The social sciences rest at the elbow of the arts/humanities and the natural sciences. The topics addressed by social scientists are familiar to the philosopher and the musician: reason, passion, and the magnificent depths of the human experience. Artists need no formal proof of their wisdom: If the words of the poet ring true to a reader, they are true. Humanists need evidence for their interpretations and claims, but it is different from the proof of physicists and chemists, who determine causal processes by ruling out alternative explanations through the experimental manipulation of variables.
Applying the methods of science to the subject matter of the arts and humanities is a challenge. The complexity of constructs such as love, knowledge, and motivation make them difficult, if not impossible, to measure, and their manipulation raises serious ethical concerns. In the decades since my own undergraduate days, students have become increasingly sophisticated in understanding these difficulties. Although I emphasize the fact that correlation does not equal causation, the need for such emphasis decreases yearly: Bowdoin undergraduates often arrive on campus knowing that associations between measures can be interpreted in multiple ways. A researcher might suggest that A causes B, but it may actually be the case that B causes A, or that a third factor, C, causes both. My concern over this increased sophistication is that a little knowledge is dangerous, and rightful skepticism of the social sciences is a short step from unwarranted cynicism.
A fundamental question in developmental psychology concerns the relative influence of nature and nurture in shaping human behavior. There are two common misperceptions of this issue. First, although it is assumed that scientific proof can be garnered for the contributions of nature (e.g., genes, neurotransmitters) to differences in behavior, some maintain that science cannot address the role of nurture in determining these same differences. The second misperception is that nature and nurture constitute fundamentally distinct forces. A more appropriate view of development is that nature frequently acts through nurture (and vice-versa).
Consider the elegant research of Dymphna van den Boom regarding relations among "difficult" infant temperament, parental treatment, and social relationships of Dutch children. This work makes use of three primary tools used by psychologists to draw inferences about causality. The first tool is convergence. Whereas a single study provides limited support for causation, multiple replications of links between irritable behavior in children, insensitive treatment by their parents, and later problems with their peers convinced van den Boom that these factors might be causally connected.
Second, van den Boom further suggested causation by deploying longitudinal designs. Researchers identified two groups of two-week-old infants: one that cried strongly to a pediatric exam and another that expressed contentment. Over the first year, mothers' interactions with their infants were repeatedly measured in the home. During the first few months, the behavior of mothers in the two groups was indistinguishable. But the experience of caring for an irritable infant evidently took a toll on mothers, as this group became increasingly unresponsive to their children's signals. This decreasing sensitivity to infants was then linked to impaired mother-child relations at the end of the year. This pattern indicates an influence of child nature (biological factors underlying neonate behavior) upon nurture (subsequent parenting).
Mothers of difficult and easy children started out the same but became different, strongly suggesting that infant behavior caused a change in their respective parenting. However, this causal inference is still limited. More persuasive was van den Boom's next study, in which she used the most powerful tool in the psychologist's kit: controlled experimentation. Fifty mothers of irritable infants received specific training in soothing and playing with their babies, and they were compared to a control group of irritable infants and their mothers who received no training. At 9 months, intervention-group mothers were more responsive, visually attentive, and stimulating than controls. Their babies were more sociable and exploratory, cried less, and displayed more cognitively advanced behavior. A follow-up demonstrated enduring effects of the intervention not only on the children and mothers, but also on others: when viewed during interaction with same-age peers, intervention-group children were sought out as play partners more often. Thus, the intervention appeared to short-circuit the developmental trajectory by which infant characteristics alter parental treatment, which then feeds back to child behaviors, which in turn determine the quality of later relationships.
This body of research illustrates the bidirectional causal connection between nature and nurture, and exemplifies the application of social-scientific reasoning to "humanities" questions. The parenting experience is fraught with anxiety, and parents understandably desire more than folk wisdom when choosing discipline styles and soothing techniques. They desire proof derived from objective analysis. The social sciences cannot match the eloquence provided by the arts and humanities in expressing the human condition, but when carefully conducted, they are able to provide this type of proof.