ABSTRACT
When we talk about the relation between theory and practice, we usually assume that the science comes first and enables the solution of practical problems. Thus, we speak of “applied science” and have no equally common noun phrase for direct attack on practical problems by scientific means, although “engineering research” comes close. Yet many historians, philosophers, and commentators have concluded that the historical path is more often from practice to science than from science to practice (Kuhn, 1977; Mokyr, 1990; Petroski, 1982). Technology usually advances by incremental trial, error, decomposition, simulation, and improvement, with general principle discovery and scientific theory occasionally sprouting from the process and occasionally helping to solve future practical problems. The Wright brothers perfected their wing shapes by hundreds of trials in a wind tunnel, not by calculations from aerodynamic theory. Aerodynamic theory mostly came later, to explain why the wing shapes that Wilbur and Orville chose had adequate lift and stability and to suggest new versions that, even today, need wind-tunnel testing before major investment. Fisher invented analysis of variance to help agronomists select among seed varieties, not to help biologists—much less psychologists—prove theories. On the other hand, the needs of practical problems often drive, or at least stimulate, science and theory. Navigation motivated astronomy; artillery and commerce motivated geometry and physics; medicine motivates molecular biology. Scientists try to discover the biochemical, cellular, genetic, or physiological processes that account for disease states in order to support rational cures for well-known diseases more often than they take discoveries from pure science and seek diseases to which to apply them. Sometimes science is stimulated by failures of practice, by the appearance of unsuccessful or dangerous technology. The collapse of large numbers of early iron bridges was eventually followed by scientific investigation of the physical properties of iron beams (Petroski, 1982).
