Also found in: Dictionary, Thesaurus, Legal.
- a relation between scientific theories in which the propositions and overall content of the theories cannot be directly compared.
- a conception of scientific theories which holds that all observations are theory-relative (see THEORY-RELATIVITY), and that there may exist no theory-neutral data language of the kind assumed by inductive, logical positivist or falsificationist conceptions of science. The conception is most associated with Thomas KUHN and Paul FEYERABEND and it is often assumed to also imply a more general RELATIVISM. However, this is not Feyerabend's view: incommensurability is seen as a possible, not a necessary, relation between theories. If theories cannot always be strictly compared in terms of a theory-neutral data language (or using any unambiguous or unchanging decision rule), the proponents of competing theories can enter into a dialogue with the aim of appreciating each other's view and reaching a decision on this basis. See also TRUTH. POSITIVISM, FALSIFICATIONISM.