This paper contrasts the scholastic realists of David Armstrong and Charles Peirce. It is argued that the so-called 'problem of universals' is not a problem in pure ontology (concerning whether universals exist) as Armstrong construes it to be. Rather, it extends to issues concerning which predicates should be applied where, issues which Armstrong sets aside under the label of 'semantics', and which from a Peircean perspective encompass even the fundamentals of scientific methodology. It is argued that Peirce's scholastic realism not only presents a more nuanced ontology (distinguishing the existent front the real) but also provides more of a sense of why realism should be a position worth fighting for.