Pluralistic Paradigms

See allHide authors and affiliations

Science  12 Apr 2013:
Vol. 340, Issue 6129, pp. 146
DOI: 10.1126/science.1236791

Irreconcilable differences have characterized the scientific study of human behavior for decades, even centuries. For some scientists and philosophers, like the late Thomas Kuhn, this inability to agree signaled the immaturity of the field. Preparadigm science, Kuhn held, “is regularly marked by frequent and deep debates over legitimate methods, problems, and standards of solution, though these serve rather to define schools than to produce agreement” (1). By way of contrast, Kuhn posited that mature science (characterized by normal, puzzle-solving activities) consisted of communities of scientists who worked unproblematically within a single research paradigm. Yet, as Helen E. Longino cogently argues in Studying Human Behavior, we would be foolish to assume such differences will vanish anytime soon. By rejecting Kuhnian expectations that one paradigm in the human behavioral sciences could ever come to dominate all research, Longino hopes to reconcile scientists, philosophers, and the popular press to the continued plurality of research programs in the study of human behavior as constituting a normal, and successful, scientific practice.

Longino (a philosopher of science at Stanford University) lucidly dissects five areas of active empirical research on the causes of human behavior: quantitative behavioral genetics, social-environmental approaches, molecular behavioral genetics, neurobiological approaches, and more integrative approaches (for example, developmental systems theory and multifactorial path analysis). As a group, scientists in these areas concentrate on the proximate causes of behavior rather than evolutionary or functional explanations [as Ernst Mayr would have put it (2)]. And they seek to understand the biological structure constraining the expression of behavior rather than exploring the immediate physiological “triggers” of an instance of a behavior. Longino's choice of research programs is doubly important: given these similarities, one could reasonably expect scientific agreement, and her researchers, from the outset, are invested in describing the genetic and developmental causes of behavior as it is expressed in the lifetime of individuals.

These differences in research questions correlate with similar disagreements over how to define or measure behavior. Take the two families of behavior on which Longino focuses. She notes that only some kinds of aggression are studied and that researchers vary widely in how they measure aggressive behavior, including through rates of “violent crime,” frequencies of “fighting” or “hitting a doll,” or even scores on psychological tests. When considering sexual orientation, behavioral researchers often follow Alfred Kinsey in defining homosexuality as a function of the frequency of sexual encounters with members of the same sex (3). But they vary in whether they consider homosexuals part of a sexual binary (as paired against heterosexuals) or existing on a sliding scale of sexual desire. Longino acknowledges that these definitions are set by social criteria. Scientists may only consider those aggressive acts that are reported to the police (surely a subset of the behavior, however defined). Kinsey's sexuality scale emerged within a specific sociocultural context and may not be useful in defining homosexuality in other cultures (globally or temporally). In both instances, there is no unitary phenomenon under investigation. Thus, despite appearing to ask the same question, “what causes human behavior?” even the five ontological approaches Longino explores are irreconcilable because they operate in differentiated causal spaces.

Discussions of human behavior in the media tend to pit one of these “explanations” against another and ask which better accounts for our foibles and actions. Longino argues that by relying too heavily on the nature-nurture debate and framing the conversation about human behavior in terms of competing research programs, the popular press has “fuel[ed] anticipation of intervention at the level of the individual”—providing (on aggression) superficial scientific support for the American incarceration culture and (on sexual orientation) contributing to a changing conversation about homosexuality as a civil rights issue. The political implications of research on human behavior are never clear-cut because they rely on a conflation of an individual's behavior with her or his identity.

One possible escape from this dilemma, Longino posits, may be to think of behavior as social. After all, most aggressive or sexual acts involve more than one person and take place in a specific cultural context. This is the one place I feel her careful analysis falls short. After she meticulously examines how scientists' choice of ontological subjects affects their analyses, she is less attentive to the consequences of her own choices. If she had selected a wider swath of behavioral research from the outset (including, for example, ethological perspectives), Longino may have seen less of an individualist bias in both the professional and popular literature (of course, such a choice would have made her project far less manageable). But this is a minor quibble with a fascinating book.

As a result of her commitment to methodological pluralism, Longino refuses to advocate for one research strategy over another. For human behavior, the obvious poles of the debate are defined by nature and nurture. But, she notes, all sides of the nature-nurture debate already agree that both genetic and environmental factors are important. So the far more interesting question is “how should scientific research on human behavior inform discussions of social policy?” In Studying Human Behavior, Longino has clearly articulated the methodological plurality of research on human behavior. She leaves to scientists and science writers the responsibility of shedding the last vestiges of their Kuhnian expectations.



Navigate This Article