In "Making Comics: Storytelling Secrets of Comics, Manga and Graphic Novels" (2006), Scott McCloud proposes that the use of specific drawing techniques will enable viewers to reliably deduce different degrees of intensity of the six basic emotions from facial expressions in comics. Furthermore, he suggests that an accomplished comics artist can combine the components of facial expressions conveying the basic emotions to produce complex expressions, many of which are supposedly distinct and recognizable enough to be named. This article presents an empirical investigation and assessment of the validity of these claims, based on the results obtained from three questionnaires. Each of the questionnaires deals with one of the aspects of McCloud’s proposal: face expression intensity, labelling and compositionality. The data show that the tasks at hand were much more difficult than would have been expected on the basis of McCloud’s proposal, with the intensity matching task being the most successful of the three.