Exploring the effects of combined subtitling and revoicing tasks on learners’ L2 acquisition of multiword expressions

In the last few decades there has been an increase in the number of studies that investigate the potential of subtitling and revoicing as didactic tools; several empirical studies proved that subtitling (Ghia, 2012; Incalcaterra McLoughlin & Lertola, 2014; Lertola, 2013; Sokoli, 2011; Talaván, 2010; 2011) as well as revoicing activities (Burston, 2005; Chiu, 2012; Danan, 2010) can enhance the learning experience by helping students to develop a wide range of L2 skills. However, the line of research that explores the use of subtitling in the foreign language classroom seems to have developed independently from that exploring revoicing since most studies focus on either one or the other practice, except for, to the best of my knowledge, a recent study where Talaván (2015) offers first insights into the combination of dubbing and reverse subtitling. Furthermore, there appears to be a lack of empirical studies that explore the didactic potential of audiovisual translation tasks as opposed to traditional translation (Lertola and Mariotti, 2017). The present paper reports on a small-scale experimental study on the effects of combined subtitling and revoicing tasks on learners of Italian as a foreign language with a specific focus on the acquisition of multiword chunks, which has been linked to a better L2 fluency and communicative competence. The experiment involved an Experimental Gr oup and Control Gr with only one variable identified as the type of translation task, that is, AVT (subtitling and revoicing) against traditional textual translation integrated with oral comprehension and production tasks; on the other hand, the dependent variable is the acquisition of six collocations selected from the first video used. The video material chosen for this study consists in clips of episodes taken from an Italian TV series which represent a rich source of contextualised L2 input (Fernández-Guerra, 2013; Frumuselu, 2015).

This study was carried out in an Irish university with 13 B2-level students and employed multiple data collection tools; qualitative data were collected through initial and final questionnaires and class observation, while a language pre-test and immediate post-tests supplied quantitative data. During six weeks, participants in the Experimental Group were asked to produce interlingual L2 to L1 subtitles for a video and subsequently to revoice another video; the Control Group watched the same videos, translated their dialogue transcripts and carried out oral comprehension and production tasks. Results obtained from the statistical analysis of the language tests show an overall improvement over time; moreover, although no statistically significant difference was detected between the two groups, the same results indicate that the experimental group outperformed those in the control group, thus pointing towards an added value of the AVT-based treatment to the acquisition and reuse of multiword expressions.