Causal necessity

Hume's treatment of the problem of natural necessity dealt mainly with the relationship between cause and effect. He argued that all reasoning concerning matters of fact is ultimately founded on this relationship. So, he thought, the more general problem of justifying all sound reasoning of this kind could be solved if the more specific one of justifying reasoning from cause to effect could be. What then, he asked, is there about this relationship to justify such an inference? Is there, perhaps, some kind of necessary connection between causes and effects?

On this question we have conflicting intuitions. On the one hand, it seems obvious that the things we do and the events that occur in nature have effects, and that these effects are somehow produced, or brought about, by these actions and events. The effects would seem to be not just subsequent happenings, but, in most cases, inevitable consequences of the actions or events that give rise to them. On the other hand, the effects that are produced do not seem to be necessitated by their causes in the strong logical sense in which, say, adding one to an even number produces an odd number, because the contrary of any cause and effect relation is easily imaginable, “as if ever so conformable to reality” (Hume 1777: 21). Therefore, if causes necessitate their effects, then they must do so in some weaker, non-logical, sense.