Introduction
In recent years, content and language integrated learning (CLIL) has been proliferating across Japan, given its positive impacts on student motivation, teacher development, learning depth, and language use (Ikeda, Reference Ikeda, Reinders, Hayo and Nakamura2019). However, Japanese English educators are still struggling to best deliver CLIL. In order to document and describe what is actually happening in a Japanese English CLIL classroom, and how multiple language use unfolds, a three-week ethnographic study, including classroom observation, interviews and fieldnotes, was conducted.
Background
The study involves eleven grade 1 and fourteen grade 4 Japanese elementary school students learning English at a private girls’ school in an urban city in northern Japan, in the sixth largest city with a small percentage (0.6 percent; 13,989) of the total prefectural population being non-Japanese (Ministry of International Affairs and Communications, 2015). The data was collected in a math and science class taught by one native English speaker (with no prior CLIL experience) originally from the UK (T1) and two Japanese-English bilingual veteran CLIL teachers, one specializing in math (T2), the other in English (T3).
Research questions
In this study, the following research questions were explored:
1. What is happening in a Japanese English CLIL classroom?
• What role does L1 play in L2 content learning?
• What role does L1 play in L2 vocabulary learning?
2. How does multiple language use unfold in a CLIL classroom with monolingual Japanese students and with different teachers?
In order to address these questions, in addition to CLIL-related studies, we also considered translanguaging (Garcia & Wei, Reference Garcia and Wei2014), sociocultural theory (Vygotsky, Reference Vygotsky1980), and activity theory (Engeström, Reference Engeström, Engeström, Miettinen and Punamäki1999).
Theoretical framework
CLIL has been reported extensively as an effective instructional approach that combines language and content learning, hence allowing for meaningful language use to take place (Coyle, Hood, & Marsh, Reference Coyle, Hood and Marsh2010; Dalton-Puffer, Reference Dalton-Puffer2011; Ball, Kelly, & Clegg, Reference Ball, Kelly and Clegg2016; Ikeda, Reference Ikeda, Reinders, Hayo and Nakamura2019; Lin, Reference Lin2019). Within CLIL classrooms, translanguaging has been observed (Nikula & Moore, Reference Nikula and Moore2016; Lin, Reference Lin2019). Translanguaging (Garcia & Wei, Reference Garcia and Wei2014) addresses the entire linguistic repertoire the language learner has and uses creatively to convey meaning and accomplish goals. Classroom translanguaging refers to translanguaging induced by instruction or orienting to content; universal translanguaging is orienting to interaction (Lewis, Jones, & Baker, Reference Lewis, Jones and Baker2012).
The notion of translanguaging is in line with that of sociocultural theory (SCT), which envisions learning from social interaction (Vygotsky, Reference Vygotsky1980). Used in interactions, translanguaging serves to be a metatalk (i.e., talk about talk) and metacognition (i.e., talk about the task), extending the learner's zone of proximal development (Lantolf & Thorne, Reference Lantolf and Thorne2006; Garcia & Wei, Reference Garcia and Wei2014).
In order to examine how different instructional scaffolding affects students’ language use, we referred to activity theory (Engeström, Reference Engeström, Engeström, Miettinen and Punamäki1999) to see the connections among different social variables that gave rise to a particular language use at particular instances. Specifically, according to activity model the subject that acts upon an object does so within various constraints that are imposed by the community, social rules and division of labor in society. Within the social confines, the individual utilizes various mediating artifacts to act upon the object. The resulting outcome is that of a particular instance of a social interaction (see Figure 1 in the supplementary material for an activity model for this study).
Methodology
The grades 1 and 4 CLIL math classes at a private elementary school were observed and audio- and video-recorded daily for three weeks, for a total of 27 hours. All classes were taught by the same three teachers (i.e., native speaker (NS) classroom teacher (T1), non-native speaker (NNS) math teacher (T2), and NNS English teacher (T3)). The audio-recordings were all transcribed, coded, and analyzed. Findings were explored using activity theory (Engeström, Reference Engeström, Engeström, Miettinen and Punamäki1999) to investigate the connections between the individual (i.e., teachers and learners) and social (i.e., in-class interactions).
Findings
For illustrative purposes, discussion will focus on two excerpts from the poster. It was discovered that the students used their first language (L1) (Japanese) effectively and strategically in order to understand mathematical, linguistic, and scientific concepts introduced within the classes. In Excerpt 1Footnote 1 we see a discussion using L1 by T3 (NNS English teacher) that results in students correctly describing the use of plural ‘s’. Later excerpts reveal their correct application of this form after this exchange.
In turn 50, T3 allows the use of L1 to better discuss plural ‘s’ use in English, which up to this point had not been introduced within the English language classes. Students begin to conceptualize the use, by noting differences between ‘big numbers’ (turn 52) and ‘single things’ (turn 53).
Within Excerpt 2, instances of effective and ineffective scaffolding occur. In turn 16, T2 provides L1 scaffolding that provides some level of understanding, yet students still do not seem to have enough information to work with the concept being discussed.
Within turn 11, S8 tries to address the meaning of ‘petroleum’ within their L1. They reform their answer into a direct translation of the two L1 characters that make up the word ‘sekiyu,’ ‘rock energy,’ into their L2. This episode results in little apparent gain, as in turn 15, S9 checks once more whether ‘sekiyu’ is correct. T2's response in turn 16 appears to provide enough scaffolding for the students to move beyond the translation of the word into L1, and to continue to focus on the concept of ‘petroleum’ as T1 moves into introducing differences in usage between countries. Unfortunately, as the interaction progresses, problems involving the instruction being provided appear within turns 27 and 29, as S7 expresses difficulty understanding what is being instructed. Later discussion reveals that, though the students know the translation of ‘petroleum’ in their L1, they do not know what ‘petroleum/sekiyu’ is, e.g., a flammable liquid, or why it is used.
Within these excerpts, we have seen multiple uses of scaffolding that varied between NS and NNS; importantly, the use of language (L1 Japanese and L2 (second language) English) as a holistic resource was shown to be effective for scaffolding the students’ understanding within their unique zones of proximal development (ZPDs). This lends credence to classroom translanguaging being an effective resource for teaching/understanding within classrooms.
Discussion
It was found that NNS teachers managed to tailor their instructions that were in line with the students’ ZPD (see Figures 4 and 5 in the supplementary material) while NS could not (Figure 3 in the supplementary material). T1 (NS) attempts to scaffold in L2 (English) but fails to address the students’ ZPD in doing so. As a result, students fail to understand both the content and the English. On the other hand, T2 and T3 (NNSs) both manage to provide scaffolding that are attuned to the students’ ZPD, at times in L1 (Japanese) as well as in L2 (English).
Conclusion
Prior to the classroom observation in June 2019, it was hypothesized that English teachers and Japanese primary schoolers frequently engage in classroom translanguaging in CLIL classes. However, it was found that classroom translanguaging happened unpredictably and rarely, with instances of teacher-initiated translanguaging sometimes, surprisingly, resulting in no student translanguaging. This is dissimilar to instances of universal translanguaging which emerged regardless of in-class instruction. Hence, it is argued that classroom translanguaging is not as naturally emergent for the learners as compared to universal translanguaging. Due to this, classroom translanguaging could perhaps be better utilized by teachers as a planned part of classroom discourse, with affordances made for impromptu and fortuitous uses.
Limitations
This study is a small-scale study which only looked at two classrooms in a small, private Japanese girls’ school for just one month. A more longitudinal data collection, together with data collected from different cities in different schools might offer more illuminating findings.
Significance
This study demonstrates how L1 use, which is often discouraged in a Japanese English as a foreign language (EFL) classroom (MEXT, 2002, 2003; Sakamoto, Reference Sakamoto2012), is important and useful in Japanese CLIL classes. While ZPD-sensitive scaffolding in L1 or L2 is found to be effective, different teachers scaffold students differently, and while NSs are often more appreciated and deemed to be more qualified language teachers in general, this study suggests that NS proficiency alone is insufficient for effective EFL teaching. It is suggested that both NSs and NNSs are to be trained not only in terms of techniques in CLIL delivery but as teachers who can build on the knowledge students bring to the classroom.
Supplementary material
To view supplementary material for this article, please visit https://doi.org/10.1017/S026144482000035X
Corey Fegan is a Ph.D. candidate at Sophia University in Tokyo, Japan. For ten years he worked as an English language teacher at a private elementary school in Japan, introducing and teaching a CLIL mathematics program. From there his interest in CLIL and translanguaging developed, and he is now focusing his research on translanguaging, CLIL, and how language integration can aid in the learning of content and language.
Mitsuyo Sakamoto is Professor in the Department of English Studies at Sophia University in Tokyo, Japan. In 2000, she received her Ph.D. from the Ontario Institute for Studies in Education, University of Toronto. She is a critical sociolinguist interested in how language learning is shaped by social interactions and socio-political structures. Specifically, she is interested in how bilinguals enliven their ethnolinguistic vitality and creatively utilize their cultural and linguistic resources. Over the years she has been collecting and documenting accounts of heritage language learners/users in Brazil, Canada, and Japan.
