A key feature of language is its use to report what people have said. In languages like English this can be done by using either direct speech (literal repetition: Corien said: 'I am tired') or indirect speech (Corien said that she was tired). This difference between direct and indirect reports and the related language-philosophical distinction between mention and use has long been considered fundamental to language. With its six different options for reporting speech, Ancient Greek challenges this traditional distinction and falsifies current semantic theories of reports. Strikingly, the language has never figured in the discussion about reports. The present project aims to fill this gap by putting Ancient Greek reports at the center of attention. Using a corpus of Ancient Greek texts, I determine the semantic and pragmatic differences between the various options to report speech. The resulting typology will uncover the problems for existing theories of reports and lead to the development of a new theory, which can handle both Ancient Greek and languages like English. With its multidisciplinary approach this project advances our understanding of Ancient Greek and at the same time delivers an entirely new view on foundational issues in philosophy of language and formal semantics.