Original Research

Multilingual literacies for learning: Shifting ideologies, developing praxis

Margaret J. Probyn

Received: 30 Apr. 2024; Accepted: 29 Aug. 2024; Published: 22 Nov. 2024

Copyright: © 2024. The Author Licensee: AOSIS.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

Background: In South Africa the majority of learners are failed by school language policies that require them to switch from learning through the medium of their home language, to learning through the medium of English from Grade 4. Yet language ideologies rooted in coloniality sustain the domination of English in education and suppress the home languages of teachers and learners as valid resources for learning.

Objectives: The context of this article is a teacher development intervention – Languaging for Learning (L4L) – which aimed to challenge prevailing language ideologies and develop bi/multilingual pedagogies based on the recognition of the learners’ home languages as important resources for learning.

Method: Science, Mathematics and English teachers from 10 township schools in the Cape Town Metro participated in the 18-month programme which included workshops, classroom-based support and the provision of some bilingual materials. Data on the teachers’ uptake of the programme were collected through observation, interviews, and video and photographic records.

Results: This article describes the shifts in language ideologies and practices of five of the participating teachers. The introduction of bi/multilingual pedagogies impacted positively on learners’ participation, motivation and opportunities to learn as well as on teachers’ sense of efficacy and agency.

Conclusion: The research demonstrated how bi/multilingual pedagogies that value learners’ and teachers’ linguistic repertoires as resources for learning can contribute to overcoming the long-standing barriers to learning resulting from language policies and practices that are rooted in coloniality.

Contribution: These findings demonstrate how bi/multilingual pedagogies can support epistemic access and biliteracy in education.

Keywords: multilingual learning; translanguaging; language ideologies; South Africa; teacher development.

Introduction

South Africa is one of many countries in the Global South where language ideologies rooted in colonialism continue to sustain the status and power of English over indigenous African languages in all spheres of life, including in education (McKinney 2017). The right for learners to learn through any of the 12 official languages is enshrined in the South African Constitution (Republic of South Africa 1996b) and in the Language-in-Education Policy (Republic of South Africa 1997) and yet school governing bodies continue to opt for language policies that favour English over indigenous African languages, perpetuating the ‘linguistically structured inequalities’ (Kerfoot & Bello-Nonjengele 2022) of colonialism and the apartheid era.

This is starkly evident in township and rural schools where the majority of teachers and learners communicate in African languages at home and in their communities; yet school language policies require learners to switch to English as the language of learning, teaching and assessment (LOLTA) from Grade 4, despite the very obvious barriers to learning that this creates – what Macdonald (1990c) described as ‘swimming up the waterfall’ with the resultant harm to learners’ self-confidence and identity:

[P]utting … children through a painful experience by making them learn through the medium of English when they were not adequately prepared to do so; … giving them the experience of failure, not only in terms of marks, but also in terms of coming to grips with important concepts … [so that] children are alienated from the process of schooling from an early age. (Macdonald 1990a:17; 1990b:47)

For learners in these schools, failure to learn is normalised and their home languages are suppressed as potential resources for learning. This runs counter to the expressed intentions of the South African Schools Act (Republic of South Africa 1996) to provide education based on the principles of access, equity, redress, and social justice.

Teachers in these contexts are faced with the conflicted situation that if they stick to the English-only policy, learners are unable to understand lesson content; but if they switch to the learners’ home languages to communicate lesson content, then they run the risk of official censure (Macdonald 1990b; Setati et al. 2002; Probyn 2009). Too often the poor performance of learners in national and international assessments is described in deficit terms: teachers and learners are blamed for what is in fact systemic failure to provide learners with opportunities to freely access and engage with the curriculum.

While the National Department of Education’s decision to introduce ‘Mother Tongue Based Bilingual Education (MTBBE)’ from 2025 (Republic of South Africa 2024) is welcomed as a major breakthrough, at the time of the research there were still mixed messages from education officials and school management, with many expressing disapproval for any use of African languages for learning content subjects beyond the Foundation Phase.

The research described in this article is based on an in-service teacher intervention, Languaging for Learning (L4L), that set out to challenge current language ideologies in education and to work with teachers to develop bi/multilingual pedagogies that embrace learners’ full linguistic repertoires as important resources for learning across the curriculum and in tandem develop their proficiency in English.

Conceptual framework

The L4L programme was informed by the following conceptual framework: socio-cultural theories of language and learning, translanguaging as a theory of language and as pedagogic practice, and the notion of Anglonormativity as an ideology rooted in coloniality.

Socio-cultural theories of language and learning, based on the work of Lev Vygotsky (1962), put engagement with learning through language at the centre of education. Plainly, children cannot learn without opportunities to engage with texts and more knowledgeable others in co-constructing knowledge through language they understand. Accordingly, the most valuable resource that children bring to school is their most familiar languages (McKinney 2017:xv). Barnes’s (1992) distinction between exploratory and presentational classroom talk and writing provides teachers with useful ways to think about the different functions of language in learning and how these can be woven into bi/multilingual teaching and learning. The L4L programme drew on socio-cultural orientations to research and to teacher development that recognise the importance of reflection on practice and the role that classroom-based mentors can play in supporting teachers and bridging the theory-practice divide (Calderhead & Gates 1993).

The term ‘translanguaging’ was coined by Baker (2001) to describe a bilingual pedagogy first described by Williams (1994) where English and Welsh were alternated as languages of input and output in order to develop learners’ biliteracy and to deepen learning. The pedagogical term has been expanded by sociolinguists and psycholinguists to describe languaging practices in multilingual contexts beyond the classroom (see Lewis, Jones & Baker 2012). It marks a shift away from viewing languages as existing in strictly compartmentalised ‘solitudes’ (Cummins 2008), to a view that the boundaries between named languages are naturally flexible and porous, and that multilingual individuals draw on their full range of linguistic resources to make meaning, according to the demands of their social context (Garcia 2009).

This article however is concerned primarily with translanguaging as pedagogic practice, described by Baker (2011) as:

the process of making meaning, shaping experiences, understandings and knowledge though the use of two languages. Both languages are used in an integrated and coherent way to organize and mediate mental processes in learning. (p. 288)

South African linguists and education researchers have pointed out that much of the translanguaging theory considered novel in the Global North is in fact reflective of pre-colonial languaging practices in Africa, where the interdependence and fluidity of languages – what Makalela (2016) has described as ‘ubuntu translanguaging’ – was disrupted by colonisation and the artificial fixing of language boundaries by missionaries and other colonial agents. So too was the setting of linguistic hierarchies with indigenous languages in subservient relations of power to colonial languages – mirroring the political relations of power of colonisers and colonised (Heugh 2014). Accordingly, translanguaging pedagogy can be enacted as a critical pedagogy that exposes ‘the injustice of forcing bilingual students to perform academically with less than half of their full linguistic repertoire’ (Vogel & García 2017:7–8) while ‘disrupting the hierarchy of languages, transforming both teachers and students’ attitudes towards their diverse meaning making resources, and enabling students’ full participation in knowledge co-making’ (Li & Lin, 2019: 212).

More than three decades of research (Macdonald 1990; McKinney 2017; Setati et al. 2002) have documented the negative impact that the sudden switch in Grade 4 from home language to English medium of instruction has on opportunities for meaningful learning for the majority of learners in South Africa. Nonetheless, school governing bodies continue to opt for English LOLTA from Grade 4. These seemingly self-defeating choices can be best understood in terms of deeply held language ideologies, rooted in coloniality which Maldanado Maldonado-Torres (2007) describes as:

long-standing patterns of power that emerged as a result of colonialism, but that define culture, labor, intersubjective relations, and knowledge production well beyond the strict limits of colonial administrations. (p. 243)

In line with this view, McKinney (2017) has coined the term ‘Anglonormativity’ to account for the deeply entrenched dominance of English in society in general, and in education in particular. McKinney (2017:80) describes as ‘the expectation that people will be and should be proficient in English, and are deficient, even deviant, if they are not’ (p. 80). Thus English, at the apex of the linguistic hierarchy, is the language of aspiration and the taken-for-granted language of education.

In addition, school language policy choices are driven by the instrumental need for learners to gain the benefits of proficiency in English and the perception that this must be achieved through ‘time on task’, by learning through the medium of English. This is, despite research, that has demonstrated that learners in bilingual programmes achieve better results in English as well as content subjects (Bambgose 1984; Thomas & Collier 2002).

While many teachers have been socialised into the prevailing Anglonormative ideologies, these are at odds with classroom realities and and therefore in practice, often resort to covert code-switching (Macdonald 1990a; Setati et al. 2002). Any intervention that aims to address language barriers in education needs to take account of teachers’ existing language ideologies as a starting point for developing multilingual pedagogies that value and engage with learners’ most familiar languages as important resources for learning (Bernstein et al. 2023).

The L4L programme drew on a view of classroom translanguaging as a planned and strategic but also flexible pedagogy that includes both ‘fluidity and fixity’ in oral and written modes, that employs strategies such as bi-directional translation for deepening understanding and that seeks to redress the systemic injustice of current Anglonormative language policies and practices.

Pedagogical translanguaging in South Africa: Recent research

There is a rich history of research in South Africa on classroom code-switching as this is common practice in classrooms where there is a mismatch between the learners’ most familiar language and the LOLTA, most often English (Setati et al. 2002). More recently such practices have been reframed as ‘translanguaging’, following the shift in understandings of bilingualism and conceptions of languaging in the literature. Much of the research on translanguaging in classrooms relates to the spontaneous oral responses of teachers to the imposition of language policies that obstruct learning. There is a small but growing body of literature on translanguaging as a planned, purposeful, bi/multilingual pedagogy that could inform the development of the L4L programme. Some examples are briefly reviewed.

Makalela (2015) describes a planned translanguaging intervention to improve reading comprehension in a Grade 6 Sepedi/English language class. The translanguaging strategies included bilingual vocabulary contrasts; alternating the language of input and output by reading a text in one language and answering questions on it in the other language; reading texts aloud in alternate languages; writing a text in Sepedi and rewriting it in English; reading the same text written in English and Sepedi; and a bilingual publication corner where learners’ bilingual texts were on display. A pre-test and post-test showed an improvement in reading comprehension in both languages.

Guzula, McKinney and Tyler (2016) describe an after-school literacy club where planned translanguaging strategies around oral and written texts were used to develop isiXhosa-English biliteracy. For example, a story told and enacted orally in English was discussed mainly in isiXhosa, then key terms for a narrative genre were elicited from learners in either language and written up for display with translations; the facilitator elicited ‘story starters’ in either language and wrote them down along with translations – as preparation for the learners to write their own stories in either language. The facilitator built on the learners’ linguistic repertoires flexibly to support their biliteracy development.

Charamba (2019) describes a planned translanguaging intervention in a Grade 11 Physics class where the dominant home language was Sesotho: the teacher taught the science content using English and Sesotho; learners were encouraged to ask and answer questions in whichever language they preferred; learners were provided with multilingual learning materials including parallel texts, transliterated texts and multilingual science dictionaries; and learners were given written tasks to do in English and Sesotho. In a post-test learners in the intervention group outperformed learners in a control group.

Tyler (2023) described a translanguaging translation activity with an after-class Grade 9 science study group of isiXhosa-English bilingual learners: the group were provided with isiXhosa explanatory definitions for concepts they had encountered in their English medium science classes and asked to translate these into written English and afterwards into the local urban variety of isiXhosa. This exercise involved learners in debate and contestation to refine their definitions, deepening their understanding, positioning them as resourceful and agentic, and engaging them in critical metalinguistic talk about language use and science learning.

The most extensive and sustained planned translanguaging intervention has been the Department of Basic Education’s MTBBE pilot in the Eastern Cape province which was initiated with 70 schools in the Cofimvaba District in 2011 and in 2016 extended to all districts in the province. From Grade 4 to Grade 7 learners were taught half the curriculum in isiXhosa or Sesotho (Mathematics and Natural Sciences) and half in English (Social Studies and Life Orientation). Grade 6 learners in MTBBE schools outperformed non-MTBBE schools with 53% average versus 40% in Mathematics and 78% versus 50% in Natural Sciences. This has provided the National Department of Education with the necessary evidence to back the introduction of MTBBE nationally starting with Grade 4 in 2025 (Republic of South Africa, 2024). However, to date there has been little published about the content and process of teacher training that took place or the translanguaging pedagogies that were developed.

Thus far there is not a large body of literature documenting details of planned translanguaging pedagogies that have worked for teachers and learners, nor details of the content and processes of teacher development programmes to support such initiatives. Nonetheless, the L4L programme was able to glean some valuable ideas from research projects outlined.

Languaging for learning: Project outline

The L4L project was funded by the Zenex Foundation and developed by a team of university-based researchers and teacher educators. It was informed by the theoretical framework and research as outlined.

Key principles that underpinned the design of the L4L programme were:

• The recognition that South Africa is a multilingual country with fluid multilingualism or ‘ubuntu translanguaging’ (Makalela 2015) as the norm.
• Understanding the role of coloniality and Anglonormative language ideologies (McKinney 2017) in the seemingly counter-productive language policy choices made in schools.
• A view of bi/multilingualism in classrooms that embraces and develops learners’ full linguistic repertoires as resources for learning. This includes moving away from the binaries of arguing for either English or the learners’ home languages as LOLTA.
• Translanguaging in classrooms is viewed as a planned, purposeful, strategic, and flexible pedagogy that operates on a continuum of fluidity and fixity in order to optimise learners’ opportunities for epistemic access and bi/multiliteracy development.
• Teachers’ existing multilingual skills that have been ignored and/or suppressed should be legitimated as a starting point for developing appropriate translanguaging pedagogies across the mode continuum.
• Reflection on practice is an integral part of the programme, for both the research team and for the participating teachers.
• Teachers are respected as knowledgeable professionals and partners in developing new pedagogies.

The L4L programme ran from April 2022 to November 2023 for Grade 8 and Grade 9 English, Natural Sciences and Mathematics teachers, from 10 ‘township’ schools in the Western Cape province, South Africa. The LOLTA in the schools was English, and isiXhosa was the home language of the majority of teachers and learners.

The L4L programme focused on three key areas:

• The development of translanguaging pedagogies that leverage teachers’ and learners’ bi/multilingual resources.
• Text-based approaches to teaching language across the curriculum.
• Independent reading, supported by classroom libraries, to develop literacy skills in English and in isiXhosa.

This article focuses on the development of translanguaging pedagogies and the process of change experienced by teachers.

The key elements of the intervention were workshops for teachers combined with classroom-based support by mentors drawn from the L4L team members. Teachers attended a series of foundational workshops over a period of seven weeks in Term 2 of 2022. These aimed to surface and unsettle teachers’ existing language ideologies and to introduce key concepts and strategies for developing bilingual language proficiency and literacy across the curriculum. The workshops created reflective spaces for teachers to share ideas and report back on translanguaging strategies they had tried out in class.

The foundational workshops were followed by two to three practice-based workshops per term, where teachers and mentors planned lessons and bilingual teaching materials for the forthcoming weeks, closely aligned to the official curriculum (see Figure 1). Teachers of all the Grade 8 classes, received multilingual dictionaries and classroom libraries for natural sciences and mathematics. The experiential activities in workshops involved teachers in translanguaging and demonstrated how discussions around written translation of concepts could deepen their own understanding.

FIGURE 1: Teachers discussing translation of concepts to develop bilingual classroom posters.

From the third term of 2022 through to the end of the third term in 2023, mentors drawn from the L4L team visited teachers on a weekly basis to provide classroom-based support and mentoring. This support took the form of help with lesson planning and materials, classroom observation and feedback, modelling teaching and co-teaching. Mentors kept in close touch with teachers through WhatsApp groups and emails.

The workshops provided teachers with the knowledge and support to engage openly with translanguaging in their classrooms. The positive responses of learners and the increase in their participation in class provided a positive feedback loop and encouraged teachers to expand their translanguaging practices. The development of bilingual materials in workshops provided teachers with a model for breaking through the ‘text frontier’ (Tyler 2023) and how to work with learners in written translanguaging. The classroom support was an important factor contributing to the uptake of the programme by teachers and the development and consolidation of their classroom practice.

Research design

The L4L intervention included a research component so that data on the progress and the impact of the intervention were systematically and ethically collected, analysed and disseminated. The research approach was qualitative and interpretive. It followed an educational design-based research approach, the purpose of which is (Plomp 2010):

to design/develop an intervention (such as programmes, teaching-learning strategies, materials, products and systems) with the aim to solve a complex educational problem and to advance our knowledge about the characteristics of these interventions and the processes to design and develop them. (p. 12)

It is an approach that:

speak[s] directly to the problems of practice and that lead[s] to the development of ‘usable knowledge’ and follows a process of planning informed by prior research and review of relevant literature, implementation, reflection and adaptation, with close collaboration between practitioners and researchers. (p. 9)

The research team met on a weekly basis to plan, report and reflect on workshop engagements, classroom support and the research process. Planning, reflection and sharing of practice by teachers was an integral part of the workshops; further opportunities for teachers’ and mentors’ reflections were during classroom support visits, and emails and WhatsApp messages provided online platforms for sharing successes, information and opinions. In this way both the research team and teachers were engaged in cycles of planning, reflection, adaptation in the process of exploring and developing bi/multilingual pedagogies.

Ethical clearance was obtained from the University of Western Cape Ethics Committee (Ref. HS22/2/10), based on the principles of informed consent by participants, with participants reserving the right to withdraw at any point; protection of the identity of participants; and respect for the experience and opinions of participants as collaborators in the research. Permission for the research was obtained from the Western Cape Education Department.

The guiding research question was: In what ways does the ‘Languaging for Learning’ teacher development programme support improved opportunities to learn English, Science and Mathematics?

The sub-questions addressed in this article are:

• What are the changes in teachers’ perceptions of language and learning in multilingual classrooms?
• What are the changes in teachers’ classroom practices when teaching English, Science and Mathematics in multilingual classrooms?
• In what ways did the changes introduced in classroom teaching and learning impact on learners’ opportunities to learn English, Science and Mathematics?

The collection of data was exploratory and open-ended and took place over the course of the intervention, during workshops, discussions, and classroom support. The main sources of data that form the basis of this article are focus group discussions with teachers, which took place in October 2022 and October 2023, observation data from workshops and classrooms including photographs, and documents such as learners’ workbooks and posters created by teachers and learners. The selection of teachers to participate in the focus group discussions was purposive as they were invited to participate on the basis that they had been among those most enthusiastically engaged in developing translanguaging pedagogies with their learners.

Of the teachers who participated in the focus group discussions (see Table 1), Ms Dyasi and Ms Lekena were both from School A, while Ms Bizwaphi and Ms Cala were from two different schools. All three schools were in township areas where the official medium of instruction was English while the home language of the majority of learners was isiXhosa. Three of the teachers identified as isiXhosa speakers whereas Ms Lekena claimed that while she was orally proficient in isiXhosa, her most familiar language was Sesotho and therefore she did not know all the isiXhosa translations for English science terms.

The focus group discussions were recorded, transcribed and translated where necessary. The transcriptions were coded and analysed according to broad themes in order to respond to the research questions.

The teachers’ responses have been extensively quoted to preserve a sense of their agency and voice.

Research findings

Surfacing and engaging with language ideologies in workshops

Teachers’ practices are underpinned by their ideologies (Bernstein et al. 2023); so the starting point for the L4L programme was to create opportunities for teachers to surface and question the ideologies underpinning school language policies and their own language ideologies and to recognise how these shaped their classroom practices. This was an important process prior to introducing bi/multilingual approaches to teaching and learning.

In the first workshop teachers were shown the documentary ‘Sink or Swim’ (Project for Alternative Education in South Africa, 2004)¹, that advocates mother tongue-based bilingual education and demonstrates the difficulties learners have when switching to learning through the medium of English after only a few years of learning English as a subject.

In the discussions following the viewing of the documentary, teachers unpacked some of their own ideological positions. Some of their responses to this documentary are shown in Figure 2 and highlighted as follows:

‘We need to unlearn that English is superior’.

‘It feels like things haven’t changed since the end of apartheid’.

‘Lack of competence in a language seems to be an indicator of intelligence: high English marks equals intelligence. This should be changed’.

‘Abazali bafuna isiNgesi [parents want English] for their children’.

‘All languages are mixed [words are borrowed]’.

FIGURE 2: Initial discussions by teachers in response to viewing Sink or Swim.

They spoke to the challenges teachers and learners faced (see Figure 2):

‘Every teacher is required to teach in English’.

‘Being taught in a foreign language makes children lose interest in school’.

‘Not understanding the language leads to disruptive behaviour’.

‘Abafundi abavi iEnglish and hide rather than ngena eklasini [children don’t understand English and hide rather than go to class]’.

A reading of the Language-in-Education Policy (LiEP) (1997) showed teachers that, contrary to popular belief, the policy does not require learners to switch to English LOLTA in Grade 4; instead it recognises multilingualism as an asset and encourages additive multilingualism in classrooms.

Many teachers expressed excitement at the prospect of opening up their classrooms to translanguaging (see Figure 2):

‘Code-switching is exciting’.

‘Understanding is easier in home language’.

Others expressed some concerns (see Figure 2):

‘Teachers have two jobs – teach science and translate into isiXhosa’.

‘Translation is complicated’.

‘Learners don’t speak ‘proper’ or ‘pure’ isiXhosa’.

‘What about words in different dialects? Can we accept these?’

The visual images and the persuasive commentary in the video revealed to teachers the powerful ideologies that continue to shape language-in-education policy choices. This provoked debate that enabled them to surface and question their own ideological positions as a starting point for initiating change in their own practices.

Expanded reflections on language ideologies prior to the languaging for learning programme

The teachers in the focus group discussions were asked about their attitudes towards the use of English as the official LOLTA, and how this played out in their classroom practices, prior to and after the L4L intervention. They were able to articulate their beliefs with a self-awareness that might well not have been possible without the initial workshop experiences.

The teachers’ beliefs and practices prior to the L4L programme were not uniform: two teachers, Ms Cala (English) and Ms Lekena (Natural Sciences), both believed strongly that it was in the best interests of the learners to teach only through the medium English, but for different reasons. Ms Cala as an English teacher expressed ideas consistent with monolingual views of languages as bounded entities and felt that any contact between English and isiXhosa would cause confusion and contamination of ‘pure’ English: learners would ‘confuse the rules’ and ‘start speaking Xhonglish [a mixture of isiXhosa and English]’. Her teaching methods were consistent with theories of second language acquisition (Krashen & Terrell 1983): learners should be immersed in English and if they were unable to answer questions in English this was because they were ‘in the silent stage of language acquisition’; so she would ‘continuously rephrase’ her questions in English so that they would ‘eventually understand’. The backwash effect of examinations in English and her feelings of professional responsibility were expressed:

‘I was sticking strictly to English. … I was not code-switching at all otherwise I would be short-changing learners … because I was paid to teach in English. The children need to learn to express themselves in English as the exams are in English.’ (Ms Cala)

Ms Lekena’s practice of teaching natural sciences only in English was driven in part by her own limited command of isiXhosa, as her home language was Sesotho. Like Ms Cala, Ms Lekena also felt that teaching in English was in the best interest of the learners because their textbooks and examinations were in English:

‘If you are using their home language are they going to perform? Because in the paper I am going to ask them in English so are they going to be able to respond?’ (Ms Lekena)

The other two teachers both reflected that they would try to stick to their school policy and teach in English, but when they saw that learners did not understand, they would code-switch to isiXhosa. However, this was done ‘secretively’ as teaching in isiXhosa was ‘not legal’ and ‘not allowed’ by curriculum advisers and school management (Ms Bizwaphi). This was echoed by Ms Dzina:

‘It was not allowed. Even our curriculum advisers were discouraging that. Even the SMTs were discouraged that, you were not allowed to do that.’

As a result the translanguaging of these teachers was limited:

‘I was the only one code-switching … they [learners] were not allowed to speak or communicate with me in isiXhosa. … I only switched to their language when I see now the room is tense, there is no learning happening … if I see they are confused then I quickly code-switch then I go back to English.’ (Ms Bizwaphi)

Two of the teachers recognised that they had assumed that learners who were more fluent in English were more academically able and conversely those who were less proficient in English were labelled as slow:

‘The children that were more fluent in English I thought of as the brightest in class, the clever ones.’ (Ms Cala)

‘I gave up on those learners [who were not proficient in English] and said ok those learners are going to be pushed to Grade 9. … I would believe that a black child is not sometimes good for learning. There are some learners that are struggling in learning, maybe they [should] work … with [their] hands.’ (Ms Lekena)

The teachers’ views on the role of English in learning across the curriculum prior to their participation in the L4L programme revealed the deep influence of Anglonormative ideologies (McKinney 2017) in their acceptance of the role of English as LOLTA – despite the obvious challenges it posed for learners and the frustrations it caused them as teachers. The learners’ struggle to learn through English was normalised and those who were less proficient in English were regarded as less intelligent or even unteachable. Two of the teachers were prepared to subvert the language policy by ‘secretively’ using the learners’ home language to clarify misunderstandings but they did this minimally, as a measure of last resort, conscious that they were risking official disapproval and censure.

Challenges of teaching through English

The teachers spoke at length about the challenges they and their learners faced when they tried to teach only through the medium of English. These included that teachers found learners did not respond: ‘They just look at me as if I am talking to myself. They don’t respond’ (Ms Bizwaphi). ‘[I] was preaching every time and you find that learners are asking to go to the loo. Sometimes they just sleep. We see that learners are getting bored’ (Ms Lekena).

Ms Bizwaphi expressed her frustration:

‘You don’t know if they are quiet because they don’t understand, or quiet because they are cheeky, are they quiet because they just ignoring you or they didn’t want maths at all.’ (Ms Bizwaphi)

Ms Cala revealed that in English lessons some learners would prefer to get zero marks rather than speak English, ‘You can give me zero Miss, I don’t want to do it’.

Ms Lekena noted that many learners were not able to understand the questions in English assessments and some would respond by just copying out the question on their answer sheets, to avoid submitting blank pages. As a result she would coach learners to answer test questions, encourage rote-learning, while ‘leaving [out] that part of understanding [and] creativeness’.

Teachers’ reflections on their practices revealed the tensions they experienced as a result of the dissonance between language policies and the classroom realities that limited possibilities for learners to actively engage in co-constructing knowledge through languaging and instead pushed them into passive rote-learning.

Shifting ideologies and developing practices

The teachers’ discussions and observed classroom practices provided fine-grained evidence of the shifts in their language ideologies and practices once they became involved in the L4L programme.

The workshops provided teachers with the necessary rationale and authority to engage freely in translanguaging:

‘I actually went back to my learners, then I said “Listen guys, I went to a workshop and it was said we can use our home language to do the learning, to learn with, to explain some difficult mathematical concepts.” Now it makes it official to them. Ja now after the workshop I teach with confidence … I took action with something that has been researched.’ (Ms Bizwaphi)

Three of the teachers confessed that they had had initial reservations about changing their practices:

‘I was not fully sold on the idea … but I was interested in trying it out.’ (Ms Cala)

‘I was asking that question, “Is it going to work?” So I was scared … But then I told myself, “I will try it and see”.’ (Ms Lekena)

‘I have, myself, an attitude, saying that how are we going to do this because we don’t have that much vocab for our content subjects?’ (Ms Dzina)

However, once they took the plunge and tried out some translanguaging strategies, the positive responses of learners led teachers to revise their ideologies and change their practices:

‘But when I saw the transformation that it did, yoh, it was very impressive. I totally changed … I’ve seen a huge difference. … Like the quiet kids, the ones who didn’t participate in class, started participating. … My results were better because … everyone is more confident and everyone felt like “I can”.’ (Ms Cala)

‘So the way they responded to me that’s what changed me but especially when it comes to results, wow!’ (Ms Lekena)

‘It is the participation I have always been longing for in my class I am getting. The positive feedback from my learners. … I am able to see their capabilities now … and actually grow what they already know unlike before when I did not even know where they are.’ (Ms Bizwaphi)

‘So I think to allow them to express themselves multilingually, it opens a way for them to express themselves more in mathematics. Before, they were so passive in my classroom. Now they are so active and engage very well.’ (Ms Dzina)

The foundational workshops provided teachers with the knowledge and confidence to openly initiate oral translanguaging in their classrooms. The positive responses of learners along with the encouragement of mentors and opportunities for reflection during workshops reinforced teachers’ ideological repositioning in a cycle of change. This process appeared to support quite radical ideological changes in teachers: from unquestioning acceptance of the hegemony of English in education, and the assumption that English proficiency was a marker of intelligence, to a realisation that learners’ home languages were an important resource for learning that could transform their classrooms.

Developing practices: From oral to written translanguaging

Initially the changes in teachers’ practices were focused on oral translanguaging and either introducing this as a new practice or expanding their existing covert code-switching into more open and planned bilingual approaches. Classroom talk was informed by Barnes’s (1992) distinction between exploratory talk and presentational talk, with exploratory talk or thinking aloud done freely in whatever languages learners found easiest and presentational talk very often done bilingually. Teachers’ and learners’ languaging ranged along a continuum of fluidity and fixity depending on the context and task at hand.

In the practice-based workshops teachers engaged in translation activities to develop multilingual teaching materials such as glossaries and concept posters in line with curriculum topics they were going to be teaching in the weeks that followed. Teachers noted that working in groups and the process of translation and contestation around precise meanings deepened their own understanding of concepts. Teachers used the materials for their teaching but the process of translation and developing written bilingual texts also provided them with a model for written translanguaging that they could introduce into their developing pedagogy. The workshop activities in conjunction with classroom-based support helped teachers bridge the theory-practice divide.

Teachers were encouraged to write down translations of key words and concepts on whiteboards during lessons and also on flipchart paper so that glossaries could be displayed on classroom walls for ongoing reference – creating print-rich multilingual learning environments. Learners copied key words and their translations into their workbooks to create bilingual glossaries (see Figure 3).

FIGURE 3: Translanguaging on whiteboards, posters and into learners’ workbook glossaries.

Teachers adapted the workshop activities to their classrooms: learners worked in groups with their English textbooks and translated unfamiliar words and concepts for themselves, using the multilingual dictionaries and Google Translate to develop bilingual glossaries which they copied into their notebooks (see Figure 4). In science lessons in particular, learners used the glossaries to design bilingual concept posters which they presented to the class for feedback and correction and these were then displayed on the classroom walls.

FIGURE 4: Learners translating terms from English textbooks to develop glossaries and bilingual posters which they present to the class and then display on classroom walls.

FIGURE 5: Translanguaging across genres in English lessons.

English teachers were very conscious that English was both the medium and object of learning and they were expected to stick to English to provide learners with maximum exposure to English language input. However, they found ways of working in more flexible bilingual ways. As there is a common curriculum for languages, teachers found opportunities for cross-curricular work. For example, a writing frame for writing a narrative text (or any other text type) could be translated and cross-referenced across languages; literary genres are studied across languages, for example, learners used a play they had read in their isiXhosa class as a model for analysing the features of a play written up on a poster for an English lesson. In essay writing learners were encouraged to write their first draft in whichever language they could best express themselves so as to focus on meaning, and then to translate it into English as a final draft varying their translanguaging from fluidity to fixity.

The nature of mathematical language is different to English and natural sciences in that much of it is expressed through numerical script which remains the same across written English and isiXhosa.

In Figure 6, a learner is working out a mathematical problem on the board and explaining her reasoning in isiXhosa:

‘Uthatha, nje ngokuba n plus 3 times 4. Kulandawo kaMiss, uMiss pha uthe Usebenzise inumbers ezohlukeneyo yilento inibhidayo [You take, as you can see it’s n plus 3 times 4. Where Miss worked, Miss said here She used different numbers. That’s what’s confusing you].’ (Grade 8 learner, Female, isiXhosa)

FIGURE 6: Explaining a mathematical problem in isiXhosa; fluid translanguaging in writing to show mathematical thinking.

Ms Bizwaphi got learners to explain their mathematical reasoning in reflective writing:

‘in any language that they’re comfortable with. Some wrote in isiXhosa, others wrote in English, others in mixed both languages in one sentence. So then that gives me their thinking. I get to be inside their heads. With solving that problem, what were they thinking?.’

Figure 6 is an example of fluid translanguaging in writing. Teachers made the point that they were deliberately using both languages as resources for learning and thus were developing learners’ knowledge and language skills in both isiXhosa and English.

‘Now [we] translanguage from isiXhosa to English from English to isiXhosa depending on the content that I’ll be teaching at the time.’ (Ms Bizwaphi)

‘We are balancing the both languages. Whatever they do in English, they must also say it in home language.’ (Ms Lekena)

‘So that was fascinating for me because they were using both languages as resources.’ (Ms Cala)

While many translanguaging strategies were common across subjects, the subject-specific content and methodologies also meant that particular strategies were developed for the different subjects.

Teachers’ views on how translanguaging pedagogies had impacted on teaching and learning

Teachers all talked about how translanguaging enabled learners to engage freely and confidently with one another and with their teacher:

‘You can’t have a maths lesson where everyone is quiet. Then there is no learning happening. But then when you hear that they’re explaining things to each other, … I know the noise is what I want, academic noise, which is what we need in a maths class. So it became free … so now they were able to actually think out loud. There was interaction between me and the learners.’ (Ms Bizwaphi)

Teachers noted that learners who were previously excluded from participation in class were now included:

‘This use of translanguaging has been like an equalizer in the classroom … those that could not express themselves well in English always held back and didn’t participate much. … The ones that I thought were not that smart were … competing on equal par now.’ (Ms Cala)

Teachers claimed that in the process of translation from one language to the other, learners would negotiate meaning and in the process deepen their understanding:

‘When we are translating, we have to come to an understanding what this sentence actually means.’ (Ms Cala)

‘I see that it unlocks the highest level of thinking in our kids.’ (Ms Dzina)

Teachers also identified affective issues relating to language, learning, identity, and motivation: as learners engaged in translating concepts into their home language ‘they ended up by owning the concepts’ (Ms Cala). This had important implications for learners’ identity:

‘The learner becomes … proud of his or her language and knows that … I can learn science using my language. Science is not only in English, but science can be done also in my language … science is no longer a foreign something.’ (Ms Lekena)

‘These learners, they see that they are valued even if they speak Xhosa. So they are taken seriously as intellectual beings.’ (Ms Dzina)

Related affective issues were those of learner autonomy and confidence:

‘What is most important is them acquiring knowledge, creating knowledge for themselves.’ (Ms Cala)

‘They’ll just say, “Miss just give me the questions. And also we’ll search. We’ll use our cellphone to search. We’ll also use our textbook and also we’ll go to the library”.’ (Ms Lekena)

‘We’ve got great minds here in maths. So to me it is a lot because they have confidence to express themselves more than to memorise, just to write … they can even argue with me because they do love and understand mathematics.’ (Ms Dzina)

‘They got to see that maths can be fun. It doesn’t have to be feared or that somebody should see it as a subject or content for smart kids.’ (Ms Bizwaphi)

‘It is like a bright light that sparks into their eyes when, they’re able to express themselves in their own mother tongue.’ (Ms Cala)

While teachers were bound by Anglonormative language policies, they were limited to teacher-centred pedagogies or ‘preaching’ (Ms Lekena) as they had to mediate the English texts for learners. However, once they introduced translanguaging pedagogies and multilingual resources, they were able to adopt more learner-centred approaches and learners were empowered to initiate and direct their own learning.

Ms Lekena reflected on this shift:

‘Education is all about involving. You can’t teach the way we were taught. So it seems to me it [translanguaging] is like a new strategy for teaching and learning. I changed the seating plan in my class. The learners started to sit in groups. Before I used to see the learner as an empty vessel that I’m supposed to pour knowledge on. But now I call them the researchers. I say to them, ‘You are here to research. Then for myself, I’m here to rectify you, or I am here to tell you which part you need to focus on for the exam.’

The two teachers who had been strongly in favour of monolingual English-only classroom practices prior to attending the L4L programme, articulated how their ideologies changed over the course of the programme:

‘Language is very, very important. Language can be a barrier to learning. Language can be a barrier to a future of the learners and language can destroy the future of the learner.’ (Ms Lekena)

‘It is totally unfair that although South Africa is a multilingual country it is still one of the few countries where people are still not taught in their mother tongue. Children are not reaching their full potential when they are taught in English. … it is just unjust for people to be taught in a language that is not theirs. It is just an unnecessary problem.’ (Ms Cala)

The key areas of impact of the translanguaging pedagogies, reported by teachers were improvement in learners’ participation, motivation, achievement and self-esteem.

Discussion

What have we learned about changes in teachers’ perceptions and practices from the interview and observation data?

The initial L4L workshops set out to challenge the Anglonormative assumptions constraining teachers’ practice and provided them with the academic authority (‘it was something researched’) to explore ways of leveraging the learners’ linguistic repertoires as resources for learning. The immediate positive response of learners in terms of participation in class increased motivation and confidence and improved learning led to changes in teachers’ ideologies, which in turn motivated teachers to further explore and develop their translanguaging pedagogies, in mutually reinforcing cycles of planning, action, reflection, and adaptation (Plomp 2010).

The L4L workshops provided teachers and mentors with the space for ongoing sharing of ideas reflection and adaptation, as part of the design-based research process. The L4L workshops were informed by small-scale research projects (Charamba 2019; Guzula et al. 2016; Hattingh et al. 2021; Makalela 2016) that had developed written translanguaging for biliteracy in language and science classrooms. The workshop experiences with translation of concepts and creation of bilingual teaching materials provided teachers with a model for moving beyond oral translanguaging to crossing the ‘text frontier,’ as Tyler (2023) describes the resistance by many teachers to engaging with written bilingual materials and practices.

The translanguaging pedagogies as described can be considered critical, decolonial pedagogies as they disrupted the prevalent Anglonormative linguistic hierarchies and enabled learners’ full participation in knowledge construction (Li & Lin, 2019). These pedagogies enabled epistemic access and dismantled the linguistically structured inequalities (Kerfoot & Bello-Nonjengele 2022) that have been central to systemic educational disadvantage for the majority of South African learners.

Limitations and conclusion

The teachers who participated in the discussions were among those who had participated most enthusiastically in the L4L programme and had transformed their classrooms practices. They and others demonstrated what is possible under current classroom constraints, with limited additional resources in the form of science and mathematics bilingual dictionaries and online resources such as Google Translate, as well as classroom support from mentors. However, a minority of teachers were not sufficiently persuaded to let go of their Anglonormative ideologies or perhaps it was simply that they were reluctant to leave their comfort zones and break new ground. Their resistance is the subject for further research, but it is also an indication that teacher education and change is not necessarily straightforward and that the decolonial project is ongoing.

One of the purposes of the L4L intervention and research has been that of advocacy for language policies and practices that fulfil the promise of access, equity and social justice in education. Since the conclusion of the L4L project, the Minister of Basic Education has announced the introduction of ‘Mother-Tongue Based Bilingual Education’ from Grade 4 to Grade 7, starting in 2025 (Republic of South Africa 2023). The findings from the L4L research as described in this article, along with the other papers cited, are timely contributions to ‘usable knowledge’ (Plomp 2010) to inform these important changes.

Acknowledgements

Zenex Foundation funded the L4L project and their support is deeply appreciated.

The contribution of the teachers and learners who participated in this project and agreed to share their ideas in discussions is greatly appreciated.

The author declares that they have no financial or personal relationships that may have inappropriately influenced them in writing this article.

M.J.P. is the sole author of this research article.

Ethical clearance was obtained from the University of Western Cape Ethics Committee (Ref. HS22/2/10), based on the principles of informed consent by participants, with participants reserving the right to withdraw at any point, protection of the identity of participants, and respect for the experience and opinions of participants as collaborators in the research. Permission for the research was obtained from the Western Cape Education Department.

The author disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Zenex Foundation.

The data that support the findings of this study are not openly available because of confidentiality prerequisites.

The views and opinions expressed in this article are those of the author and the product of professional research. It does not necessarily reflect the official policy or position of any affiliated institution, funder, agency, or that of the publisher. The author is responsible for this article’s results, findings, and content.

References

Footnote

1. See https://www.youtube.com/watch?v=1bJt5FVJYis.