Language, translanguaging, and epistemic justice: Multilingual learning across the curriculum

multilingual learning enabled processes of sustained multilingual knowledge-building among peers, as well as the emergence of new social, epistemic and ethical orders from below


Epistemic justice and contemporary language-in-education policies
Epistemic justice challenges forms of unfair treatment that relate to knowledge, understanding, and participation in communicative practices. 4,5It is concerned with relations of knowing: those that construct (or fail to construct) others as knowers and, more importantly, as producers of knowledge.From a decolonial perspective, it is profoundly entangled with questions of language, having to do with who can legitimately know and through what language. 1Epistemic injustice can thus be seen as inextricably bound to coloniality, the racialised structures of power and prescriptions of value that survive colonialism and are kept alive in contemporary structures of governance. 6From this perspective, questions of bi-or multilingual education are not just language or policy matters, but epistemic ones.
There has long been a compelling international consensus that a solid foundation in a familiar language is a robust predictor of educational success. 7,8A recent comparative study of 56 countries spanning six continents found a "very strong, positive correlation between poor basic literacy outcomes and the presence of a mismatch between the languages of home and schooling, controlling for country income" 9(p.9) .Research worldwide shows that children who are not taught in a familiar language can take 5-7 years to catch up with their monolingual counterparts, even in well-resourced contexts. 9,10[10] Confirming the wisdom of these recommendations, a 9-year system-wide assessment of education in Ethiopia found that learners who received 8 years of mother-tongue education achieved the highest scores across the curriculum. 11A range of international longitudinal studies further show that scores for proficiency in the L2, or official language of schooling, improve in proportion with the number of years of education in the first language 12 , thus allaying common fears about the effect of first language instruction on learners' ability to acquire the dominant language of economic and often sociopolitical power.Note that the terms 'mother tongue', 'home language', 'first', 'second' and 'additional' language are used here in line with South African educational terminology.These terms do not, however, capture the fact that many learners come from bi-or multilingual homes.For this reason, the term 'familiar language' is also used to index a language spoken at home and/or more broadly in everyday interactions.
Despite this extensive body of empirical research and a language-in-education policy that emphasises the importance of learning through a familiar language, South Africa, like most countries worldwide, still implements a monolingual, essentially colonial, model of education.This model requires more than 75% of learners to learn through an unfamiliar language from Grade 4 or earlier.Moreover, as urban schools face increasing diversity in learner populations, education authorities lag behind in imaginative responses to promoting the linguistic and academic well-being of all learners, instead maintaining the essentialised connections between language and ethnicity on which the apartheid system was founded, and perpetuating social and epistemic harm. 1 As a result, and almost without exception, linguistically structured inequalities persist.They are evident in substantial differences in educational outcomes for speakers of marginalised languages.The latest South African results in the Progress in International Reading Literacy Study (PIRLS 2021) stand as stark testimony.In 2021, 81% of Grade 4 learners could not read for meaning in any language -a 3% increase since 2016.Brazilian Grade 4s are 3 years ahead of South African Grade 4s. 13 These literacy outcomes are compounded by weak disciplinary knowledge 14 , which is particularly acute in science and mathematics where, for example, in 2021 only 16% of the 36% writing the Maths exam achieved 60% or more in Mathematics, a figure which converts to 6% of all those writing matric 15(table 2.5) .These figures suggest that, by the end of their schooling, the majority of those learning through a second or additional language acquire "only the most superficial ability to engage with school knowledge, that is, to negotiate it, appropriate it, transform it or transmit it effectively to others" 1(p.466) .The authors of the paper discussed here argue that such monolingually oriented language-in-education policies and practices constitute a form of epistemic injustice in that they can inhibit or prevent learners from making epistemic contributions, that is, from conveying knowledge to others. 4For Miranda Fricker, this kind of "testimonial" injustice is: "a wrong done to someone specifically in their capacity as a knower […] in a capacity central to human value" 4(p.1,5) .It is frequently tied to deeper forms of "hermeneutic" injustice through which dominant, often northern, interpretative frames can obscure "significant aspects of [individuals'] social experiences from collective understanding" 4(p154) .These language-based forms of epistemic injustice sustain social perceptions that misrecognise the cognitive capacities of second language speakers as deficient or less credible. 16In classrooms, they can result in epistemic exclusion, silencing, or resistance.Moreover, these injustices are "tracker prejudices" which are "systematically connected with other kinds of actual or potential injustice" 4(p.27) .In South Africa, they render learners vulnerable to "long-term trajectories of economic and sociopolitical exclusion and disadvantage, along with reduced confidence in their own epistemic worth" 1(p.466) .

the transformative potential of unmarking multilingualism
In contrast, the paper underpinning this commentary explores the ways in which a transformative language-in-education policy can contribute to epistemic justice by countering exclusion, reducing harm, and promoting more equitable participation.The study shows what happens when multilingualism is 'unmarked' -seen as the norm rather than an exception -and all languages are given equal value as learning resources.It examines the interactions of a group of learners in a large, multigrade, English-medium, Economic Sciences classroom in Delft, a large 'township' on the periphery of Cape Town with a uniformly low socio-economic status.All learners in this group spoke isiXhosa at home but also English and Afrikaans with friends at school.The analysis illuminates the ways in which they used the different languages and language varieties in their repertoires -a practice often known as translanguaging -to negotiate knowledge, construct others as knowers, and promote solidarity.The analysis focuses on a 13-yearold learner who had chosen isiXhosa as medium for the first 5 years of her schooling.She had successfully completed each year, a lively participant in class who obviously enjoyed learning.However, once she reached Grade 6, her mother, perhaps influenced by wider neoliberal ideologies of the low value of African languages, changed her to the English-medium stream.She subsequently failed and had to repeat the grade.In contrast to previous years, observations during Ensha's first year in Grade 6 showed her as a mostly silent and non-participating member of the class, often visibly distressed by others' reactions to her ways of speaking in English.Here her ability to be recognised as a competent knower was compromised by her inability to articulate that knowledge in a particular language.This linguistic marginalisation can be seen as a form of epistemic exclusion, reducing her ability to participate in a given epistemic community and denying her "part of what it is to be fully human" 4(p.4) .In such cases, a person "may be, quite literally, prevented from becoming who they are" 4(p.4) .
As a result of these experiences, even after passing Grade 6 the second time in English, Ensha stated unequivocally that she would have preferred to continue with isiXhosa as medium.Nevertheless, her failure and then repetition of Grade 6 had interesting consequences for her academic identity: her second year in this grade coincided with the implementation of the LTP where all languages were considered legitimate resources for learning.She took advantage of this to mediate knowledge for her Grade-6 peers in the English class: observations showed her moving transmodally between the English textbook, the diagrams, isiXhosa, and features of other urban vernaculars such as Kaaps Afrikaans, along with embodied meaning-making strategies such as gaze and gesture.In this way, she engaged in complex forms of multilingual, multimodal knowledge construction, including the meshing of academic and informal registers.Her 'schooled' register was for the most part in isiXhosa and was characterised by use of the passive form, complex and compound sentences, increased semantic density and abstraction.It was striking that isiXhosa was used for all significant discipline-specific, knowledgebuilding work (elaborating, exemplifying, evaluating, building from example to abstract concept).This finding is evidence of the fact that without support or previous learning of this content in isiXhosa, she was able to resemiotise knowledge acquired in English the previous year into this language.It belies the belief that teachers are unable to use first or home languages for content subjects without lengthy and specific training.
A further significant feature of her knowledge-negotiating practices was the use of a range of inclusive and dialogically expansive strategies which served to maintain relations of solidarity, work against hierarchical relations of knowledge, and strengthen a participation framework in which all had equal status as knowers.For this purpose, she used humour, discourse markers of familiarity, and face-saving moves; she refrained from explicit evaluations of others' contributions, instead leaving space for disagreement and seeking consensus.Here, too, moving seamlessly among languages was a key feature.The dialogic relationships created in this way embodied an emergent and decolonial ethics of knowing, premised on openmindedness, respect for others, and solidarity.Such values are associated with the relational ethics of 'ubuntu' 17 , an Nguni Bantu term meaning 'humanity', sometimes translated as 'I am because we are'.

towards epistemic justice
The implications of these findings for future policy and planning are both conceptual and practical.The argument made here is that a shift from a monolingual to a multilingual language-in-education policy is a relatively straightforward and easily implementable strategy to address longstanding issues of epistemic injustice and rapidly enhance learning outcomes.This shift will require two conceptual moves: from epistemic access to epistemic justice, and from languages to languaging and translanguaging.
In postcolonial contexts, epistemic access, defined as access to the knowledge that educational institutions distribute, is geared towards a monolingual outcome, that is, towards academic proficiency in only the official language.As outlined above, epistemic justice on the other hand, challenges us to find ways of recognising all learners as knowers, not only accessing but producing knowledge and conveying it to others.This implies a second shift to the recognition of all language resources as legitimate learning resources, that is, constructing educational systems on the understanding that multilinguality, or the multilingual capability, is an essential condition of what it means to be human. 18cal and international research has demonstrated the potential of translanguaging or trans-semiotising practices such as Ensha's for supporting scientific argumentation, conceptual understanding, and learners' confidence in their own epistemic abilities.However, the goal in the vast majority of cases researched is ultimately to promote access to knowledge in the second language, that is, to support a transition to the official language of schooling.Translanguaging, therefore, very often remains "an affirmative rather than transformative strategy, leaving underlying hierarchies of value and relations of knowing unchanged and promoting epistemic access rather than justice" 1(p.464) .The concept of languaging 19 seems better able to carry the onto-epistemological challenge of rethinking language in education.While translanguaging suggests moving across fixed language boundaries, languaging captures the flexibility of spontaneous multilingual practices, dynamically combining languages and language varieties and incorporating a plurality of expressive practices.
The practical implications of these conceptual moves include the following:

•
Begin with multilingualism as the norm and see multilingual practices as the basis for acquiring and producing knowledge.
• Recognise all linguistic varieties and semiotic modes as legitimate in producing and mediating knowledge.
• Encourage teachers to move ahead with multilingual learning even when the resources commonly thought necessary are not yet in place.
• Promote peer mediation processes, especially in highly diverse classrooms where teachers cannot be proficient in all languages spoken.

•
The choice of first language instruction and/or translanguaging pedagogies will depend on the profile of the learners.Therefore, decentralise decision-making and support teachers and schools to develop multilingual policies, pedagogies and practices from the ground up.
• Offer pre-or in-service bilingual teacher education which models processes of pedagogical translanguaging, multilingual learning and assessment, along with ethical relations of knowing.
• Engage with teachers, caregivers, and communities to enrich understandings of the value of multilingual resources and multilingual literacy practices.
Experience in the LTP further demonstrates that publishers are ready and willing to produce bi-or even multilingual texts when sure of state support.Moreover, rapid developments in technology and expanded digital access across Africa offer enriched possibilities for the support and development of multilingual literacies and learning across the curriculum (see for example, the online African Storybook project in over 200 languages 20 and work done on simulated classrooms for teacher education 21 ).
Education authorities, when well informed and willing to put learners' well-being and life opportunities ahead of wider political goals, can ease obstacles and promote swift rollout as in the case of the recent Eastern Cape pilots. 22 The announcement in May 2024 of an incremental nationwide rollout from 2025 offers long-overdue hope to future generations of South African learners. 23stly, decades of economic research have shown that the initial costs of mother-tongue-based multilingual education may be 4 or 5 percentage points higher than those for a single language of instruction across all schools. 24Technological advances can now reduce this cost substantially.Moreover, the cost of mother-tongue-based bi-or multilingual education to the state will always be far less than the cost of high repetition and pushout rates and, in the long term, of citizens who are unable to participate fully in social, economic and political processes.

Conclusion
The research presented here has highlighted the potential of multilingual learning for the promotion of epistemic justice.Such policies and pedagogies hold equal promise for progress towards the United Nations Sustainable Development Goals of quality education and reducing inequalities within and between countries, with longer term implications for reducing poverty and promoting economic growth.
While "language is not everything in education, […] without language, everything is nothing in education" 25(p.92) .Language on its own cannot transform entrenched structural inequalities.Nevertheless, unmarking multilingualism and giving equal value to all languages as epistemic resources has the potential to disrupt hegemonic regimes of language and learning, constructing new relations of knowing and new paths to knowledge.It can also begin the long process of undoing epistemic harms and lay the basis for conditions of greater epistemic justice.