Understanding intercultural virtual exchange through a translanguaging lens in Chinese as a foreign language

2.1 Translanguaging as a theoretical framework

From the translanguaging perspective, all language users can be perceived as not merely using a specific, conventionally named language on its own at a time in their communication; rather, they deploy all linguistic, cultural, cognitive and semiotic resources available to them in the meaning-making process (García & Li, 2014; Li, 2018). This leads to a shift of the focus from separate languages to a holistic view of how bilinguals and multilinguals make use of different resources in an integrated and coordinated manner. Thus, the language practices of bilinguals and multilinguals are not considered to just be sourced from one or more separate language systems, but as one autonomous linguistic repertoire with features of different languages (García & Li, 2014).

Translanguaging can be perceived both as a practical theory of language and as a pedagogical framework (Vallejo & Dooly, 2019). It reflects the fluid and dynamic practices of the linguistic reality in the 21st century that go beyond the boundaries of conventionally named languages, language varieties, and other semiotic systems, including images, gestures, emojis, facial expressions, etc (Li, 2018). Furthermore, translanguaging highlights a very common feature of human social interaction, that is, that “language users move dynamically between the so-called languages, language varieties, styles, registers, and writing systems, to fulfil a variety of strategic and communicative functions (through) the alternation between languages, spoken, written, or signed; between language varieties; and between speech, writing, and signing” (Li, 2018, p. 26). In both everyday communication and learning in formal and informal arenas, social interaction is multisensory, multimodal and multilingual in nature, “transcending the traditional divides between linguistic and non-linguistic cognitive and semiotic systems.” (Li, 2018 p. 20). The notion of translanguaging also echoes Canagarajah’s (2015) translingual approach, which involves the use of various semiotic resources. It can also be perceived as a form of inclusive multilingualism (Backus et al., 2013) and equitable multilingualism (Ortega, 2017). As Walker (2018) argued, translanguaging affords interlocutors flexibility and diversity in language use, encouraging them to create learning opportunities for all parties involved.

Translanguaging challenges the deficit model (Li, 2011) of bilingualism or multilingualism, which takes native competence as an ultimate objective, highlighting the emerging and developing nature of language acquisition. Multilingual users may not think unilingually even when they use one conventionally defined linguistic system, such as Arabic, Chinese, French or Spanish (Li, 2018). Rather, to achieve effective communication, any conventionally named language is only one of the various linguistic resources available and deployable. In any sense-making and meaning-making process, people use multiple modes including image, text, layout, speech, videos, etc. and in the digital world, these modes of communication undergo a constant shift from isolated texts as the main source of communication to image, sound and motion being used more frequently. From a multimodal perspective, these different modes, as socially and culturally shaped resources, are integrated to create meaning in human communication (Kress, 2010).

In the creative and agentic deployment of multiple resources, language users create a social space in which they bring together their personal interests, life experiences, attitudes, beliefs and value systems to interact with each other (see Li, 2011). This was termed a “translanguaging space” (Li, 2018), wherein national and ideological dichotomies can be broken and boundaries between languages and disciplines can be transcended in engaging diverse meaning-making and meaning-negotiating processes. According to García and Li (2014), education can be a translanguaging space where teachers and students engage and challenge outdated understanding, transform existing structures and practices and generate new ones. In the last decade, translanguaging has been applied in both formal and informal settings. These range from face-to-face classrooms in content and language integrated learning (CLIL) and English as a medium of instruction (EMI) (e.g., Tai, 2022; Tai & Li, 2021a, 2021b), to informal arenas such as karate clubs (Zhu et al., 2020) and self-directed virtual exchanges (Helm & Hauck, 2022). Participants in those studies created a translanguaging space for themselves to construct and negotiate meanings in their interactions and that supplemented and enhanced their everyday life experiences. However, translanguaging has rarely been explored in VE involving CFL learners. The current study aims to fill the gaps in both telecollaboration and translanguaging studies, exploring how VE participants construct their talk with each other to create a translanguaging space for themselves during online interaction.

2.2 An overview of virtual exchange

It is well-established that virtual exchange is beneficial to learning foreign languages, digital skills, teacher education and intercultural communication (Commander et al., 2022; Fuchs et al., 2022; Hauck et al., 2020; Jiang et al., 2014; Luo & Gao, 2022; Luo & Yang, 2022; Rienties & Rets, 2022; Wigham & Satar, 2021), as it offers participants an authentic learning experience withth transformative potential while interacting with each other (O’Dowd, 2011; O’Dowd & Ware, 2009; Reynolds, 2020). Most existing studies focus on bicultural and bilingual subjects, mostly concerning English and other European languages such as French, German and Spanish as the target language and culture (e.g., Belz & Müller-Hartmann, 2002; Kramsch & Thorne, 2002; O’Dowd, 2003, 2005, 2021b; Thorne, 2003). Participants are engaged in tasks including information exchange, comparing cultural practices, working on collaborative projects (The EVALUATE Group, 2019) or comparative exploration of juxtaposing materials such as parallel texts, films with similar themes (Belz, 2003) or songs with similar themes (Luo & Gao, 2022). This contrastive and comparative approach enabled the participants to better understand a different language and culture, the complex relationship between culture and language and develop their intercultural competence. O’dowd and Eberbach (2004) reported on how language students developed their intercultural communicative competence (ICC) via a message board exchange between Irish and German institutions. Schenker (2012) explored how intercultural partners gained more insights into different cultures and developed their ICC through the VE between sixteen pairs of German secondary school students and American university students. Elsewhere, through analysing texts communicated between partners, it has been found that participants developed specific language skills such as grammar form in Spanish (Fiori, 2005) as well as the lexical-grammatical features of the intercultural discourse (Liaw & Master, 2010). Similarly, Chun (2011) explored how online exchanges can impact second language learners’ development of pragmatic competence and ICC, using data obtained from an exchange between university students in Germany and the USA. She explained how culture can be embedded in language as discourse and how foreign language learners expressed facts and opinions and negotiated meanings with their partners, displaying different discourse styles. Although these studies identified the complexity of social interaction and its contextual factors in online exchange, they did not relate the findings to CFL students.

There is an emerging interest in adopting lingua franca approaches to virtual exchanges; that is, exchanges in which neither of the partners are native speakers of the target language (O’Dowd, 2021a). English has been a lingua franca in most exchanges involving students from Spain, Sweden and Israel (O’Dowd et al., 2019), and German was also used in online interactions between students from France, the Netherlands, Finland and New Zealand (Kohn & Hoffstaedter, 2017; Korkealehto, 2022). Comparison and discussion in these exchanges covered tasks that required collaboration beyond explicit bicultural comparison and were concerned with global themes. Apart from ICC, the focus was on the development of global citizenship and leadership, as well as understanding issues applicable to different countries.

Within less than a decade, telecollaborative partnerships have extended to an increasing number of locations worldwide. Chun (2015) reviewed bicultural exchanges and identified a wide range of examples involving France and China and the USA, and the Philippines and the USA. Many of the inquiries focused exceedingly on English as a foreign language (EFL) students (e.g., Angelova & Zhao, 2016; Chen & Yang, 2016; Feng et al., 2021; Liaw & Master, 2010). Feng et al. (2021) investigated the changes in feedback types offered by native-level English-speaking trainee teachers at a US university and the factors concerning the successful uptake of EFL students at a Chinese university. Through a quantitative analysis of five weekly Zoom-based group videoconferences, the authors concluded that feedback type and linguistic level did not predict successful uptake; however, peer repair was likelier to lead to better uptake. Fuchs et al. (2022), writing about a similar university context between Chinese EFL students and US trainee teachers, reported how the teachers’ mediation facilitated task completion in online exchanges in stages including task design, implementation and evaluation.

2.3 Modes and modality in virtual exchange

Mode and modality in VE depend principally on technological affordance and practicality in various institutions for different learners. The choice of modes of communication leads to varying levels of connectivity and interactivity (Liaw & Master, 2010) and creates significant differences in online dynamics (Liaw & Ware, 2018). Certain tools such as email, forum or Zoom have enabled virtual exchanges to be asynchronous, synchronous or both in order to achieve certain pedagogical and research purposes.

The dominant modality in VE studies involves written exchanges between participants via asynchronous technology such as emails, forums and discussion boards (Avgousti, 2018). They can be highly beneficial tools for cultural learning when planned carefully, administered competently and supervised effectively (e.g., Fiori, 2005; Fuchs et al., 2022; Liaw, 2007; Liaw & Master, 2010; O’Dowd & Eberbach, 2004; Schenker, 2012). Combining tools has also been found to generate more benefits in VE. Hauck and Youngs (2008) reported how students developed closer relationships with their partners through the asynchronous tool, while a synchronous audio-conferencing environment used during the exchange enabled the participants to maintain real-time contact with native speakers. Similar findings on the effective use of videoconferencing in conjunction with emails between two target language groups have been reported in many studies such as O’Dowd (2007), Lewis and Kan (2021), where the participants were given more opportunities to discuss in more life-like conversations to clarify doubts in their email correspondence about the target language and culture. Angelova and Zhao (2016) recounted that Chinese students learning English improved their English grammar and developed their cross-cultural awareness, while prospective EFL teachers in an American university improved their teaching skills by communicating with their Chinese partners via discussion boards, emails and Skype. In a similar context, Liaw and Ware (2018) found that native English-speaking trainee teachers posted videos while Chinese EFL students used mixed modes of text and the uploading of videos and audio during their online exchanges.

With more sophisticated synchronous platforms being adopted for online delivery, synchronous communication appears to be more suitable for intercultural negotiation due to its affordances for instant clarification, questioning and expansion—although it requires better linguistic, pragmatic and sociolinguistic skills on the part of students (Jauregi & Bañados, 2008)—more effective in discussing matters of cultural interests (Angelova & Zhao, 2016; Avgousti, 2018); and more feasible when intercultural interlocutors interact with each other across borders with instant and natural feedback (Liaw & Master, 2010). Chun’s (2011) study noted more engagement and greater development of ICC in synchronous chat than in discussion forums.

Hampel and Hauck (2006) argued for a new framework to inform the development of online language teaching and learning, demonstrating how meaning can be made differently in virtual spaces with the support of Computer Mediated Communication (CMC) synchronous and asynchronous technological tools. Since then, there has been a growing interest in CMC research in terms of utilising multimodal (inter)action analysis as a new method to understand online interaction due to its nature of multimodality (Lamy, 2009; Satar & Wigham, 2020). As a “combination of different semiotic modes” (van Leeuwen, 2005, p. 281), multimodality can be described as a collection of communication practices that include textual, aural, visual, spatial and physical resources (e.g., technology that is used to make meaning and facilitate understanding) (Kress, 2010). Being multimodally competent, learners become semiotic responders and semiotic initiators with full multimodal awareness, harnessing the full potential of a wider array of meaning-making resources for online communication, whether it be visual, spoken, gestural, written or a combination of different types (Hauck & Satar, 2018). Although linguistic resources could be assumed to play a dominant role, other semiotic resources are likelier to contribute distinctively to the meaning-making process (Sindoni, 2013). Whether they deploy multimodality consciously or not, participants’ meaning-making could occur holistically (Norris, 2004), and every mode could have equal potential to make meaning (Jewitt, 2016).

The perspective of multimodality justifies semiotic resources such as image, posture, gestures an objects in the environment, including technological tools in addition to language as valid data to collect and analyse. In online interaction, participants can communicate with each other through spoken language, written information on a screen, postures, gestures, head movements, and gazes. In videoconferencing, Wang (2007) found that as semiotic tools for meaning-making, facial expression and gestures were used to facilitate the completion of task, while Satar (2013) evidenced the importance of mutual gaze and eye contact in learner interaction for positive social relations. Wigham and Satar (2021) adopted the multimodal analysis of interactional features such as gazes in teacher’s instruction in online language teaching. Gaze shifts, gaze direction, hesitation markers and raised eyebrows can all carry subtle messages in engaging students as resources along with utterances and written information on screens. Wigham and Guichon (2014) examined trainee teachers’ gestures in and out of the webcam frame and the functions of these two categories in delivering online sessions during desktop videoconferencing interactions. Satar and Wigham (2017) provided a detailed multimodal (inter)action analysis of instruction-giving practices involving semiotic resources such as gaze and gesture, as well as word stress and text chat in a telecollaborative exchange between students of French and their trainee teachers. The study exemplified how multimodal analysis can be used as both a research and pedagogical tool in teacher education and how the potential of semiotic resources can be harnessed in online language teaching.

Through detailed contextual analyses of visual data, many studies found that participants in telecollaborative activities developed their digital literacy (Bezemer & Jewitt, 2010; Dooly & Hauck, 2012) and engaged with technology in an unexpected way, which could diverge from teacher’s pre-arranged tasks (Dooly, 2011, 2018). The innovative use of technology highlights the spontaneous and inadvertent learning of languages in online exchanges. Korkealehto (2022) and Korkealehto and Leier’s (2021) reported a virtual exchange project between intermediate learners of German (their common target language) from universities in New Zealand and Finland. Students were required to upload five posts that took the form of video, audio, online sites, photos and text. The study adopted multimodality as the theoretical framework as well as content analysis for the qualitative data, which consisted of Facebook logs and transcripts of semi-structured interviews. They found that multimodal collaboration enhanced students’ digital literacy and ICC development and that student engagement was fostered through collaboration, authenticity, use of tools and teachers’ support. Helm and Dooly (2017) addressed the challenges of transcribing and representing multimodal data as well as resolving them in computer-assisted language-learning research. They argued that multimodal competence appears to be complex but worth probing further to understand the micro level of VE interaction.

2.4 VE studies in TCFL

Although a few recent reviews (Avgousti, 2018; Guo, 2022; Luo & Yang, 2018; O’Dowd, 2016) of telecollaboration accentuated that less commonly taught languages and distant cultures remained under-researched, a number of studies on VE in TCFL emerged in the last decade. As an earlier enquiry, Jin and Erben (2007) examined how Instant Messenger facilitated intercultural learning among eight CFL learners from an American university who were paired with Chinese native speakers. They found that participants’ intercultural engagement and attentiveness steadily increased while critical thinking skills and respect for intercultural differences were developed. Under similar conditions, Jin (2013) reported participants’ positive target language development in a 10-week VE project between 10 pairs of university students from China and the USA. She reported that CFL students steadily increased the number of Chinese characters produced, expanded vocabulary and improved reading ability; however, the quality of writing did not improve rapidly. Adopting a sociocultural theoretical perspective, she suggested that a scaffolding and feedback mechanism must be built into future training to improve productive skills in VE. Sharing a similar perspective, Lewis and Kan (2021) recounted the benefits and challenges of six-week e-tandem learning between CFL learners from a British university and EFL learners at a Chinese university. Using datasets consisting of email messages, learning diaries, Skype conversations, online surveys and follow-up interviews, they identified a significant improvements in CFL learners’ language proficiency and intercultural learning while carrying out authentic communication with native speakers. While CFL beginners gained more understanding and increased their knowledge of Chinese culture, the primary concern of the students was the limited proficiency of the target language that affected learner engagement. Learners at the same level in Chen’s (2017) study also improved their linguistic accuracy through VE, as did students in an advanced Chinese Business class in a British university in a study carried out by Wang et al. (2013). The latter used wikis as the platform for their written exchange with students of EFL from a Chinese university. In the work of Ryder and Yamagata-Lynch (2014), seven intermediate-level CFL students tele-collaborated with partners who majored in TCFL and were chosen according to their English proficiency test results. They applied activity theory to analyse online communication, identifying tensions between participants in their VE. From the perspective of social network analysis and adopting mixed methods, Rienties and Rets (2022) explored the impact of VE between university students from China and Portugal on their intercultural relationship and multiple competences. Using the multiple data corpus, Tang et al. (2021) provided detailed analysis of types of linguistic errors made by CFL learners in terms of grammar, lexis, idiomatic expressions and error correction strategies adopted by their VE partners.

Xu et al. (2022) investigated how L2 learners of Chinese and English develop linguistic competence, social skills and digital literacy in their e-tandem project between British and Chinese universities over a period of two years. They reported that participants frequently used translation software to resolve language issues. As the only study to our knowledge that adopted CA as part of the data analysis of VE interaction; the authors proved that students developed multimodal strategies to scaffold effective communication in the target language. In their study of exchanges between EFL students at a Chinese university and CFL students at an American university, Luo and Gao (2022) explored how the use of songs with the same themes in both languages can promote intercultural learning. Adopting Byram’s (1997, 2021 intercultural competence assessment model, Luo and Gao analysed the four varieties of data comprised of transcripts of a WeChat group discussion and end-of-semester interviews, videoconferencing audio recordings and CFL students’ final reflection journals. Despite a close examination of the qualitative data, they did not reveal the minute details of the telecollaborative process through multimodal conversation analysis.

The literature review suggests that, as with existing inquiries involving other languages, most research on VE in TCFL concentrates on text as the main modality and less on other modalities. In a few studies that recorded online conversations between participants, attention was primarily paid to the development of language and cultural benefits as well as to the problems and challenges in VE. These studies reported on the issue of varying levels of target language proficiency between partners (Guo et al., 2022; Lewis & Kan, 2021; Luo & Gui, 2021; Luo & Yang, 2022; Tang et al., 2021); however, how best to design tasks to mitigate the issue has yet to be investigated. A closer scrutinisation of minute details via the MCA of naturally occurring conversations between intercultural partners during their VE interaction in TCFL has yet to occur. Sociocultural and ICC theories have been applied more than translanguaging, which is rarely mentioned in VE as an analytical framework. The current study is underpinned by the translanguaging approach to explore how CFL learners share their knowledge-building and meaning-negotiating processes in VE, using MCA to analyse online multimodality interactions at the micro level. Through a translanguaging lens, this study hopes to address the imbalanced level of proficiency in the target language between intercultural partners, incorporating various resources and engaging more multilingual and multimodal means.

2.5 Research questions

The current study aims to answer the following research questions (RQs):

3.1 Context, procedures and participants

The data were collected from a nine-week VE project between a British university and its partner university in Qingdao, China, in the academic year of 2021–22, when the pandemic made it impossible to conduct in-person exchanges. 22 CFL students at various levels were paired with native Chinese-speaking peers who were in their first year of postgraduate studies in TCFL. Microsoft Teams was the tool chosen due to its accessibility and convenience of monitoring for both sides.

Teachers of both universities paired their students, coordinated with them and had them exchange emails in Week 1 before setting up the first group meeting in Week 2 with all 22 pairs (another meeting took place in Week 9). Between Week 3 and Week 8, each pair was required to have a weekly one-hour synchronous exchange on MS teams, and they had the autonomy to arrange for mutually convenient meeting times themselves. In the task briefing, participants were given total freedom as to the topics they would like to talk about with their partners, and they were also encouraged to deploy all the resources while trying to use the target language as much as possible. Among those self-initiated topics, they had to choose one to present in the final week (Week 9), showcasing their project to all the participants in this project. Again, they could choose their preferred format, which could involve videos, PowerPoint presentations, stories, songs, etc. Among the most popular topics were food, festivals, popular songs or movies, university life, hometowns, music and history.

In addition, CFL students were required to write a paragraph in Chinese on any chosen topic that their partners would provide their feedback on in terms of linguistic accuracy. The feedback session was incorporated into weekly online meetings. As a way of recording their VE sessions, the students were encouraged to upload their weekly learning diaries, audio/video recordings of their weekly online meeting (subject to agreement of both parties) and their final reflective reports to the British university’s virtual learning platform. They were also encouraged to complete an end-of-project questionnaire on an anonymous and voluntary basis. There were two parts to the questionnaire. Part I asked questions about language learning background and Part II focused on their perceived development in terms of ICC (Byram, 1997, 2021) (see Appendix I for sample statements). Part II encouraged participants to reflect on their VE experiences so that they would become consciously aware of their achievements or areas they should improve. Additional consent was sought from those whose audio/video recordings were used for further study. Pseudonyms were used in this study to protect participants’ identities.

The final data set consisted of 22 completed questionnaires, learning diaries (LD; 110 in total) and 14 audio/video recordings of online meetings from six pairs for roughly 15 h in total. The profiles of participants who spoke more than two languages and the number of audio/videos recordings are listed in Table 1.

As one of the aims of this study is to explore how multilinguals make use of their existing resources, recordings of CFL participants who could speak more than two languages competently were chosen for analysis. English was the mother tongue for five of the participants from the British university, while one had French as her mother tongue. All of them could speak at least one other European language competently. They all learned Chinese as an optional module or part of their degree for different durations—between less than one year and five years. Two of the Chinese partners began learning English in secondary school, and, respectively, Korean and German while taking up their first degree in university. The other four Chinese participants were not listed here, as they did not have any foreign language other than English. These pairs recorded their online exchange sessions (each session lasted between 20 min and a little over 1 h), the transcripts of which constitute the main data for this study. The analysis with the findings will be explained in Section 4, along with typical LD entries about the participants’ weekly experience. For the entries in Chinese, English translation was provided.

3.2 Data analysis

In order to catch the minute details of the natural occurrences during virtual exchange sessions, the study adopted MCA as the main data analysis method. Transcription includes textual, visual and oral modes along with non-verbal cues (Helm & Dooly, 2017). Transcribing multimodal data entails multiple modes, including both linguistic and paralinguistic elements, silence and still and moving images and artefacts (Satar, 2016). In meeting the challenges of transcribing and representing multimodal data (visual, oral and textual), the choices on what to transcribe were made according to if and how they could address research questions, as it is not possible to transcribe everything (Helm & Dooly, 2017; Satar, 2016). In this study, the researchers viewed the video recordings separately a few times to figure out how linguistic items and cultural phenomena were explained and transmitted between the partners. For discussion, they then identified the most relevant episodes in accordance with the translanguaging approach. They were transcribed to include descriptions of multilingual, semiotic and multimodal cues such as gestures, emoticons, facial expressions, images and videos (see also Dooly & Hauck, 2012). The detailed transcription convention adopted by this study is listed in Appendix II (see more details of MCA transcription in Mondada, 2018). The transcription mainly includes oral interactions in Chinese and English. For utterances in Chinese, English translations are placed in curly brackets. Relevant non-verbal cues, such as gestures, and other visual modes, such as pictures and images, were also integrated into the transcript (Helm & Dooly, 2017). After the initial transcription and turn-by-turn analysis (Wigham & Satar, 2021), the researchers discussed again to decide on the final transcripts, whose accuracy of notes on gestures, images and translation was cross-checked before the final analysis was conducted. The three extracts were selected to show how intercultural partners interacted with each other in their online sessions.

4.1 Coordinating to confirm cultural norms while reinforcing linguistic elements

One of the objectives of the VE in this study was to assist CFL students in learning Chinese language and help TCFL postgraduates (prospective teachers) practice Chinese teaching skills. In the actual exchange sessions, language learning and cultural knowledge exchange were intertwined in the conversation flow between the partners. Although in some situations the CFL students are ‘language learners’, it is not always the Chinese native speakers who introduce or ‘teach’ their culture or language to CFL students. In fact, during the VE processes, both parties constantly bi-directionally exchange their knowledge and experience based on their own cultural backgrounds. There are many moments in which the Chinese native speaker students initiated topics to confirm certain cultural facts or norms from the CFL partners. Many language and cultural learning opportunities and moments are embedded in this process as well. The extract below is evidence of such exchanges on cultural norms that enhances the linguistic elements of the two partners whose target language proficiency is imbalanced. Sailin began learning English in secondary school on and off for at least 10 years, while her CFL partner Eleanor had begun learning Chinese around two years prior. In Extract 1, the partners talked about their families, and Sailin came up with a specific question and used this opportunity to seek confirmation of a British cultural norm in the process of which Chinese vocabulary items were learned or reinforced.

Extract 1. Sharing information about the family.

[S: Sailin Qingdao (China) participant and E: Eleanor (UK) participant]

1. S: Eleanor can … can I ask you a question?

2. E: enhuh?

3. S: So I … I err … in my class so I have read some … book or article just say err … in England … en

4. people don’t like to talk about your family because err … they think this is privacy (2.0) right?

5. Bu … but … you can talk about it, so maybe this is not right, yeah?

6. S: I have read some articles talk about the differences between English people … er the people

7. from England and the people from China. So sometimes in China and we just like to share

8. with our friends to talk about our family, but the article just says in England people don’t like

9. Talk about the family with other people, because this is a privacy.

10. E: Oh!

11. S: So, yeah? (( ))

12. S: So, does it right?

13. E: Honestly no, I think 我 我喜欢 oh god I don’t even know how to say this … 我喜欢谈谈 说

14. 我的家 with my friends

15. S: 啊 你喜欢 你喜欢谈你的家人

16. E: 嗯! er 我们 我的朋友 er 我们说 … 的 … 我的家 our? How do you say our family? 我的

17. S: 家?我们家 我的家人 our family

18. E: 我的家人

19. S: 对

20. E: Maybe the article is talking about it more negatively, some people don’t like to talk about their

21. family negatively to other people

22. S: ah ok

23. E: But err personally 我喜欢 er I find it very interesting er so yeah that’s interesting that you’ve

24. read an article on it but … yeah

25. S: Okay: So: I think this is a stereotype, right?=

26. E: =Hmm yeah

27. S: Stereotype of England because right now, I talk with you and I know this, the article just say,

28. is not right

29. E: Yes.

30. S: Okay.

During their conversation in Extract 1, S initiated a request very carefully and politely (Line 1, “Can I ask you a question?”) as she intended to probe a cultural fact from E. After E’s positive response in Line 2, S began to illustrate her query and confusion about a culturally specific norm that she learned from other channels, such as reading about whether English people like to talk about their families due to it being a private matter (Lines 3–5). In order to clarify what she was after, in Lines 6–8 she offered what she thought the Chinese norm was in sharing information with friends about one’s family. Her understanding was that this was very different from the British norm that she had read before. After rephrasing in different ways, thanks to her English proficiency, she managed to convey her enquiry well to her partner. Immediately after S’s comparison and illustration, E’s response was “Oh!” (Line 10) in a surprising tone, showing that what S said was something unfamiliar to her as well. Hearing this, S further pursued a confirmation and explanations by asking, “So, yeah?” and “So, does it right [sic: is it right?”] in Lines 11 and 12. Although her grammar was not correct, her partner understood her. In Line 13, E first confirmed that what S read before might not be true based on her own British cultural experiences. She even added ‘honestly’ to indicate her affirmativeness. She then switched the target language, attempting to use Chinese to explain her interpretation of the British norm. Due to her limitations in terms of Chinese language proficiency, she was only able to express her main idea in Chinese with the assistance of English (Line 14, “with my friends”), clarifying that she herself liked to talk about her family with her friends. She also switched to English to fill in gaps or pauses possibly caused by searching words and choosing between 谈谈 and 说 (the former being slightly more formal than the latter). In Line 15, S repeated what E said in a proper complete Chinese sentence to acknowledge her receipt of E’s explanation and confirm her understanding of the explanation with E. Meanwhile, she offered correct linguistic input, serving as a recast (a form of feedback) for her partner. S’s feedback might have taught or reminded E of the Chinese word for family (家jiā). In Line 16, E confirmed that S’s understanding of her previous talk in Chinese was correct and tried to pronounce it again. In Line 16, E also grabbed this opportunity to use the word “family” in Chinese that she was not very familiar with before. She initiated a question in English and proposed an answer in the form of a question, hoping to double-check with her partner if she was correct. In Line 17, S assured her partner about how to say the word in Chinese and offered another option, “家人”(family members), which was followed by the answer with an English explanation to assist her understanding. E practiced the word in Chinese again in Line 18, and S confirmed its correctness in Line 19. After this language learning moment, they went on to further discuss what S had read, with E trying to provide an explanation or possible analysis of the author’s intention in Line 20. That is, what S had read could be that English people may not share something negatively about their family with their friends (Line 1). She further commented on the fun experience in Lines 23 and 24. They finally (Lines 25–30) coordinated to come to the conclusion that the article S had read was only a stereotype or a partial reflection of a British cultural norm.

The interlocutors in Extract 1 did not have the same level of proficiency in the target language to enable them to converse effectively—half in Chinese and half in English. From the recording data, it is clear that they spoke more English than Chinese in their online exchanges. This was the case among the many pairs in this VE project. However, it was still clear that both linguistic and cultural items were exchanged between the partners, even though cultural learning seemed to have dominated in their interaction. The LD entry below, by Tina, whose mother tongue was French, illustrates similar intercultural learning:

我这周了解到万圣节在英国很重要, 但在中国不一定。他们不会看恐怖电影, 不会做“trick-or-treat”, 等等。 然而, 在中国越来越多人喜欢圣诞节因为他们可以交换礼物, 但这不是普遍现象。比如说, 我的伙伴不会做装饰圣诞树。所以我们庆祝的方式不同。 关于尊重的概念, 我了解了与英国相比, 人们的地位要重要得多。比如说, 在中国他们会用“老师” 还是“Miss”但是在[英国]用过有时候, 老师告学生他们可以叫他们用他们的名字。

English translation: I learned this week that Halloween is quite important in Britain but not so much in China. They would not watch horror movies, play trick-or-treat, etc., but more people like Christmas as they can exchange presents, which is still not common. For example, my partner would not decorate the Christmas tree. We have different ways of celebrating. With regard to the notion of respect, I learned that people’s social status in China is much more important than social status in Britain. For example, they would address their teacher as Laoshi (note: it means “teacher”) in China, while in Britain it is “Miss”; however, teachers may sometimes let their students call them by their first names. [Tina – LD week 2 entry].

The above entry from Tina, a native French speaker who studied Politics with Chinese at a British university and had been learning Chinese for five years at the time of this study, shows how participants enjoyed learning customs in different countries through online exchange.

4.2 Collaborating in joint projects while negotiating meaning

In order to fulfil their learning goals, in addition to casually chatting with each other on any topics they were interested in, the pairs also had to complete their collaborative project in a format of their choice (e.g., PowerPoint presentation, video, screencast) to demonstrate their language and culture learning outcomes in the VE sessions. The topics of the presentation could cover something that they had worked on that represented the cultures of both participants. Therefore, during the VE sessions, there were many moments when participants co-constructed their translanguaging space to discuss their collaboration projects. Since participants had different levels of proficiency in the target language, they had to employ both linguistic and other multimodal resources to scaffold each other’s understanding, reaching common ground in their project collaboration and fulfilling their communication and learning goals. Extract 2 is an example of how the partners negotiated meaning to arrive at an agreement on what to do in their joint project.

Extract 2. Deciding on the collaboration project.

[L: Liufang, (China) participant and H: Harry, (UK) participant.]

Note: see Appendix II for the explanation of transcription convention, such as ↑.

1. L: So: other groups, they have decide: what to ↑show (.) the: ↑PowerPoint in the: last two weeks

2. So, I think we should=

3. H: =Ok

4. L: Oh Oh ((puts two hands on her forehead)) the Germanish ((laughs))

5. ((both laugh))

6. H: So yeah (.) we can ↑decide (0.2) you mentioned maybe doing a little bit on (0.5) a bit a bit of

7. everything (.) So: maybe doing a bit on (0.2) a bit on music, a bit on: ermm food (.) a bit on:

8. errmm Yeah(.) is that is that what you (.) were talking about what you’re doing? Maybe a little

9. bit of (.) lots of things?

10. L: ((nods)) I think: we can err we can make the ↑PowerPoint (.) and err every—every week we

11. can talk about other ti tima-to- topic ((smiles)) topic (.) other topics (.) ↑sorry

12. H: That’s ok

13. L: So: er how about food?

14. H: oh ↑yeah

15. L: en ok: (.) I can make: er the Chinese part and you make er the English part, so then ((puts two

16. hands together)) mix

17. H: ((nods)) Perfect

18. L: OK

In Extract 2, H had only been learning the Chinese language for two months, and L’s major for her bachelor’s degree was in German (during their early sessions, she mentioned that she had limited English proficiency). However, they had to communicate in English, as it was not possible to use Chinese when they discussed their collaboration for the upcoming presentation project to fulfil learning goals. Hence, non-verbal cues were more important than words in making and negotiating meaning. H and L discussed how they would present their project, what roles they would take, and how they would communicate as they collaborated. In Lines 1–2, L suggested that the format of their collaboration project could be PowerPoint (with her voicing becoming louder and tone rising inadvertently when saying so), just as other groups would do. While H agreed with “OK” (Line 3), L commented on her own “German-ish” (coined from German and English) and apologised for not being able to articulate more clearly in English, accompanying this with hand movement (putting two hands on her forehead) and non-verbal behaviour (laugh). In Line 4, both signalled that they felt amused or embarrassed about using gestures due to the language barrier. They both laughed in Line 5 to sustain the turn-taking sequence and move it to H, who tentatively suggested in Lines 6–9 that their content could possibly incorporate all that they had talked about until then, since music was in their previous session and food was in the same video later (both extracts were not shown here). L nodded in Lines 10–11 to show her agreement, reiterating that they could use the PowerPoint from each week and confirming that H was right in that they had talked about other topics during the weeks, with her smile to indicate that she was not objecting to his suggestion. However, she thought that food could be a better idea, which she pinpointed in Line 13. This was endorsed by H in Line 14 with a positive note and a rising tone. Then, L suggested the possible distribution of work between them in Line 15, with herself being the one to prepare the PowerPoint slides about Chinese food and H for English food. In Line 16, she again applied both her simple English and hand gestures to state that the final product would be a combination (a “mix”) with one part in Chinese and another in English.

Throughout turn-takings between Line 4 and Line 17, it is clear that H and L negotiated their meanings and reached their decisions with linguistic elements, e.g., short English sentences, and non-verbal cues, e.g., nods, hand movement, smiles, raised voices, etc. They activated multimodal resources to scaffold each other’s understanding and co-constructed their collaboration project, achieving their communication purposes so as to meet the learning expectations of the project.

Data from other participants’ weekly LD entries also support the finding of translanguaging practices, in that VE enables the use of all available multimodal resources:

There was a little bit of a language barrier. My Mandarin isn’t great, so we stuck to English. Whenever one of us couldn’t understand what the other meant, explaining with hand gestures generally worked! [Otna – LD Week 1 entry].

For quite a lot of the time, we struggled with communication due to my very limited Chinese, and my exchange partner’s spoken was is not great. We resolved both problems by using visual images or using translation on WeChat, which made it easier to understand each other. [Lynn – LD Week 3 entry].

Both the LDs and Extract 2 present typical scenarios among many pairs in the VE project when both or one partner lacks proficiency in the target language to convey the meaning fully with only verbal or written texts. By utilising bilingual and other multimodal resources, it is clear from the data that they achieved a communicative purpose, i.e., reaching an agreement on the topic and format of their collaboration project.

4.3 Creating mini learning sessions while deploying all resources

During their VE conversations on self-directed topics, participants provide many opportunities for language and culture learning, whether intentionally or incidentally. Food was among the most popular topics in the VE sessions. When their linguistic resources were limited, the partners took advantage of all the means and resources surrounding them to convey their cultural and linguistic messages to their partners. These include technological tools, such as mobile phones and internet; audio, visual, spoken and written modes; and actual objects in their living environments. They created learning opportunities in an accessible and resourceful manner. With the same pairs as in Extract 2, Extract 3 provides evidence of the co-creation of learning opportunities through the deployment of all possible resources to facilitate communication.

Extract 3. Harry’s demonstration of British food.

[H: Harry, (UK) participant and L: Liufang, (China) participant]

Earlier, in the same recording as the extract below, L introduced Chinese foods by showing pictures and videos of Chinese foods she had downloaded from the internet.

1. H: Then er chips I don’t know what chips is in Chinese, but [they’re like

2. L: [Cheeps? C-H-E-E-↓P

3. H [Ermmm:

4. Er: Yes C-H-I-P-S ((uses index finger to write the letter in the air as he says each letter))

5. L: C-H-I-P-S, [Chips?

6. H [Yes tha-that

7. L: [Oh!

8. H: [That’s ermm: This is where American English is different to British English↑(0.5)

9. Ermm:(.) because in America (.) chips are like ermm crisps (.) so do you know like:a: the ones

10. you have in a packet ((uses one hand to show the packet; the other hand makes a ‘putting in’

11. gesture)) That you eat, you can just=

12. L: =Chips? ((uses one hand to hold a pen; the other hand forms the shape of the pen))

13. H: ↓Yes Umm Han- Let me show you, hanging on (.) I’ll quickly show you=

14. L: =hehehe

15. H: Errr: So::: ((starts to share his screen via MS Teams function, see Figure 1))

16. ((H typing in ‘crisps’ on the screen))

17. H: So: Erm: this (1.0) Oh no, what I’m I doing ((clicks the wrong tab)) – ((the screen starts to

18. show pictures of crisps)) So these are err in ↑England, so where I am ↑from (0.2) we call

19. them CRISPS=

20. L: =↓Oh hehehehe ((laughs)) ↑OK

21. H: But in hehe in er:America, these are called ↑chips ((sharing screen, see Figure 2))

22. L: Hmm I ↑know ((nods))

23. H: Which is why it’s confusing, but whereas if I search chips ((types ‘chips’ on the screen))

24. ermmm here (0.5) ((pictures showing up chips)) this what we would call ↓chips.

Figure 1:

Images of crisps.

In Extract 3, H initiated the topic of the food/snack “chips” to his partner in Line 1, not knowing what it was called in Chinese. Hearing the pronunciation, L tried to spell the word in Line 2, although she was wrong in spelling “ee” instead of “I”. Without pointing out the mistake straight away in Line 3, H recast it with the correct spelling while moving his index finger in the air in Line 4 and then confirmed it, following L’s request in the form of a question in Line 5. Realising that this word might still be unfamiliar to L with her “oh” in Line 7, H explained a different use of the word in American English, which resembles “crisps” in British English. To clarify what he meant by “crisps”, he showed L one pack of crisps that he might have prepared in advance or left randomly at his desk, moving his hand as if he was eating them (Lines 8–11). However, L did not seem to fully understand his explanation and maintained a fully doubtful voice (Line 12); H decided to seek other ways to explain his topic to L. In Lines 13 to 15, H started sharing his screen using the function of the platform (see Figure 1). After looking for pictures of crisps on the internet by typing the word on the screen in Line 16, clicking the wrong tab and finally showing these images until Line 19, H managed to get the message across to L, who laughed and finally understood it by Line 20. Then, H continued his further explanation in Line 21 that it was called “chips” in America. With L nodding to show understanding, H typed “chips” on the screen to explain what “chips” meant in Britain through the images that appeared (see Figure 2). H created a mini vocabulary learning session with a real learner on the spot. While showing images to accompany his explanation, he made his session very engaging.

Figure 2:

Images of chips.

This was similar to L’s earlier teaching of Chinese foods via pictures and videos, which can be further confirmed by Harry’s LD entries:

I have learned some common Chinese foods and the manner in which they are made. I also had the chance to watch a video that showed what it was like to go out for something to eat (小笼包) in China and how to eat them. This reminded me of being in Spain and going out for the Christmas dinner—tapas placed on the Table. It was through observing other people that I was able to work out the best way for me to eat different foods. [Harry-LD Week 2 entry)

I have learned a lot about the different dynasties in China and some specific Chinese phrases that can be seen above. I also watched an interesting video about the Terracotta Army. [Harry-LD Week 4 entry]

The extract and the LD entries above demonstrate that bilinguals/multilinguals are able to deploy all resources available to them to construct meanings, be it linguistic or semiotic, verbal or non-verbal, text or images, according to their partner’s specific situations. It also illustrated how intercultural partners benefit from their VE sessions, where their construction of meaning cannot be based just on vocabulary of any language, but inevitably rely on multimodal resources.

5.1 RQ1: How are meaning negotiated and communicative purposes achieved in VE sessions through intercultural interlocuters’ translanguaging practices?

This study adopted the translanguaging perspective to make sense of VE interaction, in which intercultural participants deploy complex multilingual and multimodal repertoires in the process of meaning construction. Through coordinating to confirm cultural norms, collaborating in joint projects, and creating mini learning sessions, it is clear that meaningful communication between target language learners and competent target language speakers has been facilitated by translanguaging practices and has been assisted by images, screensharing, videos, audio and text-based online tools. Applying these translanguaging practices in VE interactions, the subjects sought confirmation from each other by constantly employing various means beyond just switching between the two linguistic systems. Code-switching, as one form of translanguaging, emerged only occasionally in their online interactions. For instance, S and E in Extract 1 conducted transitions freely between Chinese and English to fill gaps due to their lack of vocabulary in either language until the message was conveyed to the other side. In the end, they learned reasonably well about each other’s cultural practices in sharing family information with friends. In the process, Chinese words were taught or reinforced and the whole conversation appeared polite and enjoyable despite their imbalanced level of proficiency in the target language.

In Extracts 2 and 3, H and L attempted to decide the topic and format of their collaboration project, which was the final task of their VE sessions. L’s non-verbal means, such as gestures, helped herself find the words she knew to sustain the turn. She nodded and smiled to show her understanding and agreement with H, when he expressed his intention about how to proceed; however, her body movement, such as putting her hands on her forehead and her slightly embarrassed facial expressions, indicated her lack of sufficient comprehension of H’s suggestion. During their conversation, she held her mobile up for translation now and then and read out the English sentences there to request that H write in the chat box so that she could read to understand him, as her reading skills were much better than her listening skills. H, on the other hand, seemed to have understood the communication barriers between them in terms of insufficient common language proficiency in either English or Chinese. He also chose the apt images to show the differences between chips and crisps and possible confusions caused by lexical differences between British and American English, as such differences are common knowledge to many English speakers in the UK. He also slowed down his flow of speech to enable a better understanding of his partner, smiling and waiting patiently for the other side to formulate her questions. All of these seem to have enabled his partner to understand his messages without compromise. With their learning goals in mind, they utilised an array of available resources and multiple modes throughout the process of negotiating meanings, despite the limitations of their proficiency in a foreign language. They also played the role of a teacher effectively, perhaps without realising that it could be fulfilling and satisfactory. These sessions offered insight into such social spaces, or “translanguaging spaces” (Li, 2011), enjoyed by CFL learners and their partners.

The data from the questionnaire supports the above finding, as all of the 22 pairs completed their collaborative projects and would like to take part in future VE projects (they either “strongly agreed” or “agreed”), although two pairs did not showcase their project but uploaded their recordings later in the final week.

Our data support previous studies in that translanguaging practices are emerging in VE sessions (Helm & Hauck, 2022) when participants bring together their linguistic repertoire, personal interests, life experiences, attitudes and beliefs to interact with each other. Our study also resonates with previous studies (e.g., Jewitt, 2016; Satar & Wigham, 2017; Wang, 2007) where multimodal resources made distinctive contributions to online task completion. The use of translation found in this study lends further support for the strategy adopted by participants in Xu et al.’s (2022) study, who frequently resorted to translation software to resolve language issues.

5.2 RQ2: How do VE partners create a translanguaging space for bilinguals and multilinguals with Chinese as a foreign language (CFL) to facilitate intercultural learning?

The analysis of the three extracts, learning diary and questionnaire data demonstrated that the translanguaging space was created when participants were informed that they had the freedom to choose their topics, making use of all linguistic and non-linguistic resources to achieve authentic communication purposes. In this space, participants can voice their thoughts and facilitate their intercultural communication with multiple modes in a more dynamic environment. Across our data, participants usually spontaneously or incidentally co-constructed cultural understanding during their casual conversation by employing translanguaging practices. This lends further support to Walker’s (2018) argument that flexibility in language use and the creation of learning opportunities are afforded by translanguaging. Virtual exchange between language learners from different cultural backgrounds becomes a prototypical translanguaging space.

From the three extracts, we can see that the translanguaging space S and E in Extract 1 enhanced their knowledge and perceptions of each other’s cultural background, renewed their knowledge and challenged their assumptions or stereotypes about the other culture. Although H and L had limitations in terms of their target language proficiency, they actively showed their ideas and suggestions to each other and constantly coordinated to resolve the issues that emerged during their conversation. They employed semiotic resources aside from linguistic resources, such as body language, eye gaze, and facial expressions, to holistically facilitate each other’s understanding and online communication (Dooly, 2018; Norris, 2004).

H, as an emerging trilingual who spoke English as his mother tongue, learned Spanish and Chinese for his degree and realised that his partner might not understand him very well; therefore, he simplified his English and reduced the speed of his speech while explaining British foods with pictures from the internet. L’s first foreign language was German, and she did not understand her partner’s English very well. She jokingly admitted herself to use her German-ish, although her pronunciation was good, and her reading proficiency in English was better than her speaking and listening proficiency. She managed to write her script in advance while preparing for each session. In another extract not shown here, she chose the right pictures and videos downloaded in advance that appealed to her partner, introducing food varieties of her hometown and other places she went before, and involving him in answering questions after watching the video. She also encouraged her partner to follow her in pronouncing the names of the food in Chinese and offered him apt feedback.

According to the traditional deficit model (Li, 2011), neither H nor L had sufficient proficiency in the target language, i.e., Chinese or English, to enable them to communicate with each other effectively in either language. This confirms the findings from existing VE studies involving CFL, where linguistic barriers were among the concerns for both sides (Lewis & Kan, 2021; Jin, 2013). However, throughout their VE sessions in this study, the participants looked very happy in the videos and expressed no anxiety in their LDs and reflections. Extracts 2 and 3 illustrate how they made good use of all the resources available to negotiate and co-construct meanings and inform each other of their own cultural products and practices through translanguaging. Their use of linguistic means appeared to not be much more useful than other semiotic and multimodal resources, such as pictures of foods, gestures, hand movement, facial expressions, etc. This coincides with Satar’s (2016) analysis of the use of facial expressions and gestures as facilitators for task completion. They also resorted to searching for information on the internet and using mobile devices flexibly, moved from one tool to another with ease to maximise their ability to communicate. To a large extent, translanguaging practice has empowered online intercultural interlocutors to bridge the gaps in communication caused by insufficient linguistic proficiency in their target language. This challenges the traditional view in bilingual research, where low proficiency leads to poor performance in a second or foreign language. For example, American high school pupils’ lack of everyday vocabulary in L2 Spanish was among the reasons for their poor word decoding and reading comprehension results (Sparks, 2015). The language deficit that Asian students have in English at Western universities negatively affects their critical thinking skills for exhibition in linguistically demanding tasks such as debate, discussion and essay writing, which are often required in academic studies (Rear, 2017). In virtual exchange, each individual is eager to make their exchange partner understand what they say rather than perform tasks such reading comprehension in a language class; thus, communication becomes the primary goal. It is clear from our data that translanguaging and multimodality have not only been taken as a strategy to address linguistic limitations but also allow flexibility and inclusivity in language use (Walker, 2018). Translanguaging practices in virtual exchange facilitate the meaning-making and learning processes; in so doing, linguistic limitations become less of an issue and are compensated by the successful completion of a communication task.

Translanguaging space not only encompasses different linguistic structures and different modalities, but also entails individuals’ sociocultural identities and life experiences, as well as the “teaching” role. Through this space, language users subconsciously and creatively construct knowledge, negotiate meaning and modify their understanding. In VE sessions, two or more sets of cultural practices and values are brought into play, and participants leverage different resources to achieve various communicative purposes—be it establishing rapport with their partners or working on collaborative projects.

The translanguaging practices of the participants in this context have been discursive and comprehensive. This finding concurs with previous studies (e.g., Dooly & Hauck, 2012; Li & Ho, 2018; Tai, 2022) in that linguistic, semiotic and sociocultural resources were leveraged together in learning. In this study, participants’ responses and interactions demonstrated that they coordinated their own way of communicating for collaboration, navigating through the translanguaging space they created for themselves. While CA used in Xu et al. (2022) revealed that peer scaffolding and multimodality scaffolding were adopted to co-construct language-related knowledge, the adoption of MCA in this current study provides insights into more varied and enriching details on how an array of multilingual, multisemiotic and multimodal resources were deployed freely to create a translanguaging space in a virtual exchange project.