Simplicity as a Criterion of Induction

Extract

There is now a well-established distinction recognized in the ways simplicity considerations enter into science. Laws of nature (and theories built out of laws of nature) may be graded either with regard to their simplicity of form or with regard to the fewness of the concepts employed to express them. I shall distinguish these as formal simplicity and conceptual simplicity respectively. Dr. J. O. Wisdom suggests that it should be fewness of non-instantial concepts that serves as the guide for making judgements of relative simplicity; a “non-instantial” concept being a concept which is constructed for the purpose of theory, its meaning not being derived simply from experience. Hence non-instantial concepts are, from a logical point of view, to be contrasted with the concepts used to express and record observations directly, that is with instantial concepts representing observables. This seems to be too narrow a restriction on the reduction of concepts. Clearly an advance towards simplicity has been made not only when e.g. a non-instantial concept like “force of gravity” is eliminated from our technical language, but also when the number of instantial concepts is reduced by redefinition, in terms of a smaller number of observables. For example a great simplification was surely made when space-traversed, time taken and mass were chosen as the dynamic “simples”, for the multiplicity of kinds of motion were found to be expressible as functions of these concepts. “Motion” is not non-instantial in the same way as “force of gravity” or “kinetic energy” is non-instantial. However, once this extension to instantial concepts is made there is no need to quarrel with the form/concept distinction as marking a pritma facie difference in criteria of some methodological importance; for a choice say to concentrate on using the fewest concepts does, and one would be inclined to think must, lead to a “decrease” in the formal simplicity of the laws of nature.