This article addresses the issue that researchers in Translation Studies (TS), who apparently come from different worlds of discourse, seem to not really understand each other. Whereas in the liberal arts the focus is on the discussion and development of ideas and theories, empirical science concentrates on objects and data. This leads to differences in the academic conventions of scholarly communication and the construction of terminology. On the one hand, there is a preference for independent argumentation, not necessarily with exhaustive evidence; on the other hand, a systematic analysis of data using rigorous logical inference is expected. The paper describes the basic characteristics of academic writing in both worlds of discourse and presents a comparative survey of research skills needed in all fields of TS. The necessary skills, such as awareness of one’s background, critical reading, personal standpoint, consistency and logic, and explicitness in the presentation are valid for doing research in general. But they find different expression in scholarly writing according to the academic background.