The Science of Demand (10) - Unofficial Translation of Steven Cheung's 经济解释 - 科学说需求

Section 2.  Theory to Restrain Behavior

Let’s not forget that a theory with explanatory power must be refutable by facts. A non-refutable theory is not useful at all. Of equal importance, to predict behavior, behavior must be scientifically restrained. If behavior is not restrained at all so that it can behave in whatever way like a non-directional wind and no prediction will ever be wrong, the theory is then not refutable by facts.

Behavior must be restrained, like specifying under what circumstances a person will turn left instead of right. Only so can behavior be inferred and explained. Certainly, specifying turning right could turn out to be turning left. A theory with explanatory power is potentially refutable by facts but has not yet been refuted. This has been explained in Chapter I. Restraining behavior increases the possibility of refuting the theory. The more restraints there are, the more precise is the inference of behavior. But then the possibility of the theory being refuted gets higher. Science is therefore a risky game. The more restraints on human behavior the better, though they must not be extended to the territory of refuting the theory. Masters in science are courageous yet cautious enough to make bold assumptions alongside careful review, extending the arbitrary boundary of restrained behavior to the brink of just not refuting the theory.