This article examines how and why participants in conversation modify the reported speech made by another speaker, and proffer a second saying in subsequent talk. Two types of modified reportings are identified: (1) explicit modifications, and (2) implicit modifications. Both types are shown to treat the first reported speech as inadequate, and lay claim to speaker’s epistemic authority or priority over the reported talk vis-a-vis other interlocutors. Further, explicit modifications of a prior report generally display a stronger epistemic stance than implicit modifications.