abandoned their traditional endorsement and enforcement of male dominance.
No society has completely dismantled the historic residue of disadvantage that women continue to carry, and few have even tried. Yet the mechanisms through which gender inequality is reproduced are qualitatively different than in the colonial era, when women worked alongside men doing equal work but had no control over the property they helped accumulate. The dynamics of modern inequality are also very different from 50 years ago, when women were hired for segregated jobs with accepted pay differentials or could be fired once they married or had kids.
We need to stay aware of important changes and variations in the form, content, and consequences of male dominance, as well as in the ways women accommodate to or resist domination. Sociologist R. W. Connell (2005) argued that men receive a “patriarchal dividend” that ensures greater income, authority, respect, safety, and control over their own lives than women. As a historian, I prefer to reserve the term patriarchy for family systems in which fathers' control over property intersects with their authority over the labor of their wives and children. But a male bonus persists even in societies that have overturned traditional patriarchy. The size of that male bonus has differed by race and class, but until recently men of every socioeconomic and ethnic group could expect to be given preference over women at work and deference from women at home. Men were not only guaranteed access to the best jobs; they were legally entitled to control many aspects of family life. They faced few social penalties even for “bad behavior” that was not officially condoned, whether that be sexual harassment, domestic violence, or infidelity.
Today, by contrast, a man who uses physical force or sexual aggression against a spouse or coworker no longer gets an automatic pass. He may find himself jailed for behaviors that used to be male prerogatives. And educational advantage now trumps gender privilege in job opportunities and pay rates. As middle-skill jobs have disappeared, many men who in the past could climb the occupational and income ladder without acquiring higher education have lost this traditional male advantage, and growing numbers of women have been able to escape their traditional assignment to low-skill work.
For men in general and working-class men in particular, the patriarchal dividend arrives much more sporadically than in the past, and it's getting harder to find places to redeem it. Women are still more likely to be poor than men, and they constitute the majority of the poorest of the poor—people whose income falls below half of the official poverty level. But in recent decades we have seen a de-feminization of poverty because the poverty rate of working-age men has risen faster than the poverty rate of working-age women. Between 1969 and 1994, real wages of female high school dropouts fell by 2.2% and by 8% for the youngest such women. But the real wages of male high school dropouts declined by more than 18% and by a stunning 27% for the youngest group of such men.
The percentage of female workers in middle-skill occupations fell by 16 points between 1979 and 2007, but the vast majority of this shift was accounted for by women moving into higher skill jobs. Women's employment in low-skill occupations rose by just 1%. During the same period, by contrast, the percentage of men employed in middle-skill occupations fell by only 7%, but half of that was accounted for by a shift of males from middle- into low-skill occupations.
Many years ago sociologist Christine Williams (1992) coined “the glass escalator” to describe the way that men in traditionally female fields such as nursing, teaching, librarianship, and social work were disproportionately elevated to leadership positions. Recently in Gender and Society, Williams (2013) revisited the concept, noting that it was connected to a job ladder that has largely disappeared in recent years, especially in low-skill occupations. Today, in low-wage retail work, she noted, “there are no glass ceilings [for women] or glass escalators [for men], only revolving doors” (p. 622). And of course many men, especially men of color, are stuck in the boiler room, doing the underpaid jobs that are among the most dangerous in the country.
In these circumstances, some of the habits and expectations that persist from the earlier era of male dominance now actively work against men. One reason for men's lower enrollment and higher dropout rates in college is the persistence of the assumption that men do not need as much education as women to make a good living. And in fact, men who drop out of college do not face an initial financial penalty in their entry-level salaries compared to male college graduates. Female dropouts, by contrast, are immediately penalized, earning an average of $6,500 less than female college grads in their starting salaries. But this initial male “privilege” imposes a heavy long-term cost. By midlife, most male college dropouts earn less than college graduates of either gender.
Discrimination against women, though still powerful, works in more circuitous ways than during the era of legally and socially sanctioned favoritism for males. Today many of the economic disadvantages women face are triggered by parenthood rather than gender in and of itself, a change that becomes obscured when we reiterate the claim that women earn just 77 cents for every dollar earned by men. Childless women are now very close to pay parity with male coworkers who have comparable education and job experience. But women are still expected—and most women themselves still expect—to make the bulk of adjustments needed in raising children. They are much more likely than their male partners to quit or cut back on work when a child is young. Their husbands, by contrast, tend to increase their work hours.
This pattern greatly reduces women's lifetime earnings and increases their vulnerability in case of divorce. Still, contrary to assertions about the persistence of “the second shift,” most wives do not put in a longer workweek than their husbands when we count paid and unpaid hours. And despite the documented discrimination against mothers in hiring and promotion, most mothers who return to work do catch up with nonmothers in hourly wages (but not lifetime earnings) by the time they reach their 40s.
Michael Omi and Howard Winant (1994) argued that racial hierarchies, once imposed primarily through coercion, now rest on “a complex system of compromises [and] legitimizing ideologies” (pp. 75–76). The result, they say, is a “messy racial hegemony” marked by contradictory, conflicted, and ambiguous relationships. I think the concept of a “messy” hegemony also applies to changing gender relations, both in society at large and within individual families. Now that open racial and gender discrimination has lost legal and cultural legitimacy, the dynamics of inequality are more complex—and uprooting such inequality requires different institutional and rhetorical tools than it did a few decades ago.
To sort through this messiness, we may need to make more efforts to differentiate types of gender inequality—for example, distinguishing between structures and beliefs that penalize caregivers, those that penalize women, and those that actually aid in the exploitation of men, such as the internalized notions that studying hard is not manly, that men should start earning income rather than take on debt for college, or that “real men” don't worry about unsafe working conditions. In some cases, institutions perpetuate historical disadvantages for most women through seemingly gender-neutral mechanisms that simultaneously disadvantage many working-class men and privilege a few women. In other cases, individual men reassert traditional privileges through extralegal mechanisms that often eventually get them into trouble.
Another challenge facing researchers in this rapidly changing and diversifying family environment is figuring out how and when to draw overarching generalizations from large data sets. Psychologists have shown that data about the average response to important life events such as marriage, divorce, or bereavement can be very misleading. Most people experience little change in well-being after a loss or a gain. So when a small subset has a dramatic change for better or worse, this creates an average outcome that does not reflect the event's impact on most people—and completely obscures the experience of any even smaller subgroup whose trajectory may go in the opposite direction.
For instance, when Mancini, Bonanno, and Clark (2011) examined people's reactions to losing a spouse, they found that only 20% of the bereaved showed the conventional pattern of grieving—a sharp dip and a gradual return to preloss levels of life satisfaction. Most grievers, 60%, reported stable levels of life satisfaction both before and after the loss. Almost 15% had chronic low levels of life satisfaction before the loss, then experienced a slight worsening, and afterward returned to their previous low levels of life satisfaction. And about 5% of grievers improved following the loss. One practical implication is that we can't assume grief therapy will be helpful for everyone who suffers a loss.
Averages also obscure important variations in and interactions of race, gender, class, age, and job structures. On average, men are rewarded when they behave in assertive or dominant ways, whereas women are penalized. On average, also, Whites are favored over Blacks in hiring decisions. In elementary school, Black girls are more likely than White girls to be punished for “unladylike” behavior. But recent studies reveal that among adult career women, dominant Black women are seen as more likable and hirable than dominant White women and may even be preferred over White men (Livingston, Rosette, & Washington, 2012; Richardson, Phillips, Rudman, & Glick, 2011).
Leslie McCall (2001) showed that depending on their degree of industrial or high-tech employment and the percentage of immigrants, cities vary greatly in their relative degree of wage inequality by race, class, and gender. Gender inequality is lower than the national average in Miami, but racial and class inequality, among both men and women, are higher. In Dallas, the greatest gender-related pay discrepancy occurs among college-educated men and women, whereas in Detroit gender discrepancies are greater among the less educated sectors of the population.
The pitfalls of averaging become especially acute in times of rapid social transformation. As old rules and generalizations change, they don't necessarily change in the same way and at the same time for everyone. Today we are seeing a diversification of experiences within as well as between the many varieties of marriage and alternative family forms. The differences among male-breadwinner marriages, two-earner marriages, cohabiting couples, and unwed parents are now often greater than the differences between those categories.
In the 1950s, a one-earner marriage was almost always a sign of economic security—at least for White Americans. Among Whites—though not Blacks—two-earner couples were concentrated in lower income families, scrambling to earn enough to survive. Today, by contrast, the only place where stay-at-home mothers are a majority is among women married to men in the bottom quarter of the income distribution. The dynamics of such impoverished male-breadwinner families are very different from those of suburban families of the 1950s—and also very different from those of today's second-highest group of stay-at-home mothers—women married to men in the top 5% of the income distribution.
Similarly, it is true that most women pay a “motherhood penalty” at work, whereas married men tend to receive a fatherhood bonus. But some studies suggest that the highest female earners now get a motherhood bonus. And when fathers attempt to utilize family-friendly policies, as growing numbers of men now do, they face stigmatization and financial penalties that are roughly equivalent to those experienced by mothers.
It is particularly important to investigate changing dynamics and differences within categories if, as the historical evidence suggests, recent transformations in family and gender arrangements are largely irreversible. Since divorced families, single-parent families, cohabiting-couple families, and married-couple families are all here to stay, it is time for researchers to move away from looking at the average outcome of various family structures and relationships and spend more time studying the variations, outliers, and divergent responses within each category. It is not only more intellectually interesting to find out what kind of variables are associated with the parenting and partnering successes or failures in each kind of family, but more practically useful as well. For we must always remember that our findings are not “academic” for families trying to cope with a rapidly changing and often stressful environment.
Adapted by Stephanie Coontz from her 75th Anniversary Address to the National Council on Family Relations Annual Conference in San Antonio, Texas, on November 8, 2013.