Video: Prof Michael Carl's lecture on empirical Translation Process Research
Watch the full video recording of the lecture on "What have we learned from empirical Translation Process Research?", by Prof Michael Carl. The lecture was delivered on the 30th of March 2022 as part of the Convergence lecture series promoted by the Centre for Translation Studies, University of Surrey.
Title of the lecture: What have we learned from empirical Translation Process Research?
Around 40 years ago, Translation Process Research (TPR) started out to investigate "What goes on in the minds of translators" (Krings, 1985). Since then, a large number of techniques have been deployed to illuminate the translator's black box and numerous cognitive models have been suggested to explain the hidden translation processes. The Monitor Model (Schaffer and Carl 2013,2015) stipulates that two concurrent (sets of) translation processes complement each other in the translator's mind. Automatic/horizontal processes are driven by priming mechanisms. They are modular and quick. Controlled/vertical processes are slower. They take into account a large(r) number of cognitive resources, including internal and external search and conscious/reflective thought.
In this talk I will point to similarities between the Monitor Model and the Global Neuronal Workspace Theory (e.g., Dehaene 2014), one of the leading cognitive accounts of consciousness. In line with Relevance Theory (RT), I will argue that principles of relevance are at the core of translational activity. RT suggests that translation is a form of communication which follows the principle of relevance (Gutt 2000). The principle of relevance stipulates that participants in a communicative situation follow a path of least effort while producing maximum cognitive effects.
Supported by observable traces of transitional effort and effects from the translation process data, I will introduce a dynamic notion of translation relevance as a multivariate space (a field) of effort and effect indicators, in which the principle of relevance defines an optimal path. Conceptualizing relevance as a multivariate field of effects and effort underpins the enactivist nature of translation that sees cognition as a direct coupling of translator-environment interaction, rather than a process of manipulating structured symbolic mental representations. It shifts focus on non-conscious (priming) and non-representational (dynamic) processes. I will argue that a notion of field of relevance is compatible with the Global Neuronal Workspace Theory. I will use data from the CRITT TPR-DB to illustrate parameters of such a relevance field. I will illustrate different trajectories through this field and their relation to automatic and reflective translation processes, in support of the Global Neuronal Workspace Theory.
Speaker's short bio: Dr Michael Carl is a Distinguished Professor at Kent State University/USA and Director of the Center for Research and Innovation in Translation and Translation Technology (CRITT). He has studied Computational Linguistics and Communication Sciences in Berlin, Paris and Hong Kong and obtained his PhD degree in Computer Sciences from the Saarland University/Germany. He has worked and published for more than 25 years in the fields of on translation studies, machine translation and natural language processing. His current research interest is related to the investigation of human translation processes and interactive machine translation.