Should Machine Translation have a role in language classrooms or not?

Abstract

Presentation: https://www.pechakucha.com/presentations/sotel-2023-neil-cowie-and-keiko-sakui-machine-translation
 Machine translation (MT) of languages has been around nearly 30 years but the importance of its role in language learning has grown exponentially in recent years. This paper summarizes recent research on teacher and learner attitudes to MT, and suggests ways that MT can be used in language classrooms.
 Studies in the 2010s (Pym, 2013) suggest that teachers were against the use of MT because of its poor quality. However, the level of MT dramatically improved from 2016 when Google Translate adopted a neural-network system. As a result, teachers’ attitudes shifted to more acceptance of MT. Even so, teacher views about MT tend to fall into two camps: those who feel it is a form of cheating (Carré et al., 2022) and those who see it as an appropriate teaching tool. The former take the general approach of “detect, react and prevent”, whilst the latter wish to “integrate and educate” (Jolley & Maimone, 2022).
 Research has shown that students use MT in different ways according to their level. More advanced students tend to check words and phrases rather than translating a whole report. They understand the limits of MT but at the same time they believe it can help learn a language (Godwin-Jones, 2022; Jolley & Maimone, 2022). Research suggests that training in the use of MT can increase chances for such students to reflect on their language learning (Pellet & Myers, 2022) and that they can become aware of and correct MT errors (Zhang & Torres-Hostench, 2022). On the other hand, lower level students use MT differently as they may lack confidence in their language abilities (Organ, 2019). There are studies that claim lower level students can be linguistically overwhelmed in trying to notice and compare their own translations with MT; therefore, they do not correct the output of MT and submit it as their own work (Lee, 2022: Niño, 2020).
 In general, the accuracy of MT has improved so quickly that many teachers who previously dismissed MT as poor can no longer ascertain whether their students have actually used it or not (Jolley & Maimone, 2022). This creates doubt in how to assess student work fairly. Furthermore, as teachers vary in their attitudes towards the use of MT for learning, students can be very confused as to whether they are allowed to use MT in different teachers’ classes; and, if they are allowed, in what ways can they do so appropriately. In order to overcome this uncertainty and confusion, it is suggested that, after Reinders (2022), institutions, students and teachers become partners in exploring MT to find the best way to use it for learning. This will vary according to each educational context, particularly concerning student level, but it is vital to create commonly accepted guidelines, approaches and practices so that MT can be best used for language learning and not just as a tool to complete tasks with little or no educational meaning. 
 References
  
 Carré, A., Kenny, D., Rossi, C., Sánchez-Gijón, P. & Torres-Hostench, O. (2022). Machine translation for
      language learners. In D. Kenny (Ed.), Machine translation for everyone: Empowering users in the age of 
      artificial intelligence (pp. 187–207). Language Science Press. Doi: 10.5281/zenodo.6760024
 Godwin-Jones, R. (2022). Partnering with AI: Intelligent writing assistance and instructed language learning.
      Language Learning & Technology, 26(2), 5–24. https://doi.org/10125/73474
 Jolley, J. & Maimone, L. (2022). Thirty years of machine translation in language teaching and learning: A
      review of the literature. L2 Journal, 14(1). Doi: 10.5070/L214151760
 Lee, S.-M. (2022). Different effects of machine translation on L2 revisions across students’ L2 writing
      proficiency levels. Language Learning & Technology, 26(1), 1–21. https://hdl.handle.net/10125/73490
 Niño, A. (2020). Exploring the use of online machine translation for independent language learning. Research in 
      Learning Technology 28, 2402. https://dx.doi.org/10.25304/rlt.v28.2402
 Organ, A. (2019, July 5). L’éléphant dans la salle / la pièce / le salon? Student use of Google Translate for L2 
      production: Student and staff attitudes, and implications for university policy. [Conference presentation
      abstract]. Translation Technology in Education – Facilitator or Risk? University of Nottingham, UK.
      https://www.nottingham.ac.uk/conference/fac-arts/clas/translation -technology-ineducation%E2%80%93  
      facilitator-or-risk/videos/conference-videos.aspx
 Pellet, S. & Myers, L. (2022). What’s wrong with “What is your name?” > “Quel est votre nom?”: Teaching
      responsible use of MT through discursive competence and metalanguage awareness. L2 Journal, 14(1). Doi:
      10.5070/L214151739
 Pym, A. (2013). Translation skill-sets in a machine-translation age. Translators’ Journal, 58(3), 487–503. Doi:
      0.7202/1025047ar
 Reinders, H. (Host) (2022, September 7). A conversation with Jim Ranalli and Volker Hegelheimer [Audio
      Podcast Episode]. In Voices from LLT. https://www.lltjournal.org/media/voices-from-llt/
 Zhang, H., & Torres-Hostench, O. (2022). Training in machine translation post-editing for foreign language
      students. Language Learning & Technology, 26(1), 1–17. http://hdl.handle.net/10125/73466