I introduce the notion of a 'control variable' which gives us a way of seeing how mental causation could be an unproblematic case of causation in general, rather than being some sui generis form of ...