The Semantic Solution to the Problem of Induction

When we say that an inductive inference is reliable, this means that it is an induction performed according to our current inductive practices. Hence those practices are, by definition reliable.

Alternatively, we are rational when we use induction because that is part of what 'rational' means.


One account is given in Edwards 1949. The bibliography of Swinburne 1974 gives further references.


The solution doesn't seem to do anything. All it shows is that we normally assume that induction is reliable, even in our normal use of language. We want to know whether induction really is reliable.