This metaphysical separation of ‘structures’and ‘causal mechanisms’ from their empirical
‘effects’ enables critical realism to endorse a sophisticated form of causal analysis. Perhaps most notably, realists reject the Humean or successionist position that ascribes causality to observed regularities in the domain of the empirical (Pratt 1994; Sayer 2000). Instead, they suggest that the relationship between causal mechanisms and empirical outcomes is contingent:that is to say it is neither necessary nor impossible for ‘cause’ and ‘effect’ to coincide in time and space. It follows from this that realist explanations require not only the identification of causal mechanisms – the inherent properties of objects that have the power to produce or force change – but also the discovery of the precise conditions or circumstances under which these have been activated. This
demonstrates the importance of ‘context’ (spatial
and temporal) to realist causal analysis.
However, it also requires the drawing of a
distinction between the necessary properties
of social objects (which exist wherever those
objects exist and are, therefore, generalisable
and predictable) and the fortuitous contingencies
that shape their form and/or influence at
specific times and places (which cannot be
known about in advance of concrete empirical
research).