Negotiation of form by EFL learners: Effect of task modality and L1 use

Abstract

One issue that faces second language (L2) teachers when they use task-based language teaching (TBLT) is how they should integrate focus on form into goal-oriented, meaning-focused tasks. This issue is particularly relevant to Japanese secondary school teachers, who need to prepare students for entrance examinations that heavily emphasize grammar. Researchers have proposed various ways to address this issue, one of which is shifting task modality from speaking to writing (Richards, 2002; Skehan, 1998). Studies have shown that learners engage in negotiation of form (i.e., an interactional sequence in which learners attempt to resolve a linguistic problem in their output) more frequently when they are required to produce written output (Adams, 2006; Niu, 2009). Another way of promoting focus on form during task-based interaction is to have learners use their first language (L1) to negotiate forms. Research has demonstrated that the use of metalanguage enables learners to discuss forms in detail and helps them maintain their attention on the forms (Fortune, 2005; Fortune & Thorp, 2001). Learners of English as a foreign language (EFL) typically develop explicit knowledge of English through the medium of L1 metalanguage. Thus, it is assumed that EFL learners will negotiate forms more frequently and effectively if teachers allow them to speak their L1 during task work. This study investigated the effectiveness of the two manipulations—shifting production modality to writing and having learners use their L1—in facilitating negotiation of form during task work. First-year university students in two EFL classes at a university in Tokyo participated in two data collection sessions. In the first session, both classes completed a picture story jigsaw task and then wrote the story in pairs. In the second session, both classes completed another picture story jigsaw task and then orally narrated the story in pairs. In both sessions, one class was instructed to speak English only during the post-task while the other class was allowed to speak their L1 (Japanese). Students’ interactions were transcribed, and language-related episodes (LREs) were identified in the transcripts. LREs refer to interactional sequences in which the learners question or correct the use of an L2 item in their own or each other’s utterance (Swain & Lapkin, 1998; Williams, 1999). When all LREs were identified, they were classified according to focus, outcome, and L1 use. Then, to investigate the effect of modality, the writing post-task and the speaking post-task were compared in terms of the frequency, focus, and outcome of LREs. To investigate the effect of L1 use, the English-only (EO) class and the English/Japanese (EJ) class were compared in terms of the frequency, focus, and outcome of LREs. In addition, to study the effect of L1 use further, the texts that students composed in pairs on the writing post-task were analyzed, and the two classes were compared in terms of the accuracy of the compositions and types of errors they made. Finally, LREs in which students used Japanese (L1 Use) and LREs in which they used English exclusively (L2 Only) were compared in terms of metalanguage use and length. The comparison between the two post-tasks showed that the writing post-task generated significantly more LREs than the speaking post-task, and this result was consistent for both classes. Regarding the focus and outcome of LREs, significant differences between the two modes were found only with the EJ class. On the writing post-task, EJ pairs focused on lexis, grammar, and discourse almost evenly and resolved 70-80% of LREs successfully. On the speaking post-task, however, they focused predominantly on lexis and resolved less than 50% of LREs successfully. The comparison between the two classes revealed that the EO class generated significantly more LREs than the EJ class on both post-tasks. For the focus and outcome of LREs, significant differences between the two classes were found for the speaking post-task, but not for the writing post-task. On the speaking post-task, EO pairs focused equally on lexis and grammar and resolved about 80% of LREs successfully, while EJ pairs focused mostly on lexis and resolved only 50% of LREs successfully. As for the compositions they wrote, the study found no significant difference between the two classes, either in terms of accuracy or error types. The comparison between LREs in the two L1 use categories revealed that students used metalanguage in only 35% of LREs in the L2 Only category. This made a clear contrast to LREs in the L1 Use category. In this study, all L1 utterances in LREs were regarded as metalanguage use. Thus, all LREs in the L1 Use category contained, by definition, at least one instance of metalanguage use. The comparison also indicated that LREs in the L1 Use category were significantly longer than LREs in the L2 Only category. In the L1 Use category, LREs that contained L1 and L2 metalanguage use were significantly longer than those that only contained L1 metalanguage use. In the L2 Only category, LREs that contained L2 metalanguage use were significantly longer than those that contained no metalanguage use. However, there was no significant difference in length between LREs that only contained L1 metalanguage use and those that only contained L2 metalanguage use. The study demonstrates that shifting modality from speaking to writing in the post-task stage is an effective means to incorporate focus on form into task cycles. Researchers argue that written production is more conducive to learning than oral production because forms are visually salient and remain permanently (Adams, 2006; Swain & Lapkin, 2001; Williams, 2012). The results imply that, on collaborative writing tasks, these features of writing help learners notice problems in their output and initiate negotiation to resolve them. As for L1 use, the study shows that allowing learners to use the L1 can reduce, rather than increase, the opportunity for focus on form. When learners have their L1 available, they might use it for addressing non-linguistic issues that they cannot easily handle in the L2, such as working out story details and identifying contents of pictures. As a result, they might negotiate forms less frequently. The study also indicates that making the L1 available while learners write together does not necessarily help them produce more accurate texts. This means that the L1 has some impact on the frequency, focus, and outcome of negotiation, but this impact might not be so strong as to affect the task product. Thus, learners in the study negotiated forms less frequently when they were allowed to speak the L1, and this might be because the L1 directed their attention to non-linguistic aspects of the task. Another explanation for this finding is that the L1 enabled learners to discuss one form longer, and this made it difficult for them to negotiate many forms in a given time. The latter explanation was supported by the comparison between the two L1 use categories: LREs in the L1 Use category were significantly longer than LREs in the L2 Only category. Closer examination of individual LREs in the two categories revealed that the difference in length came from differences in metalanguage use. By definition, all LREs in the L1 Use category contained at least one instance of L1 metalanguage use. The analysis revealed that L1 metalanguage in these LREs tended to involve Technical Metalanguage, such as grammatical terms, grammatical rules, and word definitions, and using Technical Metalanguage, learners often justified their choices or compared alternative candidates. In contrast, only 35% of LREs in the L2 Only category contained any use of L2 metalanguage, and the L2 metalanguage mostly consisted of simple response tokens such as yes and OK. Fortune (2005) and Fortune and Thorp (2001) emphasized the importance of Technical Metalanguage in negotiation of form. They argued that this type of metalanguage enables learners to articulate their explicit knowledge and discuss forms in detail, and thereby helps them engage in negotiation deeply. This study provides support to their argument and has shown that an important role of the L1 in EFL learners’ negotiation of form is to facilitate the use of Technical Metalanguage. Using L1 Technical Metalanguage, EFL learners can negotiate individual forms at length and maintain their attention on the forms. L1 Technical Metalanguage also helps them verbalize their explicit knowledge and share it with their peers. Through these, they can resolve linguistic problems collaboratively and scaffold each other’s learning.