ABSTRACT
The focus of this chapter is on how we might discriminate between theories and how we might develop a theoretical framework from individual theories, that might then provide viable and testable inferences about deeper realities. It begins by reviewing the work of Hanson, Kuhn, Popper and Lakatos on the role of theories in scientific inference. From this two concepts are deemed useful: Lakatos's “research programmes” and a modified version of Popper's fallibilism. Two other concepts are described: inference to the best explanation and reflexivity. These, along with falsification are the key elements in a realist methodological framework, described as “reflexive fallibilism”.
