Dramatic changes in family life in the latter half of this century have meant that seven in 10 mothers now work in the paid labor force, the majority of two-parent families are now dual-earner families, and three in 10 households are now single-parent households (Hochschild, 1997; U.S. Bureau of the Census, 1995). Despite these revolutions in family life and despite continuing efforts for gender equality, the task of childrearing remains primarily the responsibility of mothers. The fact that 88% of single-parent households are headed by women provides one obvious marker of this reality (U.S. Bureau of the Census, 1995). Research on dual-earner families provides further confirmation: Mothers, on average, are responsible for 74% of the total parental hours spent in direct child care (Ishii-Kuntz & Coltrane, 1992), None of this information is news to the majority of readers of Journal of Marriage and the Family. But what we should find surprising in this context are the extraordinarily demanding childrearing prescriptions and maternalist assumptions represented by the recently published "Parents' Socioemotional Investment in Children" (Bradley, Whiteside-Mansell, Brisby, & Caldwell, 1997). Bradley et al. participate in the perpetuation of a series of powerful yet outdated cultural assumptions regarding the "proper" relationship between mothers and children.
What Bradley et al. ultimately offer us is a series of taxing and exacting prescriptions for completely selfless, constant mother care. These demands, however, are cloaked in gender-neutral verbiage and the language of science and objectivity. Bradley et al.'s explicit aim is to test the validity of a new research instrument they designed, the Parental Investment in the Child Questionnaire, referred to as the PIC. Many of the questions on the PIC are drawn directly from extant forms that measure parent-child "attachment," and all the items on the PIC are derived from existing psychological theories of appropriate parental behavior. The PIC is, nonetheless, distinct and particularly useful, the authors explain, because it is expedient and efficient and because, unlike most research questionnaires in this genre, the PIC measures the parents' experience of attachment, rather than the child's experience of attachment. For these two reasons, Bradley et al. expect the PIC to be useful in parental education programs. After providing a theoretically grounded rendering of the importance of parents' investment in children, the authors dedicate most of the article to laying out their methodology and statistically demonstrating that the PIC is, overall, internally consistent and reliably correlates with connected measures of parent-child attachment and measures of mental, familial, and social health. Hence, Bradley et al. conclude, the PIC is a "valid" as well as useful measure of parents' socioemotional investment in their children.
At first glance, this looks like good science. As a research instrument, the PIC follows directly from the scientific findings of its predecessors. The authors appear to have a solid grasp of the "objective" research in this genre, and the statistical methods they use for testing the consistency and validity of the PIC appear sound. But there are three central problems with the argument made by Bradley et al.: one regarding the neglect of existing methodological and theoretical critiques of attachment theory and two involving the cultural assumptions and social implications involved in the development and use of the PIC.
First, Bradley et al. do not address the longstanding criticisms of attachment theory that arise out of debates within psychology and that have been rendered and extended in the work of Diane Eyer (1992). Second, sociologists and others will notice that the PIC prescribes a model of parental behavior so demanding that it would be extremely difficult for even the most dedicated stay-at-home parent to follow and virtually impossible for single paidworking parents or dual-earner couples. …