Abstract

In presenting a theory of representation in the documentary, this dissertation first examines the major issues central to documentary theory, then goes on to propose a typology of strategies of documentary narration. I first argue for a contextual definition, such that the distinction between documentary and the fiction film is not necessarily based on intrinsic textual differences, but on differences in the stance producers take toward texts and the ways texts are indexed. In the documentary, the producers take an assertive stance toward the states of affairs presented, asserting that those states of affairs occur in the actual world. I then go on to examine the use of moving photographs and recorded sounds in the documentary. I argue that moving pictures may look like the original scene and recorded sounds may sound like the original sound. However, their iconic nature is often overshadowed by their insertion into the text's rhetorical system. A documentary, then, is not a copy of reality, but a rhetorical construct which makes asserts and implies propositions about the actual world. I go on to examine the parameters of documentary narration, showing that structure, style, and point of view have as much relevance to the documentary as they do to the fiction film. ;The second half of the dissertation describes several strategies of documentary narration primarily according to two major factors: the epistemological position taken by the film, and its use of structure and style. ;Classical documentaries teach and explain from a position of authority and exhibit a unified structure similar to the classical fiction film. The open documentary is epistemically hesitant, content to observe and explore rather than explain phenomena. In form it is relatively disunified and meandering. While the classical and open documentaries are the central sub-genres, several alternative groupings are important, including the poetic documentary, avant-garde documentary, metadocumentary, and parodic documentary