his project adopts a coherent framework developed to enable assessment of the performance of the press in upholding democracy. The framework was developed by Pippa Norris for a developed world context, but proved by this project to be equally applicable in the developing world context of South Africa. The main functions of the press in relation to elections for public office are deemed to be facilitating pluralistic competition, public participation and the protection of civil liberties and political rights. The main research method employed in this project is content analysis; applied to press coverage of the election of 2004, but the results are correlated through interviews subjected to narrative analysis. The sample for the study includes a wide range of newspapers, from catch-all publications to publications with more segmented readerships and from provincial to national circulation newspapers. Despite the differences in the profiles of the various newspapers, the cross-title press election coverage in the case of the 2004 South African general election was internally very similar indeed. All publications covered the African National Congress more than the other parties and all gave the ANC much more parity between positive and negative coverage than the other parties. The Democratic Alliance and the Inkatha Freedom Party received a negative slant in the coverage over all. The Independent Democrats experienced the opposite. The publications had different directional bias in their editorial coverage to the one presented in the letters to the editor. The publications gave similar amounts of practical information to voters and only to a very small degree did they explicitly encourage people to get directly involved in the electoral context, for example by casting their ballot. All publications focused strategically on the electoral context and all allowed a few dominating themes to run through their coverage. The press carried a relatively small number of watchdog-category articles. While the benchmarks in the framework employed proved very useful indeed in uncovering characteristics of the election coverage provided by the press, uncovering determining factors behind the coverage was not an aim of this study, even though theoretical perspectives on some of the background to the coverage are discussed. Given that this is, to the best knowledge of the author, the first application of the employed framework to a specific case of media coverage, the results are just that, a first set ofbenchmarks. Only future research can determine if the results obtained here indicate that the press did well, or badly, in upholding their role in democracy when covering the 2004 general election in South Africa.

Thesis (M.A.)-University of KwaZulu-Natal, Durban, 2004.