The notion of Turn-Constructional Unit (TCU) in Conversation Analysis has become unclear for many researchers. The underlying problems inherent in the definition of this notion are here identified, and a possible solution is suggested. This amounts to separating more clearly the notions of TCU and Transition Relevance Place (TRP). In this view, the TCU is defined as the smallest interactionally relevant complete linguistic unit, in a given context, that is constructed with syntactic and prosodic resources within their semantic, pragmatic, activity-type-specific, and sequential conversational context. It ends in a TRP unless particular linguistic and interactional resources are used to project and postpone the TRP to the end of a larger multi-unit turn. This suggestion tries to spell out some of the assumptions that the seminal work in CA made in principle, but never formulated explicitly.