In The interpretation of dreams Freud famously claimed to have finally solved the riddle of dreams. Yet amidst all the heated debates and intense controversies that ensued in the wake of this groundbreaking work, one fundamental question has been entirely overlooked, namely: in what sense, exactly, are dreams analogous to riddles? It will be the burden of this paper to show that a critical investigation of this seemingly simple question reveals a fundamental and hereto unnoticed discrepancy between Freud's rhetoric on (...) his method of dream interpretation and its application in practice. More specifically, whereas Freud argues that the psychoanalytic method can effectively solve the riddles of dreams by uncovering their pre-existing solutions, careful examination reveals that dream interpretations of this kind are the product of a very different solving technique, one that proceeds along a retroactive timeline rather than a linear one. Drawing on Wittgenstein's distinction between two kinds of riddles and the manner in which they are solved, I expound on the nature of retroactive riddle-solving, thus generating a distinctly different picture of psychoanalytic dream interpretation than the one envisioned and advocated by Freud. (shrink)