ABSTRACT
This chapter is in the spirit of Nancy Cartwright's assertion that “causality is one word, but many things”. The chapter reviews a number of approaches to causality and the methodological strategies that arise from these. In social science practice, these are often operationalised in models that depend on narrow philosophical definitions, or statistical linearity, and they consequently under-describe the complexity of the social world. An alternative view of causality, is then advanced, that is grounded in the probabilistic approach to complexity presented in the previous chapter. The conclusion, however, is a permissive one: that once we accept that causes are grounded in probability, as a feature of the social world itself, no one methodological strategy can adequately capture social causes.
