2. The idea of translanguaging

The idea of translanguaging originated in the Welsh revitalization programmes, where the medium of instruction was supposedly Welsh and the content of the teaching was not language per se but other subject matters – that is, Welsh medium schools and classes. It is important to remind ourselves that Welsh monolingualism does not exist in the community. The pupils and the teachers in the Welsh medium schools and classes are bilingual in Welsh and English and many, especially the pupils, feel more proficient and comfortable in English. Cen Williams, an experienced teacher and teacher trainer, who was doing a doctorate at the University of Bangor, observed a commonly occurring classroom situation where the teacher would give instructions and ask questions in Welsh, as the school policy expected, but the bilingual pupils would respond in English, naturally. Occasionally they would read something in English, and discuss it in Welsh. So listening, speaking, reading, and writing were done alternatively in either Welsh or English and sometimes a mixture of the two. To the eyes of school administrators these practices were not desirable, as to revitalize a minoritized language through schooling, the assumption often was (and continues to be) to insist on using the target language monolingually to maximize both input and output in the target language.

Cen Williams, however, had a different view and made a different argument. Rather than seeing this alternating between the two languages as a problem or barrier to learning, he viewed it as a way of maximizing the bilingual potential in learning. Thus, as an agentive learner strategy to start with, translanguaging was initiated by the learners, rather than the teachers. Cen Williams was suggesting to respect the fact that the pupils were bilingual, not monolingual, and they lived in a bilingual environment. The idea of translanguaging was, therefore, developed out of considering the students’ bilingualism as integral to the students’ lives. And in this respect, weaving these lives into classroom practice became the bedrock of translanguaging pedagogy.

As translanguaging scholars, we build on Cen Williams’ (Reference Williams1994) early ideas of trawsieithu (as he originally coined it in Welsh), and Colin Baker's (Reference Baker2001) translanguaging, which invoked the extensive scholarship on languaging. Critically, the most important point is that it was a learner initiative that emerged out of their natural instinct as bilinguals to resist the imposition of monolingual instructional policies. The question then is, and that was precisely the question Cen Williams wanted to address: should teaching be aimed at encouraging and facilitating positive (in the sense of ‘productive’ and ‘beneficial’) use of what comes naturally to the learner, or imposing some other behavioural norms on them? This is a genuine question because the classroom, whether it is a language or a content classroom, is a specific, socially constructed space, often divorced from the everyday social realities of the students’– and indeed the teachers’ too. Home schooling during the COVID-19 pandemic reminded us of that very vividly. And we must not forget that not all learners in the world have the privilege of having access to what we in the Global North understand as ‘the classroom’. To what extent should we try to bring a bit of the social and sociolinguistic reality of the community in which our learners live into the classroom is a question about inclusion and social justice to me.

For an exploration of the development of the concept of translanguaging, see Lewis et al. (Reference Lewis, Jones and Baker2012a, Reference Lewis, Jones and Baker2012b). Cenoz and Gorter (Reference Cenoz and Gorter2021) present an up-to-date account of the work that follows the original conceptualization of pedagogical translanguaging. Key landmark works that expand the concept of translanguaging beyond pedagogical practices include García and her colleagues’ work (Reference García2009) in the US on the education of minoritized and racialized bilingual learners in a decolonizing education through an equity and social justice agenda (García, Reference García2009; CUNY-NYSIEB project: https://www.cuny-nysieb.org). In the UK, Angela Creese, Adrian Blackledge, and others have led major research projects on translanguaging, not only in various educational contexts including complementary schools for ethnic minority pupils, but also in the community and everyday social contexts (e.g. Reference Creese and Blackledge2015; see also: https://tlang.org.uk/). There are many other applications of the concept in a variety of fields such as: visual arts, music, gender studies, and so forth. A very important contribution of such studies to the translanguaging scholarship is taking the concept beyond language to embrace a wide variety of semiotics and modalities for meaning-making. Language, as in the conventional sense of speech and writing, is but one of many meaning-making resources human beings can deploy. There is now a great deal of discussion of translanguaging creativity and the linguistic and semiotic innovations multilinguals produce through translanguaging (Baynham & Lee, Reference Baynham and Lee2019; Bradley et al., Reference Bradley, Moore, Simpson and Atkinson2018; Zhu & Li, Reference Zhu and Li2022; Li, Reference Li2020). What seems to be clear is that the very term, translanguaging, has caught the imagination of people from very different disciplinary backgrounds with diverse research and professional interests.

At the same time, the concept of translanguaging has raised some very important, indeed fundamental questions about language, bilingualism, and education (Li, Reference Li2018). I want to comment on two of them briefly before I move onto pedagogy. The first comment concerns the nature of language. Building on the extensive and well-established research on the notion of languaging (Becker, Reference Becker and Tannen1988, Reference Becker1991; Thibault, Reference Thibault2020a, Reference Thibault2020b), a key argument in translanguaging scholarship regards language as a meaning-potential. What people do when they make sense and meaning should, therefore, be the main focus of linguistic research. Michael Halliday made that very point in the 1980s in his work on language as social semiotic when he said, ‘Language is a “meaning potential”, and linguistics is the study of how people exchange meanings by languaging’ (Halliday, Reference Halliday, Benson and Greaves1985, p. 193).

Named languages and language labels are socio-politically constructed. Linguists today have inherited named languages from the colonial projects that politically and ideologically constructed languages and gave them names and labels to manipulate them, including categorizing some as dialects and others as languages, and indeed some as standard and others as vernacular, with serious social consequences. These names and labels are being used as tools to categorize individuals, communities, and nations, way beyond describing sounds, words, and sentences.

To illustrate what I mean by named languages are socially constructed, let me retell a story that I was told by a field linguist in China in the 1980s. He and a team of language documentation researchers were tasked to help non-Han minority ethnic groups in remote mountain villages in Yunnan province in the southwest of China to (re)construct their languages by designing writing systems, compiling dictionaries, and writing grammar books. They recorded some chants from the elders of one of the ethnic minority communities and tried to construct a lexicon and the grammatical rules for that language. In analysing the data, this field linguist noticed words and phrases that did not seem to belong to the language that he was studying – some had known origins in other named languages, so they were classified as ‘borrowings’, but others had no known origin, not to this linguist at least. So he asked the community elders, ‘What language is that?’, referring to transcribed words and phrases. The people just looked at him, and said, ‘You tell us; we thought you were the linguist.’ So, the linguist asked, ‘What does it mean?’ ‘Ah, that's a different question,’ the people said. ‘We know exactly what they mean.’ But it was clearly not their job to know and even to care about which language the words and phrases belonged to and what name the language should have. To repeat: named languages are socio-politico-ideological constructs.

That brings me to the second point: bilinguals and multilinguals do not think in a named language, one at a time. They think beyond named languages, and that is one of the very important senses of the notion of trans in translanguaging: transcending the artificial boundaries of named languages. To me, it is inconceivable that when a multilingual person is engaged in producing utterances, like the following, that they are switching between different frames of mind or ways of thinking, that are determined by the named languages:

If we were to follow the ‘we think in the language we speak’ logic, then we think in an individual's idiolect, not a named language as Otheguy et al. (Reference Otheguy, García and Reid2015) pointed out. Indeed, distributed and embodied cognition scholars have argued that thinking does not need (named) language at all. This is, of course, still a controversial point and some of the colleagues working on bilingual cognition and intercultural communication would disagree. But lots of the so-called cognitive advantages of bilingualism are owing to the fact that bilinguals are capable of distancing objects and concepts from the linguistic labels in phonological forms that may be given in different communities to describe the objects or concepts. It seems to me that a potentially very useful benefit of a translanguaging pedagogy is to enhance bilingual learners’ distancing and dissociation capabilities by encouraging them to think beyond named languages. Much of the traditional literature on intercultural communication assumes that speakers of different languages from different cultures have naturally different patterns of thought, causing communication barriers between languages and cultures. The apparent different ways of articulating thought are just that: different ways of articulating thought, from the ‘thinking for speaking’ (Slobin, Reference Slobin, Gumperz and Levinson1996) theoretical standpoint, and they are an outcome of socialization within specific communities. Human beings are very capable of adapting themselves to different ways of expressing their thought in different languages so long as the social conditions allow. But that assumes a universal base for human thought; otherwise, people will never be able to learn to understand each other. I am making these points because they are directly relevant to the idea of translanguaging as a pedagogical practice, which to me must go beyond named language and the use of multiple named languages.

Linking named languages and thinking together is always a tricky business. The strong form of the linguistic relativity hypothesis, otherwise known as linguistic determinism, that suggests that linguistic categories and structures limit and determine cognitive categories and processes, can also lead to erroneous and dangerous assumptions that speakers of some languages are cognitively superior than others. Considering the above, it is crucial to repeat that when I say that named languages are a by-product of ongoing languaging activities, and that bilinguals do not think unilingually or indeed only linguistically, we do not say or mean that named languages do not exist; nor that bilinguals and multilinguals are not aware of the differences between named languages. The strawman argument that translanguaging scholars uphold that named languages do not exist is very problematic, and it appears to me that we have had enough opportunities to educate ourselves on this key issue. That named languages exist is a truism. Our emphasis, though, is on named languages as political constructs, and the naming of sets of linguistic features as language, or dialect, or something else has serious social, political, and of course, educational consequences – wars have been fought and blood has been shed over the naming and categorization of languages. Bilinguals and multilinguals are absolutely aware of the socially constructed differences between named languages because the development of language awareness is an integral part of one's socialization experience. It is a crucial part of bilinguals’ sociolinguistic sensitivities. Translanguaging is about transcending and breaking such boundaries and differences.

Drawing upon the above argument, I now move onto translanguaging as a pedagogy. Translanguaging pedagogy can help to maximize the bilingual learner's learning potential by making good use of their awareness and knowledge of different named languages, and their sociolinguistic and socio-political sensitivities, or Critical Language Awareness as elaborated by some (Clark et al., Reference Clark, Fairclough, Ivanič and Martin-Jones1990). To advance the idea of translanguaging pedagogy, in the remainder of this article, I focus on two aspects: co-learning and transpositioning.

3. Co-learning

As said earlier, to use translanguaging as pedagogy for inclusion and social justice requires a different mindset on behalf of the teachers, and a different discourse from researchers and practitioners alike. For me, that different mindset is best captured through the notion of co-learning.

I first wrote about co-learning in 2014 in the context of heritage language education in complementary schools in the UK, which has been my main research base in bilingual education (Li, Reference Li and May2014a). It was inspired by Edward Brantmeier's work. The concept of co-learning came not from education but from other disciplines ranging from artificial intelligence and computer simulation to global security systems and business information management. In essence, co-learning is a process in which several agents simultaneously try to adapt to one another's behaviour so as to produce desirable global outcomes that would be shared by the contributing agents. In the classroom context, co-learning changes the role sets of teachers and learners from what Brantmeier calls (Reference Brantmeier, Lin, Oxford and Brantmeier2013, p. 97) ‘dispensers and receptacles of knowledge' to ‘joint sojourners’ on the quest for knowledge, understanding, and wisdom. In particular, the teacher would become ‘a learning facilitator’, ‘a scaffolder’, and ‘a critical reflection enhancer’, while the learner becomes ‘an empowered explorer’, ‘a meaning maker’, and ‘a responsible knowledge constructor’.

In Brantmeier's words, a facilitator ‘guides the process of student learning’ and ‘does not get in the way of learning by imposing information’. A scaffolder ‘assesses the learner's knowledge and builds scaffolding to extend that knowledge to a broader and deeper understanding’. And a critical reflection enhancer asks the learner to ‘reflect on what is being learned and the process of learning (meta-reflection about process)’. At the same time, an empowered explorer is ‘an independent or collective explorer of knowledge through disciplined means’. And a meaning maker and responsible knowledge constructor is ‘one who engages in meaningful knowledge construction that promotes relevancy to their own life’. In co-learning, mutual adaptation of behaviour is the key. In order to achieve desirable learning outcomes, all co-participants need to constantly monitor and adapt their actions and learn from each other. Brantmeier was writing about and to those of us in the teaching profession.

It is very important to emphasize that co-learning in the classroom does not simply involve the teacher's developing strategies to allow equitable participation for all; co-learning requires much unlearning of cultural conditioning because, as Brantmeier points out, ‘it challenges the traditional authoritative, dominant and subordinate role sets in schooling environments and the unequal power relationships in wider spheres of our world’ (Brantmeier, Reference Brantmeiern.d., n.p.). It aims to build a genuine community of practice, by overcoming the teacher and the learner dichotomy or divide, moving towards a more ‘dynamic and participatory engagement’ in knowledge construction. According to Brantmeier (Reference Brantmeier, Lin, Oxford and Brantmeier2013), the characteristics of a co-learning relationship include: (i) all knowledge is valued; (ii) reciprocal value of knowledge sharers; (iii) care for each other as people and co-learners; (iv) trust; and (v) learning from one another. Additionally, the characteristics of a co-learning classroom environment include: (i) shared power among co-learners; (ii) social and individualized learning; (iii) collective and individual meaning making and identity exploration; (iv) community of practice with situated learning; and (v) real-world engagement and action.

Translanguaging and co-learning share a great deal of common values. In particular, both translanguaging and co-learning emphasize that all knowledge, acquired through all languages and in all cultural contexts, should be valued. For instance, refugee children have acquired knowledge through languages other than English before they ended up in the UK and entered schools and classrooms in the country, just like some children from socially disadvantaged backgrounds will have acquired knowledge in their own communities through what are classified as non-standard English. In an inclusive classroom, co-learning is to show reciprocal value of knowledge sharers, and care for each other as people and as co-learners. Their mutual trust is essential in order to learn from one another. In what he describes as ‘Pedagogy of Vulnerability’, Brantmeier (Reference Brantmeier, Lin, Oxford and Brantmeier2013) urges the instructors to open themselves up and contextualize that self or selves in societal constructs and systems. For him, co-learning entails admitting one does not know everything and be human. It also entails taking risks: risks of self-disclosure, of change, of not knowing, of failing. Whilst translanguaging focuses on respecting all languages and promotes the learning of each other's languages and perspectives on the world, co-learning advocates respecting the knowledge, values, and insights of all involved. Both advocate creating opportunities of different ways of learning and different ways of talking about learning, and both point to the need to unlearn cultural conditioning and to dismantle asymmetrical power relationships.

In Britain, it is commonplace for language classes, and indeed all classes, to have learners with very different linguistic, cultural, and educational backgrounds, some having very complex migration and language-learning experiences. They bring with them ‘funds of knowledge’ – ‘the historically accumulated and culturally developed bodies of knowledge and skills essential for households and individual functioning and well-being’ (Moll et al., Reference Moll, Amanti, Neff and Gonzalez1992, p. 133; see also Moll & Gonzalez, Reference Moll and Gonzalez1994). Such funds of knowledge contain rich cultural and cognitive resources that can be used in the classroom in order to provide culturally responsive, meaningful, and effective teaching. Teachers, as well as the learners, have much to gain from using these funds of knowledge in the classroom as co-learners, not only to make the classrooms more inclusive, but also to engage in real-world meaning making and identity exploration, which are crucial yet often neglected aspects of learning.

Let us look at an example (from Li, Reference Li2014b, p. 167):

• • T: 曲奇(quqi). 一种饼干，知道吗？

• • Quqi. A kind of cookie, you know?

• • G1: What?

• • T: 曲奇 (quqi).

• • G2: kuk-kei. kuk-kei.

• • T: Yes.

• • G1: So why did you say qiu…something qiu…

• • T: 曲奇(quqi).

• • G2: No. kuk-kei.

• • G1: 广东话是 kuk-kei。

• • In Cantonese it is kuk-kei.

• • G2: It's Cantonese. kuk-kei is Cantonese.

• • T: 是吗？

• • Is it?

• • G2: 係啦！

• • Yes.

As discussed in Li (Reference Li2014b, pp. 167–8), this was recorded in a Mandarin class in a Chinese complementary school in Manchester, UK. The children are mainly British-born Chinese of Cantonese-speaking parents. The Mandarin teacher is from mainland China and does not speak Cantonese. She mentions a term for biscuits or cookies in Mandarin, 曲奇 (quqi). It is, in fact, a Cantonese transliteration of the English word cookie, but the teacher does not seem to know this historical sociolinguistic fact. In Cantonese it is pronounced as kuk-kei, which sounds similar to the English word cookie. But in Mandarin, it is pronounced as quqi, which has nothing to do with the English pronunciation, but it is written with the same two Chinese characters. This is where the fun starts. The pupils from Cantonese-speaking families know what it means and where the term came from, but the teacher actually does not – she knows the meaning, as the textbook tells her, but she does not know the origin because she does not know Cantonese. And the whole social, colonial history behind this simple term is absolutely fascinating, except that the teacher does not know it. If the pupils did not query it and if they did not collectively contribute to the discussion – the sharing of knowledge that is clearly gained outside the classroom, then a great deal of knowledge and understanding will have got lost, a real missed opportunity.

Of course, the transcript of this short episode of classroom interaction merely points to the need for co-learning and how co-learning could take place in class. The teacher did tell us in our post-observation conversations with her that she found the exchange extremely interesting and checked out the history of the term and how it got transliterated herself. More importantly, she learned that she needed to give the pupils more opportunities to teach her what they know of the Chinese language from their point of view and she ended up actively pursuing a comparison between Mandarin and Cantonese in her own postgraduate studies.

4. Transformative translanguaging pedagogy for inclusion and social justice: Two examples

So, what does a translanguaging pedagogy entail in practice? I explore this question by recounting the experiences of two teachers with whom I have worked in the last couple of years to show how they transformed their pedagogical practices as well as their subjectivities and positioning through co-learning. Both teachers are based in state secondary schools in London. One is an EAL teacher (teacher of pupils with English as an additional language) and the other is a MFL (Modern Foreign Language) teacher teaching Mandarin and German.