This article argues that the end of Tacitus's Dialogus de Oratoribus is inconclusive in ways that draw attention to the difficulty of interpretation not only of the dialogue, as by modern scholars, but also in the dialogue, as by its leading characters. The inconclusiveness is especially marked by a commonly noted, but little discussed, feature of the end: when the rest of the characters laugh at the point of departure, Tacitus himself does not. Arguing that this difference of affective response (...) on the part of the characters prefigures differences in interpretive response on the part of readers, the article identifies different strains in recent scholarship: pessimistic and optimistic. Both forms of response entail an attribution of a “poetics of conspiracy” (Hinds) to the ultimate speaker of the dialogue, the author Tacitus, and a “hermeneutics of suspicion” (Ricoeur) to its reader. At the same time, the author's double-position, as character and author, between narrated event and narration of the event to the reader, suggests that the other characters in the dialogue may, like the author and reader, also exercise such poetics and hermeneutics on one another and themselves. The article ends with the comparandum of the first satire of Tacitus's near contemporary, Juvenal, suggesting that, in the case of these works that can look with hindsight on the social and political past of the Early Empire, their modes of transmission and reception may be politically determined (e.g., as conspiratorial, suspicious) but may also demonstrate, within the restrictions of social and political determinations, a high degree of contingency, reflexivity, and autonomy. Such possibilities suggest that the text itself is part of a pragmatic and performative tradition of the kind enacted by its characters, in addition to a tradition of the production of (comparatively static and unfree) “literary” works. (shrink)