I do use the principle of induction, even though it can't be rationally justified; that's why I call it an axiom. The same thing goes for the assumption that thinking is any good.

I'm slightly at a loss - I can't now see how you can't see that your justification of the principle is circular.