This paper forms part of an ongoing investigation into the English subtitled version of the South African film Tsotsi (2005), released in the UK with the subtitled and dubbed DVD versions released in France. The film has won various awards, including an Oscar for Best Foreign Film in 2006. The screenplay draws its inspiration from an Athol Fugard novel published in 1980 but has been adapted to present-day South Africa, centring on life in Soweto, a sprawling township on the outskirts of Johannesburg. The characters speak Tsotsitaal, which mixes various local African languages, Afrikaans and English. Code-switching is another feature of the original dialogues.
A previous study examined the translation strategies in operation, referring to similarities and differences between the versions whilst taking account of any additional influences as a result of the interaction between auditory and visual signs. The current study builds on these findings by examining the DVD released in the United States. To what extent has the translation been shaped by the target audience? The subtitled and dubbed versions of Tsotsi will also be compared with another contemporary South African film set in Soweto, entitled Hijack Stories (2000), which appeared at the Official Selection of the Cannes Film Festival in 2001. The characters in Hijack Stories generally speak English, although code-switching also features highly and provides an accurate depiction of the multilingual setting which is characteristic of the South African ‘Rainbow Nation’. The visual setting and thematic concerns of each of the films are similar. In both films the viewer is shown the realities of life in a South African township briefly contrasted with wealthier urban areas. However, there are differences, with a higher frequency of dialogue in Hijack Stories, which also explores questions surrounding sense of belonging and personal identity in post-apartheid South Africa.
Do these similar representations of a South African township differ with regard to the translation strategies in the French versions? To what extent are the resulting translations defined by the signs that make up the audiovisual text? In the dubbed versions, do verbal and visual signs connect? Do the translations bridge the inevitable gap between an image belonging to the source culture and dialogue situated in the target culture? Such questions will be explored, taking into account the multimodal nature of the audiovisual text.

Zoë PETTIT is a senior lecturer in French at the University of Greenwich (London). She holds a PhD in Audiovisual Translation (2000) from Paul Valéry University, Montpellier III (France). Her research interests include interlingual subtitling and dubbing, multimodality, verbal and non-verbal communication. She has published various articles on audiovisual translation in, for example, Y. Gambier (ed.) Special Issue of Meta; JosTrans; Perspectives: Studies in Translatology. Other recent publications include chapters in, for example, A. Serban, J-M. Lavaur (eds.) Le sous-titrage des films: approches pluridisciplinaires and in J. Diaz-Cintas (ed.) New Trends in Audiovisual Translation (Topics in Translation). She has participated in various international conferences on audiovisual translation and is a member of the European Society for Translation Studies and the Chartered Institute of Linguists.