The proposition that every vector space has a basis is equivalent to the
Axiom of Choice (the implication from AC is easy: apply Zorn's Lemma to the collection of subspaces-with-a-basis ordered by inclusion -- this is due to Hamel). There is no 'natural' basis for
R over
Q and the existence of one must surely require some weak form of Choice: it is an open question whether it is equivalent to AC. See Peter Johnstone, "Notes on Set Theory and Logic", Cambridge University Press.
Incidentally there's some confusion in the use of the term
Hamel basis: some authors use it precisely to mean a basis for
R over
Q whereas others use it to mean a basis in the algebraic sense as opposed to a
Schauder basis in a topological vector space which has the property that every element is uniquely a possibly infinite, but convergent, sum.