Hilary Putnam and Saul Kripke have proposed two famous puzzles about refere
nce. These puzzles are generally thought to give us very good reasons to be
lieve that linguistic meaning has nothing to do with cognitive content. I a
rgue instead that we can give a cognitive account of them. The general idea
is that meaning includes the specification of ontological domain to which
the referent belongs. I sketch this idea through the analysis of Frege's pu
zzle about Evening/Morning Star. Actually, I maintain, theories of meaning
that argue against cognitive content are based on a normative (versus descr
iptive) approach, and it is doubtful that they are in a position to reject
any descriptive theory of meaning.