Understanding Normative Concepts in Accounts of Causation
Metaphysics, Philosophy of Science, Normative Theory
My current research concerns the recent trend of employing normative concepts in counterfactual accounts of causation.
David Lewis’s counterfactual account of causation has had a tremendous impact on the field. According to Lewis, for one event to be the cause of another event it must be the case that a chain of counterfactual dependence relations holds between the two events. This account is simple and speaks to our intuitive understanding of causation.
However, this account has been subject to an immense amount of criticism that has exposed a number of serious problems. Consider the following example, suppose Alice is going on holiday and gets Bob to promise to water her plant while she is away. However, Bob is lazy and neglects to water the plant, so it withers and dies. Bob is clearly the cause of the plant’s death and the plant’s death is counterfactually dependent upon Bob’s not watering it. But then suppose that there is a third person, Carlos, who doesn’t know about the plant but who still could have watered it. The plant’s death is also counterfactually dependent on Carlos’s not watering it, but it doesn’t seem like the latter is a cause of the former. Lewis’s theory doesn’t seem to have any way to account for this difference.
The natural response to this problem case is to say that the difference lies in what we judge to be normal. It is normal for people to keep their promises and it is abnormal for people to go around watering plants for no reason. This is the route taken by Joseph Halpern and Christopher Hitchcock who argue that our causal judgements involve an appeal to a normality ordering of possible worlds as well as an idea of a default course of events. The hope is that incorporating these normative notions into an account of causation will solve a number of existing problems with the counterfactual account.
My issue with this is the lack of any detailed understanding of these normative notions. The proponents of these normative accounts will use words like “normal” and “default” without really spelling out what it is that those words mean. My aim is to provide an in-depth understanding of these notion and to then examine the effects this will have on our accounts of causation.