This paper proposes an alternative view on different types of exercises usually pro-posed by teachers in history classes. Based on intertextual theory, it discusses the rele-vance of radically distinguishing between 'researched' and 'taught' history. It identifies different levels of analogy these exercises share with the historiographical gestures: dealing with information, the logics of conceptual and causal structures, argumentative work. By pointing out that more complex exercises presuppose that simpler ones have been previously solved, it discusses the different textual products: texts considered suf-ficiently analogous to the original in order show the acquisition of knowledge and also texts in which memory interacts with imagination in different ways.
Teaching of history; discursive thickness; logical reconstruction; methodological reconstruction