La théorie de l'Esprit chez des traducteurs

The ability to explain and predict other persons’ behaviour by attributing to them independent mental states, such as beliefs and desires, is known as having a ‘Theory of Mind’(ToM). This ability is mandatory in tasks where other peoples’ thoughts have to be anticipated. Interpretations such as translations of texts from one language into another are such a task: Translation can be considered as an inter-lingual transfer during whom a source text is interpreted, a target text is produced and an equivalence relation is established between the two . Such a process necessitates linguistic, technical, cultural and mental and communicative competences and it does not only develop strategies of self observation but also such relating to the observation of others . This concept of awareness of the others’ thinking is found as well in several models of translation, as in the form of the so-called “association competence” , and as an important parameter in the “relatively uncontrolled processing centre” . It is therefore tempting to postulate that translators necessitate developing a special form of Theory of Mind processes, i.e. processes which allow to anticipate the expectations of the target audience concerning the translated text and to fully understand the intentions of the author of the source text. Such ability is not thought to be present in naïve bilinguals without translation training , despite the fact that bilingual children may, in certain situations show more efficient ToM than monolinguals .

Since there are currently no studies on ToM in translators, our aim is to propose paradigms which could allow for answering the question whether the interpretation work in multicultural contexts leads to the development of special ToM processes. Moreover, there are some data which suggest that ToM is not a general meta-representation, but is composed of lower-level domain-specific mechanisms for processing social information. The hypothesis the authors want to test is about the presence of domain specific ToM processes shown by translators during interpretation, (cf. the basic paradigm of translation studies Homo Interpres, which understands man as interpretative being ) and could also be used in other similar mental activities not requiring translation in its pure sense. According to this hypothesis, i) such processes would not (or less) be present in bilinguals without translation training ii) the translator would use these abilities in similar monolingual activities necessitating comparable forms of interpretation.

To address the first question, this project will combine a behavioural and neurophysiological analysis of inter-lingual and intra-lingual translation paradigms. First, one group of translators (20 subjects) and a group of naïve non translator bilinguals (20 subjects) matched for linguistic combination will undergo inter-linguistic and intra-linguistic translation tests and a domain-specific evaluation of ToM abilities related to text interpretation. Second, the same group of translators and naïve bilinguals will perform inter-lingual and intra-lingual translation tasks relying on a validated protocol in translation research (Think Aloud Protocol, TAP) in an fMRI block design paradigm. Our prediction is that:

i) Translators will have higher translation scores than naïve bilinguals. This difference will be due to higher empathy scores (supposed to reflect ToM abilities) in their translation. On the opposite, translators and bilinguals without translation training will not differ in classic “general” ToM scores used to evaluate social cognition. The evaluation of translation quality will be possible using validated methods developed in our groups.

ii) Empathy sub scores will be higher in translators than in naïve matched bilinguals in both inter-lingual and intra-lingual translation evaluation.

iii) In the fMRI experiment, translators will show larger activation of ToM specific areas than non translators. During inter-lingual translation, but neither in intralingua translation nor in repetition, translators will activate more areas implicated in ToM (anterior paracingulum, amygdala, temporo-parietal junction, orbital frontal cortex) to a larger extend than non translators.

Keywords: Bilinguals, Translation, Theory of Mind, fMRI