INTRODUCTORY REMARKS

In the previous chapter, the notion of the independence of theory from data was introduced and defined. This notion was intended as a partial measure of the lack of ad hoc dependence of a theory on the data it explains. It was also argued that there is reason to believe this notion will help us to explain the phenomena with which we are here concerned. And in later chapters it is indeed the notion of independence that plays the most prominent role in the explanation of the phenomena. But, as we will see in this chapter, the notion of independence, as defined, is not by itself sufficient to account for all cases of the phenomena: some more notions are needed. Neither is it, by itself, a complete account of the features of a theory that make it more likely to enjoy subsequent empirical success. The aim of this chapter is to describe and justify more of these success-conducive properties of scientific theories.

Although this chapter describes a range of different properties of theories, they all have something in common: they all indicate that a theory with these properties has an increased chance of future empirical success. Moreover, for all these properties, the reasoning justifying this conclusion is the same: it is highly unlikely that it should merely be due to chance that the data should be explainable by a theory with these properties, so it is likely not due to chance that the data is so explainable, so it is reasonable to believe there is a propensity for the data to be so explainable and that it will continue to be so explainable when observations are made of other locations within the data.