Using the Gravitational Pull Hypothesis to explain patterns in interpreting and translation: The case of concatenated nouns in mediated European Parliament discourse

Lefer, Marie-Aude [UCL] De Sutter, Gert [UGent]

In this chapter, we present a corpus study of the French rendition of English concatenated nouns, such as climate change, comparing two modes of interlingual mediation at the European Parliament, namely simultaneous interpreting and written translation. Using parallel corpus data extracted from the European Parliament Translation and Interpreting Corpus, we examine how frequently English concatenated nouns are rendered with semantically equivalent items in the two mediation modes, and which factors stimulate the use of these equivalent (vs non-equivalent) renditions. Alongside the complexity and lexicalization of English concatenated nouns, we consider several frequency-related variables inspired by Halverson’s (2017) cognitive linguistic model of translation, the gravitational pull hypothesis. The model posits three cognitive sources of translation effects: gravitational pull (source salience), connectivity (cross-linguistic link strength) and magnetism (target salience). The results show that there are far fewer semantically equivalent renditions in interpreting than in translation. In addition, the regression analysis provides strong evidence that connectivity and magnetism play a crucial role in the selection of semantically equivalent vs non-equivalent renditions in interpretations and translations, alongside the length of source concatenated nouns, with stronger effects in interpreting. By contrast, source-language variables related to gravitational pull and lexicalization do not seem to influence renditions in French. The study brings to the fore key commonalities between translation and interpreting and shows that the three cognitive sources in Halverson’s gravitational pull model can be successfully disentangled in a multifactorial research design.