Full Idea
We have no need to turn to an a priori explanation of our knowledge of mathematics and logic. Our intuitions that this knowledge is not justified in some direct empirical way is preserved. It is justified in an indirect holistic way.
Gist of Idea
We explain away a priori knowledge, not as directly empirical, but as indirectly holistically empirical
Source
Michael Devitt (There is no a Priori [2005], §2)
Book Reference
'Contemporary Debates in Epistemology', ed/tr. Steup,M/Sosa,E [Blackwell 2005], p.107
A Reaction
I think this is roughly the right story, but the only way it will work is if we have some sort of theory of abstraction, which gets us up the ladder of generalisations to the ones which, it appears, are necessarily true.