My goal in this paper is to make progress toward a unified theory of intentionality that does not fall prey to either of the most serious species of puzzles facing any theory of intentionality; indeterminacy puzzles (e.g. modal-theoretic arguments, the qua problem, rule-following problems) and puzzles concerning thought and talk directed at things that, prima facie, do not exist (e.g. empty names). Much attention has been payed to these two types of puzzles separately, but it is fruitful to consider how a theory of intentionality could simultaneously handle both sorts of puzzles, since many of the resources appealed to in solving puzzles of one kind often do not help (and sometimes hinder) our efforts to solve puzzles of the other kind, or so I will argue. I will then propose a novel account of intentionality that can handle both kinds of puzzles. The central idea is to reduce intentionality (the relation supposed to stand between thoughts and linguistic items, and their objects) to relations of intentional identity (the relation of being about the same thing that stands between the thoughts and linguistic items themselves).