ABSTRACT
I think that the first paper of Nancy Cartwright’s that I ever read was ‘Causal Laws and Effective Strategies’ (Cartwright 1979, [1983]) shortly after it appeared nearly twenty-five years ago. At the time I was still in a deep dogmatic Humean slumber about causation, and my first reaction was that the central claims of the paper about the irreducibility of causal laws to what Cartwright called laws of association couldn’t possibly be right. But, like a number of other philosophers of science at about this time, I had independently decided to try to learn something about the so-called causal modeling techniques widely used in the social and biomedical sciences. I assumed that these techniques would have something interesting-at that point I didn’t know what-to teach philosophers about causation and its relation to probability. As I worked through this literature, and began to appreciate the causal character of the additional assumptions that were required to get causal conclusions out of statistics, it gradually dawned on me-I was a slow study-that Cartwright was absolutely right about the issue of irreducibility. ‘Causal Laws and Effective Strategies’ was a brilliantly original paper that fundamentally changed the thinking of many of us about causation.
