Capitalizing on the ChatGPT hype to rebrand the use of corpora in translator education: suggestions for a strategic shift from “corpus tools” to “data literacy” - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2024

Capitalizing on the ChatGPT hype to rebrand the use of corpora in translator education: suggestions for a strategic shift from “corpus tools” to “data literacy”

Rudy Loock

Résumé

This presentation draws upon a decade of experience in teaching the use of electronic corpora for translators, and critically examines their actual integration in the translator’s toolbox. In spite of extensive research showing the relevance and efficiency of using corpora of authentic texts for translation, revision, and post-editing (e.g. Frankenberg-Garcia 2015, Kübler et al. 2022, Loock 2016, Zanettin et al. 2003), it has to be acknowledged that electronic corpora, present in a number of tools used by translators (online dictionaries, translation memories, machine translation, and now large language models) and now frequently introduced in translation programs, are not among the most popular tools used by professional translators (see ELIS results). Our proposal advocates a strategic shift in the way electronic corpora are introduced in translation programs. In particular we suggest: - Focusing on the development of data literacy rather than on the use of corpora as translation tools, already developed in association with machine translation (Kruger & Hackenbuchner 2022) - Making a clear distinction between corpora as translation tools and corpora as research tools - Rebranding corpora as linguistic databases, and in doing so, capitalizing on the current hype around large language models and tools like Chat-GPT - Promoting metacognitive approaches Our reflections are guided by surveys and experiments conducted in 2022-2023, involving current and former students (2013-2023) from a master’s translation program, as well as professional translators who attended a continuous training program on the use of corpora for translation (2019-2023). Participants in the different surveys were questioned about the skills gained after their training on the use of corpora and the kind of corpus tools that they use today, sometimes years after their initial training at university. Results show that, although the skills are often retained, the use of corpora remains marginal, even though it is more important for freelancers (vs. employees in an agency or an organisation) and those with at least one area of specialisation. We also noted a difference between corpora available online and specialized, DIY (Do-It-Yourself) corpora to be used with a concordancer.

Domaines

Linguistique
TALC2024_Loock.pdf (810.17 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04632976 , version 1 (12-07-2024)

Identifiants

  • HAL Id : hal-04632976 , version 1

Citer

Rudy Loock. Capitalizing on the ChatGPT hype to rebrand the use of corpora in translator education: suggestions for a strategic shift from “corpus tools” to “data literacy”. 16th Teaching and Language Corpora (TaLC) Conference, Manchester Metropolitan University, Jul 2024, Manchester (UK), United Kingdom. ⟨hal-04632976⟩
2 Consultations
0 Téléchargements

Partager

Gmail Mastodon Facebook X LinkedIn More