On Methodology
The scientific method is often presented as proceeding stepwise from hypothesis to testing to conclusion. Yet this simplified view fails to characterize in full the nature of the scientific enterprise. The primatologist Frans de Waal (2001) has concisely presented one fault in this approach:
The urge of behavioral scientists to proceed in a straight line from theory to data, hence presenting themselves as more naïve to the truth than they actually are, derives from a desire to be like physics, a science that has reached the lofty stage of armchair prediction. … It is unclear whether the behavioral sciences will ever reach the point when logically derived predictions drive progress. Behavior is more variable than the dance of photons, and its explanation involves multiple layers, from the physiological to the mental. We cannot afford to look through a single pair of glasses; we need lots of different glasses to see reality. Theories do assist in this effort, by guiding our attention and making large amounts of data graspable, but they also induce selective blindness. (182)
The first point is that hypotheses must derive from somewhere. In large part they emerge from a researcher’s training, experience and interests. Yet hypotheses are often the direct result of observation and supposition. They are not, in most cases, purely objective matters. Rather they reflect the opinions and assumptions of the researcher. This in itself is not a bad thing. The danger lurks in failing to recognize or acknowledge this.
As de Waal’s metaphor of multiple glasses implies, a multidisciplinary approach is most likely to succeed when studying the complex matter of behavior. Anthropology and ethology tend toward observational techniques, while seeking to disturb the natural flow as little as possible, with the risk of an ever-shifting focus. Experimentalism contributes to our ability to investigate underlying features in isolation and in fine detail, but at the risk of unnaturalness. The former might be seen as viewing the forest; while the latter is pinpointing spots on the bark of the trees.
One is not inherently stronger than the other. While each researcher may lean toward any given approach, I see this as reflecting their style, at best not limiting their scope. But methodology must serve the ultimate goal of answering questions. The questions themselves ought therefore drive the choice in methology. Each tack provides its own strengths and pitfalls. Borrowing techniques from various fields, and applying them across these domains, we are best able to concentrate on these questions.
References
de Waal, Frans. The Ape and the Sushi Master. New York: Basic Books. 2001
