I n "Family Values: The Sequel," Arlene Skolnick raises two important questions. First, is the trend toward family fragmentation-understood as a steady decline in the proportion of children growing up with their two parents-a harmful one? And second, should progressives and other people of goodwill seek to reverse this trend?
Skolnick's answer to the first question hovers somewhere between "maybe" and "probably not." Much of her essay is a defense of the idea that, from a child's perspective, divorce and unwed child bearing are not so bad after all-or at least not as bad as some people (like the two of us) say, and certainly not as bad as some other bad things, such as unemployment, low income, or parents who squabble or are unhappy. Her answer to the second question is a flat "no": "We [liberals] should have no part of efforts to hold children hostage to a narrow definition of family that looks only at form and not at love, care, and responsibility." We disagree with both of these answers. Let us briefly explain why.
On the question of whether today's disintegrating-family trend is harmful to children, Skolnick's determined optimism is increasingly rare today, both in the society at large and among the social scientists who study these issues. Indeed, her sanguine view evokes the 1970s, when the family structure revolution was still new and, in the area of sexuality and procreation, everything seemed possible. (Remember books from that era with titles like Creative Divorce?) Perhaps this time-warp factor explains why one of Skolnick's few concrete recommendations is that, when devising family policy for the next century, "a good place to start would be the old memos and reports issued in the 1970s."
Perhaps. But as Skolnick herself points out, nostalgia is no substitute for analysis. A quarter century has passed since the early 1970s. During that interval, scholars have accumulated a great deal of knowledge about the dimensions and consequences of the family structure revolution. And despite whatever nostalgia we may feel for a bygone era, surely we must recognize that, analytically speaking, we can't go back to a more innocent time.
Indeed, to examine the current academic literature in this area is to encounter again and again the story of fair-minded scholars who, on the weight of accumulating evidence, have substantially altered their previous judgments about the benign consequences for children of divorce and unwed child bearing. This list includes Norval Glenn, David Eggebeen, Lynn White, Peter Uhlenberg, Ronald Angel, Sar Levitan, and numerous others-most of whom, by the way, are Democrats.
For example, Norval Glenn, a former editor of the Journal of Family Issues now at the University of Texas, reports that "in the 1970s, the prevalent scholarly view was that such changes as the increase in divorce, out-of-wedlock births, single-parent families, and stepfamilies were benign and adaptive, if not distinctly beneficial." Yet, "by the 1990s, this view began to change, as evidence accumulated about the negative effects of marital disruption on children and about other social costs of the family changes. Not all family social scientists participated in this shift, but it is significant that the most prominent scholars and those most directly involved in the relevant research were most likely to do so."
Sara McLanahan and Gary Sandefur tell a similar story in their 1994 book, Growing Up with a Single Parent: "The idea that single mothers were to blame for producing a class of criminals, drug addicts, jobless men, and long-term welfare recipients seemed wrongheaded, given what we had learned as graduate students in the 1970s. Hadn't social scientists demonstrated that the negative effects attributed to single motherhood were really due to poverty and racial discrimination? So we thought when we began our study."
But by the 1990s, after years of careful research, they had changed their view: "in our opinion [today] the evidence is quite clear: Children who grow up in a household with only one biological parent are worse off, on average, than children who grow up in a household with both of their biological parents, regardless of the parents' race or educational background, regardless of whether the parents are married when the child is born, and regardless of whether the resident parent remarries."
Other recent large-scale studies have come to remarkably similar conclusions. Children raised outside of intact marriages are more likely to be poor, to have trouble in school, to report psychological problems, to commit violence against themselves and others, to use drugs, and to experience sexual and physical abuse. As adults, children of divorce report lower levels of satisfaction, more depression, and more physical health problems; they also, on average, obtain less education and hold less-prestigious jobs. They are also more likely to get divorced themselves and to bear children outside of marriage.
Even studies that "control" for income-a technique that tends to minimize the effects of divorce, since lower income is in itself one of the most reliable consequences of family breakdown-repeatedly find that, as Paul Amato and Bruce Keith conclude in their meta-analysis of 35 studies on divorce, " . . . parental divorce (or permanent separation) has broad negative consequences for quality of life in adulthood."
Consider also, as a matter of special importance for progressives, the well-established relationship between growing family fragmentation on the one hand and growing income inequality and child poverty on the other. Overall, children of single mothers are five times more likely to be poor than children living with married couples. They are also nine times more likely than children in married families to experience "deep poverty," with incomes of less than half the official poverty level. Moreover, poverty for these children is far more likely to endure. To take the two extremes, a child whose mother never marries is 30 times more likely to be poor for most of childhood than the child of a lasting, intact marriage. And according to McLanahan and Sandefur, a child who is not poor to begin with experiences, on average, a 50 percent drop in his or her standard of living after divorce.
Child poverty, like crime or school failure, has many causes. But surely the declining number of children raised by both parents is one of the most important. Other countries may have done a better job at narrowing the economic gap that separates children in disrupted families from those in intact families. But no family policy anywhere has succeeded in eliminating the gap. Under these circumstances, how can we fail to be concerned about family disintegration?
S kolnick answers with a familiar refrain: It's not divorce that injures children, it's the family conflict that predates the divorce. Like many others who make this claim, Skolnick relies heavily on a single controversial article, published in the journal Science in 1991. The article summarizes findings from two studies of children of divorce, one from Great Britain, the other from the United States. In her essay, Skolnick seriously misreports the results of these studies: "[C]hildren whose parents had divorced in the interim did have more problems, but they had shown those problems at age 7, before the parents had divorced." Wrong. Though margin of error considerations make these numerical estimates less than completely reliable, the British study actually concludes that pre-divorce family problems account for about half of the increased problems with behavior and school achievement experienced by boys from divorced families. The other half of the problems didn't arise until after the divorce. For girls from divorced homes, three-quarters of the drop in school achievement, and all the increase in behavioral problems, took place after the divorce.
The U.S. survey, based on a far smaller sample, also found that about half the increased behavioral problems of boys could be attributed directly to the divorce. But the U.S. study also reports that the behavior of U.S. girls actually improved after divorce, an oddly anomalous finding that was never adequately explained by the researchers (though the study's lead author Andrew Cherlin theorized plausibly that daughters of divorce tend to internalize their divorce-related problems, becoming anxious or depressed rather than misbehaving or "acting out" in ways that the study was able to measure).
Granted, high levels of family conflict are harmful to children. But Skolnick declines even to address one of the most basic questions: Has the increase in divorce over the last generation taken place primarily among such acutely troubled families? Or alternatively, are we increasingly ending marriages that, at least from the standpoint of child well-being, could and should be saved?
There is no definitive answer to this question. But many researchers have reached conclusions similar to those presented by Andrew Cherlin and Frank Furstenberg in their book, Divided Families: ". . . we doubt that such clearly pathological descriptions apply to most families that disrupt. Rather, we think there are many more cases in which there is little open conflict but one or both partners finds the marriage personally unsatisfying. . . . Under these circumstances, divorce may well make one or both spouses happier; but we strongly doubt that [divorce] improves the psychological well-being of the children."
Offering divorce as the "solution" to parental conflict is far more problematic than Skolnick realizes. Indeed, her implicit suggestion that divorce somehow ends parental conflict is one of those easy assumptions-again, this idea was very common in the 1970s-that the experience of the past three decades is forcing us to relinquish. Numerous studies of middle-class divorces, from scholars such as Eleanor Maccoby, Robert Mnookin, and Frank Furstenberg, find that "good divorces" are rare indeed. Even Constance Ahrons's decidedly optimistic book, The Good Divorce, finds that just 12 percent of divorced couples enjoy low-conflict divorces, and almost all of these couples had enjoyed close, friendly relations prior to divorce. By contrast, about 50 percent of divorced parents engage in bitter, open conflict. Good divorces, her research suggests, do not typically heal angry, conflict-filled marriages as much as they terminate relatively friendly ones.
Today's divorce rate, far from reducing the family conflict that children experience, probably increases it. For example, Skolnick ignores analyses by Richard Gill and others suggesting that, in a high-divorce society, not only do more troubled marriages end in divorce, but more marriages become troubled and unhappy. A divorce-oriented society thus generates precisely that "parental conflict" that Skolnick embraces as a reason not to worry about divorce. In addition, the divorce experience itself frequently opens up vast new territory for parental conflict, on issues ranging from late child support checks to who owns the toys. Not surprisingly, parents who cannot contain their conflicts when they are married do not usually uncover large new reserves of patience, understanding, and empathy for their former partners after divorce.
Finally, if preexisting conflict, rather than divorce itself, explains most of the problems that these children face, then children of never-married mothers ought to do better than children of divorce on psychological, behavioral, and academic measures. But they do not: Children in both kinds of single-parent homes appear to face a roughly similar set of disadvantages.
O verall, experience should teach us to be cautious about pinning all our hopes for children's well-being on the idea that we can make parents more cooperative after the divorce than they were before. Accordingly, doesn't it make sense to devote at least part of our attention to the cultural messages, corporate practices, government policies, and economic conditions that may be destabilizing marriage and thus exposing more of our children to the risks of family fragmentation?
Skolnick would like to attribute almost all of the explosive rise in single-parent homes to changing economic conditions. "The largest source of family change and family stress," she maintains, "is the shift to a postindustrial, globalized economy. . . ." To illustrate this idea, she points out that "living together unmarried-which in a legal sense counts as single parenthood-has been the low-cost way to start a family." Is that so? One might begin by asking: Low-cost for whom? For the man, who typically ends up keeping more of his money and time for himself? Or for his child and the mother of his child, who typically get less of both?
More generally, the best efforts of scholars to measure the impact of economic factors like male wages and unemployment have come to the same conclusion: Economic factors do explain part, but only a small part, of the recent decline in marriage. For example, several scholars have concluded that trends in unemployment and wage rates can explain no more than 20 percent of young black men's retreat from marriage. A recent study of the economic determinants of divorce by the economists Saul Hoffman and Greg Duncan concludes that "male incomes, [female] wages, and AFDC benefits did not play a large role in the change in the divorce rate over the past few decades and that the trend reflects primarily either changes in behavior or changes in noneconomic factors."
T o answer the second key question raised by her essay-can or should anything be done to reverse the trend of family fragmentation?-Skolnick suddenly shifts her mood. She even shifts her epistemology. If contemplating the consequences of the current family trend puts Skolnick in an unusually optimistic state of mind, contemplating the possible intensification of that trend in the future causes her to become a complete fatalist. For Skolnick, nothing-absolutely nothing-can or should be done to slow down or reverse the trend of family fragmentation, or what she terms family "diversity." The fix is in. It's here to stay. Anyone who says otherwise is actually hurting children.
Much of Skolnick's argument here depends upon our acceptance of an overly simplistic dichotomy between demographics and culture, between the "families we live with" and the "families we live by." For Skolnick, the former consist of real, flesh-and-blood human beings. The latter, on the other hand, consist only of ideas in our heads, fragments of ideology that, unless we are careful, will divert us from the material reality. From this perspective, caring about "the families we live with" means caring about "demographic changes," "economic shifts," and issues like jobs, housing, health care, and condom distribution. But caring about "the families we live by" can only mean yielding to "moral panic" and politically harmful "hysteria."
This way of describing family life-this way of describing any aspect of life-is intellectually unserious. A first principle of social analysis requires understanding the inevitable tension, in all human affairs, between the social "is" and the moral "ought"-between the material and the ideal, between how we act and what we believe. But Skolnick wants to wave this tension away, pretending that it does not exist. In her survey of contemporary family change, the "is" emerges as triumphant and utterly sovereign. The "ought" becomes at best an epiphenomenon, at worst a dangerous distraction from the core task of accommodating the "is."
This recommendation that, in Dietrich Bonhoffer's phrase, we become "servile before fact" is especially unsuited to progressives, since it represents such a sharp break with the American tradition of social reform. Sixty years ago, for example, the union organizers who founded the Congress of Industrial Organizations declined to be servile before the fact of workplace exploitation, even though the smart money and the smart analysts strongly favored current management. Thirty years ago, civil rights leaders in the South declined to be servile before the fact of segregation, even though the laws, the police, and local elite opinion overwhelmingly favored segregation-which was, in those days, one of the "the customs we lived with." Similarly, there is absolutely no reason why progressives today should embrace Skolnick's advice that we become servile before the fact of family fragmentation-unless, of course, we believe that family fragmentation is a good thing.
We reject the pessimistic, even reactionary, claim that current rates of family disintegration constitute an untouchable fact before which reformers must scrape, bow, and make excuses. We reject as well the charge that concern for the health of marriage as an institution somehow signals a lack of compassion for children growing up in single-parent homes. We concede that, in the relatively short time since today's scholars and reformers have begun to focus seriously on this problem, no one has come up with a perfect or definitive solution for reversing family fragmentation and fostering hands-on fatherhood, while at the same time defending and strengthening the ideal of equal regard between men and women. But anyone who cares about the prospects for our children must surely confront these challenges in the years ahead.
ARLENE SKOLNICK
RESPONDS
M aggie Gallagher and David Blankenhorn respond as if we were on Crossfire, and try to force the argument into a falsely polarized either/or mold. If they are "against" "family fragmentation," I must be "for" it. The result is a caricature of what I wrote and a distortion of the social science literature. They pounce on any evidence that divorce or family structure affects children's well-being, then discard or downplay evidence of other factors that may be even more important.
Most researchers take a shades-of-gray position on family structure. The evidence does show that children in divorced, remarried, or unmarried families are at greater risk for a number of problems, but there is little support for the frightening picture of such families painted by these authors and their colleagues. The vast majority of children in single-parent families turn out reasonably well. Alan Acock and David Demo, who examined a nationally representative sample of children and adolescents in four family structures, reported "few statistically significant differences across family types on measures of socioemotional adjustment and well-being."
It is true that other researchers have found higher rates of divorce-related problems. Mavis Harrington observes, for example, that about twice as many children from divorced families have behavioral problems as those from continuously intact families-20 percent to 25 percent, she estimates, as opposed to 10 percent. "You can say 'Wow, that's terrible,'" she comments, "but it means that 75 to 80 percent of kids from divorced families aren't having problems, that the vast majority are doing well." Sara McLanahan, who also protests the use of her data to scapegoat single parents, makes a similar point. Of course the doubling of risk is worrisome, but it's important to keep in mind that many of the problems experienced by children from divorced or never-married families are caused by poverty, lack of education, psychological problems in the parents, poor parenting skills, and other preexisting factors.
D espite their insistent claims of concern for children, Gallagher and Blankenhorn dismiss factors having a greater and more direct impact on children's well-being than family structure. For example, study after study shows that the key determinant in a child's adjustment is the quality of the relationship with the primary parent; Janet Johnston of Stanford has found that even in hostile, high-conflict divorces, a good parent-child relationship can buffer a child from the effects of a difficult environment.
Still, family conflict strongly affects children's development. Gallagher and Blankenhorn reveal the surprising limits of their familiarity with the research literature when they write, "Like many others who make this claim, Skolnick relies heavily on a single controversial article, published in the journal Science in 1991." It's hard to understand how anyone who prescribes national policy on the basis of the social science evidence can ignore-or, worse, fail to know about-an entire body of research, dating back to the 1950s, showing that marital conflict is more strongly linked to adjustment difficulties among children than to the marital status of their parents. Nobody should divorce casually, but the question is, Divorce compared to what? An angry or conflict-ridden marriage can be more harmful to children than a loving single parent.
Recent research shows that a widespread but far quieter problem-parental depression-is also a significant risk factor for child development. And depression, particularly in women, can result from marital unhappiness. As Richard Weissbourd observes, "Whether parents are chronically stressed or depressed often more powerfully influences a child's fate than whether there are two parents in a home or whether a family is poor."
In short, whether a family consists of two parents is less important for a child than how well the family (in whatever form) functions. Of course, it would be better for children if more marriages were successful. But rather than adopt policies to make divorce more difficult (which may actually discourage marriage in the first place), we ought to address the problems afflicting children in all our families.
