Full Idea
On the whole it is not worthwhile preserving the word 'cause': it is enough to say, what is far less misleading, that any two configurations allow us to infer any other.
Gist of Idea
We can drop 'cause', and just make inferences between facts
Source
Bertrand Russell (The Principles of Mathematics [1903], §460)
Book Reference
Russell,Bertrand: 'Principles of Mathematics' [Routledge 1992], p.486
A Reaction
Russell spelled this out fully in a 1912 paper. This sounds like David Hume, but he prefers to talk of 'habit' rather than 'inference', which might contain a sneaky necessity.