In a Recent article [9], Alexander Rosenberg attacks the “manipulative” view of causation as being unilluminating and as being beset with difficulties. As a proponent of that view [3], I have felt it necessary to take up cudgels in its defence.Rosenberg's criticisms are directed at Gasking's version of this view [5]. Gasking's recipe for causation is, “one says ‘A causes B’ in cases where one could produce an event of the A sort as a means to producing one of the (...) B sort”. Here it is taken for granted that we have general manipulative techniques for producing events of the A sort and that this is implicitly assumed when one speaks of A as a cause. There is one qualification to this formula for causes which arises from the necessity for individuating processes, namely, that, for A to be the cause of B, it must not always lead to B, but only in special cases. (shrink)