Causal Modelling

[Introduction] 'Causal modelling' is a general term that applies to a wide variety of formal methods for representing, and facilitating inferences about, causal relationships. The end of the twentieth century saw an explosion of work on causal modelling, with contributions from such fields as statistics, computer science, and philosophy; as well as from more subject-specific disciplines such as econometrics .and epidemiology. In this entry I will focus on two programmes that have attracted considerable philosophical attention, one due to the computer scientist Judea Pearl and his collaborators, and the other to the philosophers Peter Spirtes, Clark Glymour, and Richard Scheines. Unlike the more traditional philosophical accounts of causation canvassed in Part II of this volume, the causal modelling programmes of Pearl and Spirtes, Glymour, and Scheines do not attempt to analyse causation in terms of anything else. Nonetheless, they do establish interconnections between causal relationships on the one hand, and regularities, counterfactuals, interventions, and probabilities on the other; hence the causal modelling programmes make contact with more traditional programmes at a number of points. While the most common use of causal models is to facilitate causal inference, this application will not be the focus of this chapter. Causal inference is discussed in detail in Ch. 23. Instead, this chapter will offer a much simplified presentation of causal models that emphasizes various points of philosophical interest