I argue in this paper that anyone who accepts the ontology of scientific realism can only accept a pragmatic theory of truth, i.e. a theory on which truth is what it is right epistemically to believe. But the combination of realism with such a theory of truth is a form of internal realism; therefore, a scientific realist should be an internal realist. The strategy of the paper is to argue that there is no adequate semantic or correspondence theory of truth compatible with a realist ontology; that a redundancy theory cannot account for the value of truth; and that the only kind of truth theory which can account for the value of truth, and is compatible with a realist ontology, is a pragmatic theory. The kind of truth theory I wish to defend is objective and naturalistic, and the ontology is realistic. My position is, therefore, one of objective, naturalistic realism.