This thesis addresses the relationship among Chaucer, critics, and theories of audience. Drawing on Michel Foucault's concept of the author-function, I argue that scholars use a parallel 'audience-function' to limit and enable their criticism. I argue that this audience-function is crucial to interpretation, and that different ideas about audience produce different readings of literary texts. To prove this assertion, I analyse in detail the work of six prominent Chaucerians in the twentieth century, outlining both their latent and their explicit assumptions about audiences, and showing how those assumptions affect and enable their criticism. This analysis provides a tool for students of Chaucer, allowing them to discern some of the reasons behind the widely varying interpretations of Chaucer's works; it also provides theoretical insight into the way in which particular ideas about audiences are inherent to certain theoretical stances and approaches. In my study, I argue that several abstract categories of audience definition are fundamental in limiting and conditioning these critics' readings, and best explain the diversity of interpretation in Chaucer criticism. First, critics can be divided according to whether they include both medieval and modern readers in Chaucer's 'audience' (Kittredge, Donaldson, Dinshaw, or whether they limit their definition of audience to Chaucer's medieval audience (Lewis, Robertson, Patterson). Second, critics' ideas of audience can be categorised according to whether the audience is seen as relatively trusting and 'straight' (Kittredge, Lewis), or suspicious and ironic (Donaldson, Robertson), or somewhere in the middle (Dinshaw, Patterson). Third, images of audience can be divided according to assumptions that the audience is homogeneous in composition (Kittredge, Lewis, Donaldson, Robertson) or heterogeneous and multiple (Dinshaw, Patterson). With each critic, I explore the way in which these criteria for audience definition condition, circumscribe or prompt particular interpretations of Chaucer's works. The six scholars studied in this thesis are George Lyman Kittredge, C.S. Lewis, E. Talbot Donaldson, D.W. Robertson, Carolyn Dinshaw and Lee Patterson. Each established or represented a particular approach or trend in Chaucer studies, and each presented arguments that the following generation of scholars had to 'answer' before proposing alternative interpretations.