On the process of choosing in translational logogenesis

Abstract

The creation of a text, i.e. logogenesis, involves dynamic processes of choice. This chapter examines the process of choosing in translation, which requires negotiating the demands of both the source and the target language systems. Halliday’s model of probabilistic choice is the starting point for an exploration of how the two distinct sets of choices involved are managed in the production of the translated text. Empirical evidence from a translation experiment involving the language pair English and German offers insight into the variation between unmarked and marked translation choices. The analysis integrates product data in the form of translation corpus data complemented by translation process data from keystroke logging. The keystroke logs provide the opportunity to look for evidence of choices made during the translation process, in the form of measures such as typing duration, pauses, and revisions. It is the combination of product and process data that provides a plausible description of probable choices in translation. The chapter suggests that the inclusion of empirical evidence on the text production process in systemic functional descriptions is needed in order to understand more fully the question of choice. It contributes to an understanding of how questions of language processing and production can be integrated within the system functional linguistics framework.