Taken together, the research presents a powerful challenge to the prevailing view of family change as social progress. Not a single one of the assumptions underlying that view can be sustained against the empirical evidence. Single-parent families are not able to do well economically on a mother's income. In fact, most teeter on the economic brink, and many fall into poverty and welfare dependency. Growing up in a disrupted family does not enrich a child's life or expand the number of adults committed to the child's well-being. In fact, disrupted families threaten the psychological well-being of children and diminish the investment of adult time and money in them. Family diversity in the form of increasing numbers of single-parent and stepparent families does not strengthen the social fabric. It dramatically weakens and undermines society, placing new burdens on schools, courts, prisons, and the welfare system. These new families are not an improvement on the nuclear family, nor are they even just as good, whether you look at outcomes for children or outcomes for society as a whole. In short, far from representing social progress, family change represents a stunning example of social regress.
All this evidence gives rise to an obvious conclusion: growing up in an intact two-parent family is an important source of advantage for American children. Though far from perfect as a social institution, the intact family offers children greater security and better outcomes than its fast-growing alternatives: single-parent and stepparent families. Not only does the intact family protect the child from poverty and economic insecurity; it also provides greater noneconomic investments of parental time, attention, and emotional support over the entire life course. This does not mean that all two-parent families are better for children than all single parent families. But in the face of the evidence it becomes increasingly difficult to sustain the proposition that all family structures produce equally good outcomes for children.
Curiously, many in the research community are hesitant to say that two-parent families generally promote better outcomes for children than single-parent families. Some argue that we need finer measures of the extent of the family-structure effect. As one scholar has noted, it is possible, by disaggregating the data in certain ways, to make family structure "go away" as an independent variable. Other researchers point to studies that show that children suffer psychological effects as a result of family conflict preceding family breakup. Consequently, they reason, it is the conflict rather than the structure of the family that is responsible for many of the problems associated with family disruption. Others, including Judith Wallerstein, caution against treating children in divorced families and children in intact families as separate populations, because doing so tends to exaggerate the differences between the two groups. "We have to take this family by family," Wallerstein says.
Some of the caution among researchers can also be attributed to ideological pressures. Privately, social scientists worry that their research may serve ideological causes that they themselves do not support, or that their work may be misinterpreted as an attempt to "tell people what to do." Some are fearful that they will be attacked by feminist colleagues, or, more generally, that their comments will be regarded as an effort to turn back the clock to the 1950s--a goal that has almost no constituency in the academy. Even more fundamental, it has become risky for anyone--scholar, politician, religious leader--to make normative statements today. This reflects not only the persistent drive toward "value neutrality" in the professions but also a deep confusion about the purposes of public discourse. The dominant view appears to be that social criticism, like criticism of individuals, is psychologically damaging. The worst thing you can do is to make people feel guilty or bad about themselves.
When one sets aside these constraints, however, the case against the two-parent family is remarkably weak. It is true that disaggregating data can make family structure less significant as a factor, just as disaggregating Hurricane Andrew into wind, rain, and tides can make it disappear as a meteorological phenomenon. Nonetheless, research opinion as well as common sense suggests that the effects of changes in family structure are great enough to cause concern. Nicholas Zill argues that many of the risk factors for children are doubled or more than doubled as the result of family disruption. "In epidemiological terms," he writes, "the doubling of a hazard is a substantial increase. . . . the increase in risk that dietary cholesterol poses for cardiovascular disease, for example, is far less than double, yet millions of Americans have altered their diets because of the perceived hazard."
The argument that family conflict, rather than the breakup of parents, is the cause of children's psychological distress is persuasive on its face. Children who grow up in high-conflict families, whether the families stay together or eventually split up, are undoubtedly at great psychological risk. And surely no one would dispute that there must be societal measures available, including divorce, to remove children from families where they are in danger. Yet only a minority of divorces grow out of pathological situations; much more common are divorces in families unscarred by physical assault. Moreover, an equally compelling hypothesis is that family breakup generates its own conflict. Certainly, many families exhibit more conflictual and even violent behavior as a consequence of divorce than they did before divorce.
Finally, it is important to note that clinical insights are different from sociological findings. Clinicians work with individual families, who cannot and should not be defined by statistical aggregates. Appropriate to a clinical approach, moreover, is a focus on the internal dynamics of family functioning and on the immense variability in human behavior. Nevertheless, there is enough empirical evidence to justify sociological statements about the causes of declining child well-being and to demonstrate that despite the plasticity of human response, there are some useful rules of thumb to guide our thinking about and policies affecting the family.
For example, Sara McLanahan says, three structural constants are commonly associated with intact families, even intact families who would not win any "Family of the Year" awards. The first is economic. In intact families, children share in the income of two adults. Indeed, as a number of analysts have pointed out, the two parent family is becoming more rather than less necessary, because more and more families need two incomes to sustain a middle-class standard of living.
McLanahan believes that most intact families also provide a stable authority structure. Family breakup commonly upsets the established boundaries of authority in a family. Children are often required to make decisions or accept responsibilities once considered the province of parents. Moreover, children, even very young children, are often expected to behave like mature adults, so that the grown-ups in the family can be free to deal with the emotional fallout of the failed relationship. In some instances family disruption creates a complete vacuum in authority; everyone invents his or her own rules. With lines of authority disrupted or absent, children find it much more difficult to engage in the normal kinds of testing behavior, the trial and error, the failing and succeeding, that define the developmental pathway toward character and competence. McLanahan says, "Children need to be the ones to challenge the rules. The parents need to set the boundaries and let the kids push the boundaries. The children shouldn't have to walk the straight and narrow at all times."