Time series, nonsense correlations and the principle of the common cause

This paper examines some recent defences of the principle of the common cause (PCC) against Elliott Sober’s famous counterexample. There are two lines of attack: attempts to defuse the counterexample, that is, to show that the scenario described by Sober only apparently conflicts with the PCC; and attempts to demonstrate that the counterexample has no practical consequences. I show in this paper that there are problems with both strategies. In response, I formulate an alternative version of the principle that avoids the known counterexamples and that makes its status as fallible epistemic principle explicit.