In this paper we introduce a method for generating interactive documents which exploits the visual features of hypertext to represent discourse structure. We explore the consistent and principled use of graphics and animation to support navigation and comprehension of non-linear text, where textual discourse markers do not always work effectively.