Karl Popper and the Social Sciences

BUILDING MODELS

approaches is so great that the unity of scientific method can only be retained by describing methodology at a highly abstract (and therefore largely uninformative ) level. But in this chapter, I want to present the concept of situational analysis as proposed by Popper, including its relationship to other Popperian ideas on social inquiry, especially his support for methodological individualism and his rejection of psychologism and methodological collectivism. The following discussion will draw mainly upon Popper’s lengthiest and last sustained explanation of situational analysis—his “Models, Instruments, and Truth” essay. However, I will also draw liberally upon Popper’s other discussions of situational analysis and social science generally. BUI LDI NG MODELS Popper begins his discussion of situational analysis by positing that the fundamental goal of science is problem solving and that there are, broadly speaking, two types of problems in need of explanation: singular events and types or kinds of events (MF, 162–166; PS, 357). Explaining a singular event—such as the collision of Shoemaker-Levy 9 comet with Jupiter in 1994, the eruption of Mount St. Helens in 1980, the French Revolution, or the near collapse of Asian economies in 1997—merely requires identifying some relevant initial conditions along with some universal laws in order to predict (or retrodict) and explain the event. For example, to explain Shoemaker-Levy 9’s collision with Jupiter, one would need to identify such initial conditions as the position, mass, and velocity of the comet and other celestial bodies at successive points in time, combined with some relevant universal laws, including gravity and Newton’s laws of motion. Explaining a kind or type of event—that is, an event that recurs in a more or less predictable pattern—requires a somewhat different approach, Popper says. Examples of types or kinds of events would be lunar eclipses in general (not last month’s eclipse), cycles of economic expansion and recession (rather than the U.S. recession in 1991–92 and the following expansion), political revolutions in general (not the French Revolution or the American Revolution or the Iranian Revolution). The best way to explain types of events, Popper suggests , is to construct a “model,” which, he says, is merely a simplified representation of reality. Being a simplification of reality, it will of necessity be a false depiction of reality. For instance, in order to simplify calculations, a model of the solar system might assume that the various planets are points and that comets and other extraplanetary objects have no gravity, even though such assumptions are plainly false. No model can incorporate all elements of the phenomena to be explained, nor would such a model be desirable. Rather, a good model represents the most important features of reality, given our explanatory interests. Popper acknowledges that there is probably no formal way to state beforehand how those features should be selected; rather, a model’s value will ultimately be proved by its usefulness. “I think we have to 6 KARL POPPER AND THE SOCIAL SCIENCES admit,” he says, “that most successful scientific theories are lucky oversimplifications ” (MF, 171–172). However, the elements or structural features alone of a model are not enough to explain a typical event. To “animate” the model, Popper says, we need universal laws. Thus the planets in a model of the solar system are set in motion by Newton’s laws of gravity and momentum, and a model of an atom is animated by the strong and weak forces, and electromagnetism. No model can do without animating universal laws, Popper claims, for we can “never reduce animating laws to structural properties of the model” (MF, 164).This is not to say that we can never offer a deeper explanation of a universal law by developing a model of the law itself—a mechanistic description of the elements and structures that explain how the law operates and produces its effects. In fact, Popper encourages such mechanistic reductions; indeed, he says, they are an important goal of science (RAS, 134). Popper’s point is rather that a model, no matter how fine-grained, can never animate itself, for new, deeper laws will be required to set it in motion and the process will begin anew. For Popper, there are no ultimate explanations that are “neither capable of any further explanation , nor in need of it” (OK, 194). This is one way of characterizing Popper’s anti-essentialism, which claims that there can be no explanation of phenomena that is self-evident, intuitive...