This chapter looks at Donnellan’s referential–attributive distinction from a communication-theoretic perspective, which distinguishes between utterance production and utterance interpretation – in this case between the referential and the attributive use of definite descriptions and their referential and attributive interpretation. The framework is MSDRT, an extension of DRT that provides mental state descriptions (MSDs) for utterance producers and recipients. MSDs consist of propositional attitude representations (PRs) and entity representations (ERs). ERs represent entities from the outside world (their referents), to which they are linked by causal relations and which they can contribute to the contents of the agent’s PRs. The referential use and interpretation of a description are analyzed as those which producer and interpreter take to refer to the referent of one of their ERs (while the attributive use and interpretation take it to refer to whatever satisfies its descriptive content). This approach differentiates more finely between different use scenarios than other approaches and throws new light on the question whether the referential and the attributive use are mutually exclusive and whether they are jointly exhaustive.