Studies in Second Language Learning and Teaching Translanguaging in a Chinese university CLIL classroom: Teacher strategies and student attitudes

Pedagogical translanguaging has been extensively researched over the past dec-ade. Yet, little is known about the attitudes of students towards this practice. Students constitute an integral part of classroom interactions and their learning process is significantly affected by teachers’ classroom discourse. This action research (AR) study, situated in a Chinese university Content and Language Integrated Learning (CLIL) reading classroom and aided by lesson recordings and two sets of questionnaires, explores the translanguaging strategies employed by the teacher as well as the students’ attitudes to such strategies. Through incorporating feedback collected from students regarding the teacher’s modifications of language use, the study has demonstrated how the teacher mobilizes her full linguistic resources, in the form of translanguaging, to achieve pedagogical outcomes, which eventually leads to the establishment of a mutually beneficial classroom ecology. The study also indicates that advanced EFL learners, highly motivated to improve language proficiency and acquire subject content unanimously reject the traditional monolingual approach to teaching. The findings call for further research into the impact of pedagogical translanguaging on students’ learning process in multilingual classrooms .


Introduction
The use of first language in second or foreign language learning environments has been extensively researched in the past three decades, challenging the tenets of the monolingual teaching approach (Antón & DiCamilla, 1998;Brice, 2000;Ferguson, 2003;Lin, 2013).Terms such as code-switching, code-mixing, and code-meshing were initially used in naturalistic bilingual settings to describe the "systematic alternating use of two languages or language varieties within a single conversation or utterance" (Liebscher & Dailey-O'Cain, 2005, p. 235).Extensive research into the functions and patterns of such language alternation has led to the scope of empirical studies widening to include pedagogical domains (Cook, 2001;Lee & Lo, 2017;Macaro & Tian, 2015;Moodley, 2007;Qian et al., 2009).
Meanwhile, whilst our understanding of bilingual and multilingual competences and practices is continuously deepening, the concept of translanguaging has captured the idea of interlocutors' flexible maneuvering between two or more language entities.The Common European Framework of Reference for Languages (Council of Europe, 2018) has specifically emphasized pluricultural competence and translanguaging, with the purposeful juxtaposition of the languages of input and output.Unlike codeswitching, which assigns language choices to a particular code, translanguaging practices draw on multiple codes.This involves "the speakers' construction and use of original and complex interrelated discursive practices that cannot be easily assigned to one or another traditional definition of a language, but that make up the speakers' complete language repertoire" (García & Li, 2014, p. 22).Furthermore, instead of maintaining a dichotomous view that looks at different languages involved as separate entities, translanguaging, being intrinsically interactive, performative and creative (Canagarajah, 2011), approaches bilingualism/multilingualism from a holistic perspective.It softens the boundaries between languages, thus offering a new paradigm for second/additional language teaching and researching (Cenoz, 2017;Li, 2018).
In spite of the expanding interest in pedagogical translanguaging in foreign language learning contexts, little attention has been paid to students' attitudes to teachers' translanguaging practices.Our argument in this paper is that collecting student feedback represents an additional informational resource.This provides teachers with a straightforward opportunity to increase their awareness of how their use of classroom discourse can be adapted in order to maximize pedagogical impact.Therefore, this mixed-methods action research (henceforth AR) study sets out to investigate the teacher's translanguaging strategies as well as students' attitudes.Adapting teaching in response to students' views can help teachers to be better informed regarding engaging translanguaging strategies for the purpose of effective and efficient learning, fostering an up-close and "ecological" (van Lier, 2000, p. 251) understandings of classroom interaction.

Literature review
Originating in the context of Welsh bilingual education in the 1980s (Williams, 1996), translanguaging began to gain global popularity as a term against a background of changing views on bilingual competence in the late 20th century (Lewis et al., 2012).Defined as "the process of making meaning, shaping experience, gaining understanding and knowledge through the use of two languages" (Baker, 2011, p. 288), it has been recognized as a significant feature of bilingual/multilingual communities and educational domains.The flexible and creative employment of multiple linguistic resources which transcends the boundaries of named languages has been investigated in naturalistic interactions (Creese & Blackledge, 2019;Li & Zhu, 2020), including Internet spaces (Li & Zhu, 2019), as well as pedagogical contexts, both of which involve the communicative challenges and opportunities to which translanguaging responds.Particular attention has been paid to pedagogical translanguaging, in the hope of exploring the possibilities of increasing understanding and improving learning effectiveness by drawing on both teachers' and students' whole linguistic repertoires (Cohen, 2015;Creese & Blackledge, 2015;King & Ridley, 2018;Leonet et al., 2017;Lin & Lo, 2017;Mazak & Herbas-Donoso, 2015;Wang, 2019a;2019b, to name a few).Recently the concept of translanguaging practice has been further broadened by transcending interlocutors' linguistic competence to take into account their diverse semiotic repertoires, including body language and visual aids.This adds a spatial dimension to the understanding of the connotation of translanguaging (Blackledge & Creese, 2017;Canagarajah, 2018aCanagarajah, , 2018b;;Pennycook, 2017).
Translanguaging research in the educational domain has been widely conducted with varied groups, from young bilingual children (García, 2011;Kirsch, 2020;Seals & Olsen-Reeder, 2020), students in secondary education (Lin & He, 2017;Lin & Wu, 2015;Nikula & Moore, 2019), university students (Mazak & Herbas-Donoso, 2015;Moore, 2014) to adult learners (Li & Ho, 2018).A recent trend has witnessed the concept of translanguaging being implemented in professional development courses for pre-service and in-service teachers.The aim is to raise their awareness of the multilingual and crosslinguistic approach to teaching language and subject content (Cenoz & Santos, 2020;Gorter & Arocena, 2020;Makalela, 2015;Woll, 2020).In such studies, a translanguaging perspective has been found to have a positive impact on exploring one's multilingual repertoires, facilitating interaction and enhancing learning.
Translanguaging became conceptualized as a pedagogical practice at the beginning of the 21st century (Baker, 2011).Given the consideration that the target language is often prized in traditional L2 classrooms, challenges exist in terms of determining the extent to which translanguaging can be seen as a variable pedagogical resource and whether translanguaging strategies can be developed with the purpose of multilingual language acquisition (Canagarajah & Gao, 2019).Issues such as errors or mistake correction, rhetorical considerations and the ways students negotiate power remain to be considered in assessing the effectiveness of translanguaging as a pedagogical strategy (Canagarajah, 2011).Nonetheless, efforts have recently been made to explore instances where translanguaging is employed as a classroom teaching strategy.Researchers have found that in a number of bilingual/multilingual CLIL or second/additional language education contexts, teachers actively deploy translanguaging in a variety of classroom scenarios to fulfill communicative and pedagogical purposes.Examples include explaining subject content (Mazak & Herbas-Donoso, 2015;Wang, 2019b), eliciting students' L2 output (Makalela, 2015), elaborating on contrastive linguistic knowledge (Makalela, 2015), facilitating interpersonal communication and creating a welcoming atmosphere (Yuan & Yang, 2020), and managing classroom discipline (Probyn, 2015;Wang, 2019b).
Previous research into classroom translanguaging practice has revealed a dynamic and rich picture of interlocutors' maneuvering between multiple languages, demonstrating translanguaging as a resource to make meaning and facilitate learning (Li & Ho, 2018;Nikula & Moore, 2019;Poza, 2018).However, its contribution to pedagogical refinement is limited by the paucity of knowledge about students' attitudes to teachers' translanguaging practices.As most classrooms are dominated by teacher talk, the teacher might well be the most visible translanguager (Walsh & Mann, 2015).On the other hand, it is of equal importance to pay attention to the students who are most directly affected by translanguaging activity.It is important to know their views on the teachers' translanguaging practices, so as to enable teachers to modify classroom discourse in order to enhance learning and teaching.In the existing literature, Wang (2019b) and Galante (2020) attempted to gauge students' attitudes to classroom translanguaging.Wang (2019b) explored students' views on the desired and actual multiple language use in class by the teachers as well as by themselves, whereas Galante (2020) investigated students' perceptions of challenges in pedagogical translanguaging.Nevertheless, neither of these studies employed action research and hence neither provided space for teachers to modify their multilingual practices based on students' feedback.There is still a paucity of empirical evidence which could further our understanding of students' attitudes and how it could help teachers make amendments and improvements.
In order for translanguaging to be recognized and perhaps developed as part of appropriate pedagogy (Holliday, 1994), it would be instructive for teachers, who are in most cases the main translanguagers, to have access to students' feedback on their translanguaging practices in order to modify their language use accordingly, and thus eventually achieve a situated classroom ecology where students can benefit fully from their teachers' language input.In order to achieve this, an AR study, which offers the teacher/researcher the opportunity to gain students' feedback, incorporate it into modifications and observe their attitudes to such modifications, would serve as a useful addition to our understanding of translanguaging practices.

Research questions
As discussed previously, existing research into pedagogical translanguaging has largely focused on teachers' translanguaging practices but has rarely addressed students' perceptions of these practices, and how, in their view, translanguaging practices in classrooms can aid the learning process.It is of benefit to understand in what way translanguaging can be conducted in order to create a mutually beneficial learning ecology for both teachers and students.Therefore, this study aims to answer the following three research questions: 1. What translanguaging strategies does the teacher/researcher employ? 2. What are the students' attitudes to such translanguaging practices in class? 3.In what way can the teacher/researcher learn from students' attitudes and adapt her teaching accordingly?

Context
The research took place in a CLIL classroom where a theme-based reading course was taught to help first-year English majors (CEFR B2 level) improve English language proficiency and gain a comprehensive spectrum of humanities-related knowledge in the fields of literature, culture, history, arts etc. in English-speaking countries.The teacher/researcher in this case was a young female practitioner in her mid 30s with six years of teaching experience and higher education degrees from an English-speaking country.The class consisted of 25 students aged 18 or 19 years old who had just graduated from high schools.They had been learning English as a foreign language since the age of six and thus possessed a good command of English speaking, listening, reading and writing skills.Though both the teacher and students were capable of communicating exclusively in English, in reality, constant switches between the two languages were still evident and, on occasion, essential.This was for two reasons.Firstly, some of the course content posed linguistic challenges to students.Secondly, cross-cultural awareness is a vital learning outcome prescribed by the National Guidance on Curriculum for English Majors.This led to discussions on literature and culture from both Chinese and Western perspectives, prompting a shuttle between English and Mandarin in order to elaborate such subtle and nuanced relations.

Data collection methods
This was a mixed-methods AR study where qualitative analysis of classroom discourse provided a detailed picture of a teacher/researcher's translanguaging practices, thus opening possibilities for particular patterns or categories to emerge (Burns, 2005).At the same time, quantitative data from questionnaires provided a general yet holistic picture from a sample of 25 participants.In the data that follows, the teacher featured was the first author (therefore also referred to as teacher/researcher) and the study offered an evaluation of the value of this kind of practitioner research, as well as providing the basis for an evaluation of the pedagogic nature of translanguaging practices.
Action research is interventionist in nature and the "teacher/researcher" is at the heart of the research, focusing simultaneously "on action and research" (Burns, 2005, p. 58).Thus, by definition, action research is based on some kind of intervention.This creates the possibility of change in the participants and the educational setting, and consequently generates theories and insights anchored in that specific setting.The strength of the methodology is that there is a close connection between theory and practice, but the limitation is that results are not generalizable because the dataset is focused on one context.However, because action research pays close attention to the interaction, learning and impact and investigates "a small-scale intervention," there can be "a close examination of the effects of such an intervention" (Cohen et al., 2017, p. 226).Such insights can at least be transferable to other contexts (Richards, 2003).
An AR process was undertaken in the current study in order to gain a better understanding of the classroom dynamics and improve the effectiveness of the teacher/researcher's teaching practices (Dörnyei, 2007).Through studying students' attitudes towards the teacher's translanguaging activity and then modifying the pedagogical use of translanguaging accordingly, it conforms to the two central characteristics of AR: enhancement of practice and iterative introduction of change (Burns, 2005).The process also allows the practitioner to reflect on actions at the micro-level (van Lier, 2000).There were two AR cycles in this study: AR1: analysis of translanguaging practices and collection of students' attitudes.AR2: reflections on and modifications of translanguaging practices, as well as further collection of students' attitudes for contrastive analysis.
The analysis drew on lesson recordings, questionnaires, and teacher reflective notes.Lesson recordings provided transcripts for reflection and helped to gain insights into the nature of the classroom community so that informed decisions could be made before any changes were put into practice.Questionnaires, on the other hand, are a standard approach to understanding the learners' mind and obtaining specific information from a group of participants (Richards et al., 2012).

Data collection procedures and analysis
The data collection process for this AR study was conducted over the course of two terms in one academic year.Lesson recordings, questionnaires and teacher reflective notes were utilized to provide details and insights into the teacher's translanguaging practices and gauge the students' attitudes.A total of 60 sessions (30 sessions per term), each lasting 90 minutes, were audio-recorded.Two sets of questionnaires were completed by the participants (N = 25) at the end of the first and second terms respectively, with the intention of making a comparison of their feedback before and after the teacher/researcher's modification of classroom language use.To be more specific, the first set of questionnaires aimed to gain a preliminary understanding of students' views on the teacher's language use, whereas the focus shifted to students' attitudes towards the teacher's modification of language use in the second set of questionnaires.A synthesis of key factors in these two sets of questionnaires can be found in Table 1.

Attitudes towards the changes
Impact of the teacher's language use on students' language use and learning process Impact of changes on students' learning process All instances containing translanguaging practices involving both the teacher and students were transcribed, closely studied, and subjected to microanalysis for patterns of purposeful translanguaging strategies.Categories of translanguaging strategies which emerged from the current study were further examined in reference to those in previous studies (Li, 2014(Li, , 2016;;Makalela, 2015;Moore, 2014;Poza, 2018;Wang, 2019a) as a form of comparative analysis.Proportions of English/Mandarin spoken by the teacher in class were calculated based on word count and timed analysis methods (Macaro, 2001).This served as an important reference point when evaluating students' views on the teacher's bilingual practice.Meanwhile, items of multiple choice in the questionnaires were analyzed quantitatively, whereas open-ended questions in the questionnaires were subject to content analysis.

Findings
Broadly speaking, results from the current study were presented from two perspectives: the teacher's translanguaging strategies and students' attitudes towards them.Firstly, categories of teachers' translanguaging strategies were illustrated with examples from lesson transcripts.Secondly, longitudinal findings relevant to students' attitudes in the two AR cycles were reported from four perspectives: their general views about the teacher's translanguaging practices, attitudes to translanguaging practices in different pedagogical moments, the impact of the teacher's translanguaging practices on students, and lastly, how the teacher/researcher modified her translanguaging practices based on students' attitudes and how such attitudes changed after the modifications.It is worth noting that the first three above-mentioned perspectives entailed results from AR 1 and the fourth concerned AR 2.

The teacher's translanguaging strategies
The analysis of lesson transcriptions revealed that the teacher's pedagogical translanguaging practices could be categorized into three types: explanatory strategies, attention-raising strategies, and rapport-building strategies.The following sections provided data-led descriptions on these three areas, with a focus on how translanguaging was used to achieve interactional means in those pedagogical moments.

Explanatory strategies
This type of strategy occurred when the teacher employed a combination of English and Mandarin to explain textbook-related content.This focus might be on specific features of the English language, such as grammar and vocabulary (in textbook and after-class exercises), or might be related to the understanding of cultural differences between China and English-speaking countries.The following two examples of extracts aim to illustrate the purposeful use of translanguaging in these particular circumstances.

S:
What is the difference between "despair" and "desperation"?T: "Despair" is more about disappoint and frustration after losing hope, 灰心丧⽓ (losing hope and motivation), feeling hopeless after being disappointed.

S:
And "desperation" is having nowhere to go? T: Well, sort of . . ."desperation" means making reckless and perhaps irrational efforts knowing that you might fail, 是一种孤注一掷 (kind of putting all the money in my bet and giving it a go).
In the above teacher-student interaction, the teacher actively deployed translanguaging as a pedagogical strategy, by incorporating two commonly used four-character Chinese idioms (underlined) which were known for expressing rich meaning in the most concise and economic manner.In this particular moment, translanguaging offered a more effective explanation with the minimal possibility of misinterpretation.

T:
We Chinese are passionate in pursuing loyalty as the value of our lives.
Loyal to whom?To the emperor, the sovereign, the highest power.We talk about 忠孝价值，忘我的精神境界 (value of loyalty, a mental state where one forgets his/herself completely).Death has to be a form of self-sacrifice, like in 杀⾝成仁舍⽣取义 (to die a martyr, to sacrifice one's life for a noble cause).We Chinese are very much afraid of death because we fear that we can't achieve 立德立功立言 (firstly to set a moral example, then make contributions to society, lastly to document one's experiences and self-reflections for descendants to read), such so-called great cause.
In Example 2, the teacher aimed to explain why Chinese people regarded loyalty as their life value and assumed that students had acquired the content in their previous learning of Chinese history and culture.Hence, she switched between languages to facilitate understanding so as to achieve the pedagogical purpose in that moment, namely explaining cultural differences and developing students' intercultural awareness.

Attention-raising strategies
Observation in previous classrooms and casual conversations with students had made the teacher/researcher aware that students would gradually lose concentration and interest if 100% of the teacher's talk in class was conducted in English.Thus, translanguaging was employed when student attention needed to be raised and focused on important learning points as well as for instructional and managerial purposes.This kind of language was what Willis (2013) called the "outer language" of the classroom.In the following two examples, the teacher proactively deployed translanguaging to gain students' attention while maintaining the flow of the classroom discourse.Example 3 illustrates an attention-raising translanguaging strategy in the context of calling students' attention to key learning points in the textbook.In the monologue paragraph provided above, the teacher did not resort to their shared L1 because she felt under pressure.Instead, she strategically employed translanguaging as a linguistic resource to ensure that students were engaged in the learning of the content when there was no teacher-student interaction present.By doing this, the aim of keeping the class on track and directing students to particular aspects of the text in that particular pedagogical moment was achieved.

T:
In this mid-term examination, as usual, one of the ten sections on the test paper is sentence paraphrasing.请⼤家⼀定注意 (Please pay attention), when you are completing this section, that you should only paraphrase 划线部分 (the underlined part).千万不要 (Please never ever) paraphrase the whole sentence.
Example 4 depicted a scenario where the typical "outer language" in terms of classroom management (Willis, 2013) was used.Having failed to help students avoid careless mistakes when using full L2, the teacher prioritized the pedagogical outcome and switched to translanguaging to raise students' attention.

Rapport-building strategies
This type of translanguaging strategy mostly occurred in teacher-student interactions and took places in two typical ways, as illustrated in the following examples.The first (see Example 5) occurred when one or more students initiated the use of L1, generally because of an inability to understand or express certain words or concepts.The teacher, aiming to elicit more output from the students without intimidating or discouraging them, would employ translanguaging that incorporated the student's use of L1, and maintain a natural flow of interaction.The second circumstance was normally characterized by the teacher participating in students' group discussions.When challenged by ideas too complicated to be expressed in English, students tended to rely on the use of Mandarin, and since all students shared the same first language, such use would not obstruct communication or raise complaints.However, given the course objectives being the acquisition of both language and content, it was expected that students should make sufficient use of the opportunities in classroom discussions to improve their capability of expressing complex ideas in English.Consequently, the teacher would join the group discussions upon noticing students speaking 100% in Mandarin and through initially using translanguaging, would gradually guide students towards a fuller use of English without disrupting the smooth course of communication (see Example 6).

T:
What is your understanding of the quotation: "acceptance without proof is the fundamental characteristics of Western religion, rejection without proof is the fundamental characteristics of Western science"?S: . . .I am not sure . . .

T:
What about the phrase at the beginning: "acceptance without proof"?S: 没有证据就接受? (literal Chinese translation of "acceptance without proof") T: Yes, exactly.Can you think of an example of 没有证据就接受 in religion?
In the situation illustrated by the above example of interaction, the student encountered some difficulty in offering his own explanation of the text (line 4), which was why the teacher opted to break it down for a step-by-step comprehension.When the student was unable to explain the phrases in English, he resorted to Mandarin for assistance (line 6).In this pedagogical moment, the teacher, instead of criticizing or correcting him, employed translanguaging which acknowledged the student's L1 use, but continued to elicit a detailed answer.In this way, the flow of interaction was maintained, more effective communication was accomplished, and the student was not discouraged.
The same strategy could be seen in Example 6 where the tactic seemed slightly different, given that the conversation occurred in a group discussion with the teacher participating later as the fourth speaker.

T:
Exactly.So when one is alone in that situation, one can be as close to nature as one wishes to be.It can be very enlightening to you, you know, stimulating your thoughts . . .

S3:
Yeah, and a lot of ancient Chinese poets, they all lived in places close to mountains and lakes.
When students shifted from English to Mandarin for the ease of communication, the teacher joined in with the pedagogical aim of navigating the conversation back to full use of English and meanwhile acting as a supportive and encouraging team member in the discussion.Therefore, she chose translanguaging, rather than the monolingual use of English which might unnerve the students in any way, to negotiate meaning and eventually achieve her objective.

Students' attitudes
The nature of the AR study required that the analysis of students' attitudes be presented in two parts: attitudes before (the first questionnaire) and after the teacher's modification of her translanguaging practices (the second questionnaire).Students' attitudes in the first part were reported from three perspectives focusing on their feedback on the teacher's bilingual use in class.

General views about the teacher's translanguaging practices
Analysis indicated that before the teacher's modification of her classroom discourse, 80% of all 25 participants expressed satisfaction about the current balance in relation to the teacher's translanguaging practices, whereas 20% believed that the teacher's reference to Mandarin exceeded their expectations.With respect to the use of English in the teacher's classroom talk, 28% of the participants wished that it could be increased.In addition, 52% (N = 13) and 41% (N = 9) of the students expressed that their ideal proportion of the teacher's use of English was 70-80% and 80-90% of classroom talk respectively.At this point, the average proportion of English employed by the teacher throughout the first term, at the end of which the first set of questionnaires was distributed, was 85.6% as recorded using word count and timed analysis methods.

Translanguaging practices in different pedagogical moments
The analysis showed that, despite the disparity of English proficiencies among the participants, the majority of respondents were in favor of the teacher's translanguaging strategies in class.With regard to translanguaging in specific pedagogical moments, the majority believe that translanguaging should be avoided when the teacher explained the literal meanings of the text (N = 21), discussed literature or cultural contents directly or indirectly related to the text (N = 17), and conducted teacher-student interactions (N = 21).On the other hand, the results showed a tendency to favor the use of translanguaging in pedagogical moments where the teacher was assigning homework and briefing teaching arrangements (N = 17), as well as explaining answers to after-class exercises (N = 17).The suggestion that the teacher should use more Mandarin in the above two situations was also well evidenced in the results of the open-ended question on the questionnaire.As for other pedagogical moments, including explaining grammar and vocabulary, and telling anecdotes, there was no significant preference as to whether or not translanguaging practices should be adopted.

Impact of the teacher's translanguaging practices on students
The impact of the teacher's translanguaging practices on students could be looked at from two points of view: the impact on students' language choice and on their learning process as a whole.72% of the respondents (N = 18) stated that the teacher's translanguaging practices would not have a negative impact on their spoken language choice in class, whereas the remaining 28% (N = 7) believed that the teacher's use of Mandarin could affect them negatively in the sense that they no longer felt obliged to speak English in class the whole time.Furthermore, 96% (N = 24) of the respondents considered that a small amount of translanguaging was more conducive to the learning of the English language and content compared to an exclusive use of English.Meanwhile, 24% (N = 6) of the participants also believed that over-reliance on the use of translanguaging could potentially interfere with the students' learning.

Modifications of the teacher's translanguaging practices
Analysis of the first set of questionnaires prompted the teacher to make a number of modifications to her translanguaging practices, by incorporating views and expectations expressed by the participants.The process of classroom talk modifications lasted throughout the second term of the academic year, by the end of which the second set of questionnaires had been distributed and completed by the same group of participants.The modifications involved clear quantitative shifts in the use of a number of strategies.The Appendix provides examples to illustrate the nature of the changes involved.Below is a list of the three features of the modifications: • The teacher reduced the employment of translanguaging strategies in a number of pedagogical moments, including explaining the literal meanings of the text, discussing literature or cultural content directly or indirectly related to the text, and conducting teacher-student interactions.• The teacher increased the employment of translanguaging strategies in other pedagogical moments, including assigning homework and briefing teaching arrangements, as well as explaining answers to after-class exercises.• The teacher maintained the current balance of translanguaging practices when explaining grammar and vocabulary, and telling anecdotes.The researchers calculated the number of instances where each translanguaging strategy was used and the total number of translanguaging instances in all observed sessions.Table 2 below presents the percentages of each translanguaging strategy employed in these two AR cycles respectively.Although there appear to be slight changes regarding the frequency of using explanatory and attentionraising strategies, these were not significant.The data collected regarding students' attitudes based on the second set of questionnaires focused on two areas: their thoughts on the teacher's modifications of translanguaging practices and their current beliefs on the teacher's classroom translanguaging practices in general, as compared to those before the changes.Of all the 25 respondents, 88% of them (N = 22) had noticed certain changes in the teacher's language use in class and all 22 participants believed that the changes were positive with regard to their learning effectiveness.When asked to be specific about the changes they had noticed, their answers included the modifications the teacher planned and deliberately implemented but were not exclusively restricted to those measures, such as: • Fewer translanguaging practices in teacher-student interactions and text explanation.
• The teacher's use of full English to comment on students' presentation performance and more translanguaging practices in recounting anecdotes.Students' feedback on the teacher's translanguaging practices was further explored in the second set of questionnaires with the specific aim of comparing the results before and after the modifications were implemented.The percentage of students who were satisfied with the teacher's translanguaging practices increased from 80% to 92%.It is worth mentioning that the teacher's actual use of English by word count and timed analysis in the second term was 87.1%, in contrast to 85.6% in the first term.In this case, the teacher's translanguaging practices count for less than 15% of the total classroom talk in both the first and second terms.It can thus be argued that such calculations have clearly demonstrated the potential of translanguaging as a valuable linguistic resource -it can be employed in a variety of ways, helps to achieve a wide range of pedagogical purposes, and pays dividends when set against a small proportion of classroom time dedicated to it.

Discussion
One of the most noteworthy findings revealed by the results was as follows.In a classroom where both language and content are the reflexive learning objectives, obtaining insights into students' attitudes to the teacher's translanguaging practices and incorporating those attitudes in a process of modifications is highly beneficial.Students are more satisfied and supported with the role translanguaging plays in enhancing learning effectiveness and the classroom has become a more mutually beneficial learning ecology.This finding underlines the value and potential of translanguaging as a linguistic and pedagogical resource.In the current research, instead of fixating on which particular language should be prioritized, the priority is placed on the pedagogical moments where translanguaging is actively deployed to facilitate communication, enhance understanding and maintain classroom rapport.This AR study has demonstrated how students' feedback on the teacher's translanguaging practices and their views about classroom talk in general can be collected, analyzed and further integrated into the changes that the teacher effects.
Meanwhile, translanguaging theory suggests that boundaries between different languages are blurred when facing the challenges of the Post-Multilingualism era.The focus is now placed upon the context in which languages are used interchangeably and are seamlessly woven in the meaning-making process (Canagarajah, 2011;Cenoz, 2017;Li, 2018).The effectiveness of communication becomes the key focus.However, this article takes the view that the artful weaving of two languages is not only a valid strategy to maximize students' learning but also a useful focus for reflection (Mann & Walsh, 2017).These pedagogical translanguaging strategies encompass not only the specific blending of one's multilingual resources on a meta-linguistic level, but also the pedagogical moments (in relation to classroom tasks) in which one decides to employ translanguaging practices, based on a holistic consideration of a range of factors such as learning outcomes and rapport building.The following two sections discuss the findings from the current study based on two main aspects of the research questions: the teacher's translanguaging strategies and students' attitudes.

Translanguaging strategies
In line with the findings of previous studies conducted in CLIL contexts, the present study also provided evidence of translanguaging strategies employed for the purpose of simultaneous acquisition of disciplinary content and linguistic knowledge (Lin & Lo, 2017;Moore, 2014;Poza, 2018).Three major translanguaging strategies were evident in this study: explanatory strategies, attention-raising strategies, and rapport-building strategies.Of these, the explanatory strategies, which are usually used to elaborate on grammar, vocabulary and subjectrelated contents, have often been identified in previous studies (Li, 2014(Li, , 2016;;Mazak & Herbas-Donoso, 2015;Makalela, 2015;Wang, 2019a).Attention-raising strategies for the purpose of keeping students alert have been mentioned rarely in previous research, but when used in giving instructions, they serve similar purposes to managerial translanguaging practices (Probyn, 2015;Wang, 2019a).In addition, rapport-building strategies, though categorized as interpersonal strategies in some contexts with the aim of maintaining a good flow of classroom interactions (Nikula & Moore, 2019), convey the intent of constructing a supportive and beneficial learning ecology within a group and as a whole in the current research.This was echoed in Yuan and Yang (2020) where students' L1 was used as a facilitative semiotic resource to create an engaging learning atmosphere.
Research into classroom codeswitching has revealed similar functional bilingual/multilingual practices (Ferguson, 2003(Ferguson, , 2009;;Lin, 2013;Macaro, 2001;Van der Meij & Zhao, 2010).The use of teachers' and students' L1 is regarded as a pedagogical strategy to enhance learning effectiveness.However, as previously discussed, though translanguaging and codeswitching strategies may bear similarities in form, the theoretical premises and pedagogical implications of conducting research from these two analytical lenses differ significantly.While codeswitching studies tend to accentuate the microanalysis of the functional usage of each language involved as well as the discussion around an optimal L1 (Chavez, 2016;Macaro, 2009), studies in classroom translanguaging embrace a holistic and progressive view on the linguistic repertoire and previous learning experiences of teachers and students.This is conducted without prioritizing or devaluing any language involved, thus opening up more spaces for teachers to explore pedagogical opportunities which can potentially help students achieve their learning outcomes more efficiently.The findings of the current study indicate that although students may claim that they have a preference for a certain type of teacher talk, in reality it is how and when the teacher proactively puts into service her entire language repertoires, rather than emphasis on any given language, that has an impact on students' feedback on their learning.It can thus be suggested that in classroom discourse research, the focus should not be limited to the functional usage of one language, but broadened to translanguaging as a pedagogical asset for interactional and communicative purposes.
Additionally, these three translanguaging strategies may reflect the disciplinary characteristics of this particular educational context, namely the learning of humanities-related knowledge through the medium of English.In order to understand the extent to which discipline-specific differences influence teachers' translanguaging strategies, more research into translanguaging in EMI contexts would be required.

Students' attitudes and the monolingual approach to teaching
As discussed in the literature review, the lack of data concerning students' attitudes in translanguaging classroom research is evident and requires further study.Though Lee and Lo (2017) attempted to investigate students' attitudes towards classroom language choice, they approached the issue from the perspective of code-switching and placed the emphasis on its co-relation with learner motivation and English proficiency.To explore teachers' translanguaging practices in detail privileges one side of the interactional scenario.With little knowledge of how students perceive their teachers' shuttling between multiple languages, it is not feasible to justify its necessity and evaluate its effectiveness.In the current study, students' attitudes were obtained from two sets of questionnaires, before and after the teacher made modifications to her classroom language use.It is a step further than Wang (2019b) and Galante (2020) in the way that students' feedback is not only collected and analyzed but also incorporated in prompting changes to be implemented.An action-research orientation has provided an extra dimension, where the teacher/researcher had the opportunity to observe and modify her translanguaging practices so as to situate the students in a more efficient and satisfying learning environment.Wang's (2019b) study revealed that a large number of students opted for a monolingual approach to classroom discourse, both from the teachers and the students themselves.However, in the current study, participants unanimously expressed the opinion that making full use of one's linguistic resources in the form of translanguaging is much more conducive to learning than the traditional monolingual approach.In fact, students consider the teacher/researcher's rather infrequent (less than 15% of the classroom talk), but well-selected use of the L1 to be successful.One possible interpretation of this finding is that the participants in the current study expect not only an improvement in their English proficiency but also gaining knowledge of the subject content.In this case, their whole linguistic repertoire and previous learning experiences could be a valuable asset in classroom teaching and learning.This might explain why the participants in this study tended to favor a translanguaging pedagogy over the monolingual approach to teaching.

Conclusion
To conclude, the current research serves as a preliminary attempt to explore translanguaging practices and students' attitudes in a university CLIL classroom, and more importantly, to examine to what extent and in what way the use of translanguaging strategies can facilitate students' learning process.Data-led consideration of translanguaging can inform "decision-making for building L2 classroom environments that are engaging, demanding, and supportive of learners' development" (Hiver et al., 2019, p. 1).Due to the intrinsic nature of an action research framework, the study is limited by the subjectivity of the teacher/researcher and the small group of participants.The lack of data from students' classroom performance could be considered another limitation of the study.Future directions in researching classroom translanguaging can focus on employing translanguaging as an analytical lens to explore teachers' use of multimodal resources, as well as the implications and effectiveness of drawing upon one's whole cognitive and linguistic repertoires in EMI contexts.It would also be interesting to situate translanguaging studies in foreign language learning contexts where three or more languages are involved.This would deepen our understanding of the interplay of multiple languages and how such interplay could help teachers achieve pedagogical purposes and students improve their learning outcomes.In this sentence, the author wishes to say that 你会成为机械化野蛮人的新 物种 (you will become new species of mechanized savages), and at the same time, 也会成为会按开关的尼安德特 人 (will also become push-button Neanderthal) 这里 (Here)，the species of mechanized savages 和 push-button Neanderthal 是同位语 (are appositives).
Here Russell uses humor and irony to say that one of his ancestors died of a disease which is now rare.Because 被 斩首 (having his head cut off) is not really a disease, correct?But he is using this example in order to contribute to his previous point that he has had many long-lived ancestors.
To increase the employment of translanguaging (i.e., more frequent shuttling between Mandarin and English) when assigning homework and briefing teaching arrangements, as well as explaining answers to afterclass exercises.
In the first half of the term, we are going to cover the first, second and fourth chapters of this textbook.Within each chapter, we are going to study the first, second, third and fifth parts in class, leaving the fourth and sixth parts to be studied on your own.
From this month on till the end of this term, we will finish 第五、六、八， 这三个单元 (the fifth, sixth and eighth chapters).And as usual, in each chapter, 第四部分和第六部分是自学内 容，考试不考 (part four and six are self-study parts and won't be included in the exams).To maintain the current balance of translanguaging practices when explaining grammar and vocabulary, and telling anecdotes.
"Live off" means to provide for one's needs by taking advantage of someone or something else, 也就是靠什么来供 养，依赖什么而生活的意思 (which is to depend on something, rely on something).它和live on的主要区别 在于live on表示以什么为主食，靠 吃什么而生活 (Its main difference from to "live on" is that to "live on" means relying on eating certain food to survive)."Average out" 注意它的用法是 (pay attention that it is used as) average作 动词，且是主动形式 (a verb, and in its active form), so you say: People's wealth may differ but their sense of happiness tends to average out.或者 (or) Time spent on homework for Chinese kids average out to 3 hours per day.而不会说 (But you don't say) be averaged out.

Table 1
Key factors in two sets of questionnaires In this paragraph what the author mainly wishes to express is that . ..) the freedom most people claim for themselves is a form of right, but the freedom they adhere to others is a form of toleration.但是作者认为这是错的，为什么呢?(But the author thinks it is wrong.Why?)Because we let them speak not because we are magnanimous but because we must hear what they have to say.

Table 2
Percentages of each translanguaging strategy used in two terms