Abstract
Multilingual education is an urgent and pressing concern in the Indian educational scenario. While the National Education Policy (2020) acknowledges multilingualism as a resource in educational contexts and reiterates earlier policies calling for mother tongue-based education in elementary classrooms, it does not provide guidance in terms of how to productively accommodate multiple languages in the classroom. Multilingual education will be much stronger if it is based on a strong understanding of multilinguality—the idea that the human mind is fundamentally multilingual in nature. A new, but substantial paradigm of scholarship addressing multilinguality is that of ‘translanguaging’, which views named languages as socio-political constructs and argues that multilinguals have a unified linguistic repertoire that they flexibly, creatively and adaptively draw upon. Accepting the grounding assumptions of translanguaging would has important implications for curriculum, pedagogy and assessment in educational spaces. In this article, we describe and critique the translanguaging perspective, even while acknowledging its positive contributions. We point out, especially its failure to provide guidance in terms of how to productively accommodate translanguaging in classrooms.
The recent National Education Policy (Government of India [GoI], 2020) has stated that mother tongues should be the basis for instruction in classrooms ‘…until at least Grade 5, but preferably till Grade 8 and beyond…’ (p. 13). It has also stated that multilingualism has great cognitive benefits for young children, and therefore, children will be taught multiple languages with high quality. The ‘Three-Language formula’ (suggested by the Central Advisory Board of Education in 1956) calling for the simultaneous teaching of mother-tongues/regional language, English and Hindi will continue. Much has been written in the literature about the inadequacy of the Three-Language Formula to address the needs of multilingualism in our country. The critique includes the fact the formula does not distinguish between regional languages and mother tongues but is often interpreted in practice as a ‘two-modern languages formula’ for speakers of Northern/Hindi-speaking regions of the country who use Sanskrit as the third language. Furthermore, languages are taught in a formulaic, examination-based manner that does little to add value to the multilinguality of the country. The formula itself is also not enforceable, leaving private schools free to provide English as the medium of instruction (Annamalai, 2001; Groff, 2017; Jhingran, 2005; Khubchandani, 1981; Rao, 2021). In this context, it is necessary to question whether a continuation of the Three-Language Formula would serve the stated intentions of providing high-quality instruction in multiple languages using the mother tongue as the basis for doing so.
India has seen relatively few efforts at theorising and researching multilingual education. Mohanty and his colleagues have written prolifically and convincingly on the strengths of a mother-tongue based multilingual pedagogy (e.g., Mohanty & Panda, 2015; Mohanty et al., 2009a; 2009b; Nag, 2005). Their work conducted largely in tribal settings in Odisha used the local tribal languages as the basis for curriculum and pedagogy, even as children were introduced to Odiya and English. Mohanty and colleagues have argued persuasively that multilingualism is the norm world-wide and deserves attention and study, instead of principles and practices for language education being derived from monolingual norms that do not apply to the multilingual Indian reality (e.g., Mohanty, 2019). Similarly, Agnihotri (1995, 2007) has also challenged monolingual ideologies in education and urged that multilingualism be viewed as not just the norm, but also as a creative and counter-hegemonic resource in the classroom. Pointing out that multilinguality is not specific to India or South Asia or Africa, Agnihotri states that ‘…multilinguality characterizes being human and living in social groups and networks’ (Agnihotri, 2007, p. 8). While multilingualism is recognised and given space, however inadequately, in public discourses related to Indian education, multilinguality is an underdeveloped idea in the policy arena. The two terms represent slightly different concepts. Multilingualism is the state of knowing multiple languages; whereas multilinguality refers to the idea that multilingualism is a universal human condition. Some societies may appear to be monolingual due to the active suppression of different languages and language varieties; but, at their core, all human societies are multilingual. Thus, multilinguality is the normative condition, and in order to nurture it, we need to transform conditions that actively suppress it.
In keeping with this line of reasoning, Agnihotri (2007) has argued that language use is dynamic, diverse and hybrid in its formulations and usage; thus, the very idea of ‘a language’ should be rejected and we need to reconceptualise what is understood as language. Agnihotri’s questioning of what constitutes ‘a language’ echoes and builds upon an emerging, but substantial body of work referred to as ‘translanguaging’ (Lewis et al., 2012). Despite the relatively large amount of scholarship that has devoted its attention to the translanguaging framework, it remains a relatively new and poorly understood construct in the Indian education context. In this article, we take a close look at what translanguaging means and what it has to offer to the multilingual education discourse in India. In particular, we argue that while it has important insights to offer to the space of multilingual education, it remains limited both by certain conceptual blind-spots as well as by failing to offer viable curricular guidance.
What is Translanguaging?
Translanguaging seems to have brought about, what Ke and Lin (2017) calls, ‘a multilingual turn’ in the study of languages. Language practices of multilinguals are now often explained in terms of translanguaging, leading scholars to struggle with pertinent questions regarding the validity of monolingual frameworks that have been traditionally employed in the field. However, as a construct, translanguaging does not have a long history in the study of languages and linguistics.
The term translanguaging was first used by Cen Williams to refer to an intentional and systematic pedagogical shift between languages. Williams (1996) defined trawsieithu (‘translanguaging’ in Welsh) as a pedagogical strategy in which facilitators provide the input to students in one language, and students process the input and produce the output in another. It was assumed that students would be able to use the information in another language successfully only if they understood it fully in the language in which the input was made available (Lewis et al., 2012). Baker (2001), who translated the term from Welsh to English, stated that reprocessing the content (in another language) through translanguaging led to expansion, extension and intensification of learning. The term thus referred to the pedagogic practice of language switching that was related to meaning making and facilitation of learning.
Translanguaging gained global attention after Ofelia Garcia reconceptualised the term to widen its scope beyond the domain of education. In her book titled Bilingual Education in the 21st Century, Garcia (2011) extended the term translanguaging from the context of pedagogic practices to refer to the general way of using languages in bi/multilingual contexts. In her view, bi/multilingual contexts are marked by heterogeneity of language practices that involve intricacies which cannot be understood in terms of discrete, named languages. Named languages (such as Hindi, Tamil, English, French and Spanish) were assumed to be artificial socio-political constructs that are imposed externally on complex linguistic behaviour. Real language practices of multilinguals go beyond the fixed boundaries of named languages and are fluid, dynamic and ever-changing in nature. The morpheme trans- in the term translanguaging indexes the ability of multilinguals to transcend the boundaries that are traditionally drawn between named languages, whereas the present continuous tense marking morpheme –ing emphasises the continually changing nature of language practices. Hence, translanguaging for Garcia (2011) is not just a pedagogic strategy; it refers to discursive practices of multilinguals that are fluid, hybrid and dynamic.
It must be noted that the construct of translanguaging is not synonymous with concepts such as interlanguage (Selinker, 1972), approximative systems (Nemser, 1971) or transitional competencies (Corder, 1981) that have also been used to describe bi/multilingual behaviour. The term interlanguage refers to the use of a dynamic, rule-based, but deviant linguistic system that is developed by a learner of a second language who has not yet achieved proficiency (Selinker, 1972). In this sense, the concept of interlanguage highlights a deficiency in competence rather than a norm. Concepts such as approximative systems and transitional competencies, likewise, refer to stages of development while the learners shift from their first language to the second. They are used to explain deficits that learners overcome (or fail to overcome) as they become proficient in the target language. The construct of translanguaging, in contrast, is based on fundamentally different presumptions.
In its initial conceptualisation (Williams, 1996), the term was used to support the simultaneous development of two languages (Welsh and English) in the classroom. In Garcia’s reconceptualisation, it established as a norm representing the fluidity and hybridity of multilinguals’ language practices. In other words, translanguaging does not signify lack of proficiency in languages or a state of development; multilinguals may engage in translanguaging even when they are fully proficient in all (or one of) the languages involved. Translanguaging demonstrates the complex ways in which a speaker’s multilingualism manifests in their utterances. Consider the following conversation between a multilingual mother and her 8-year-old daughter, both of whom are moving fluidly among Hindi, English and Jaunsari (a language spoken in the foothills of the Uttarakhand region in India) while preparing for dinner.
[plain text- Hindi; bold- English; italics- Jaunsari]
Child: Maa, Mother: Aachha? Kauso? (Really? How?) Child: Mother: Haan, ped pe to latakta nahi hai (laughs). (Yes, it doesn’t hang on trees.)
This conversation provides an interesting example of fluid movement among three so called ‘discrete’ languages. In the first utterance, the child inserts English vocabulary within a sentence in which Hindi governs the syntax of the sentence. When her mother questions her in Jaunsari (in utterance b), the child responds to her in Jaunsari in the first part of her utterance, and then moves back to Hindi in the second part of her utterance (c). Interestingly, both parts of the utterance (produced in Jaunsari as well as Hindi) have English lexemes, ‘packet’ and ‘factory’ which have been inflected with plural marking morphemes ‘-o’ of Jaunsari and ‘-on’ of Hindi, respectively. The corresponding words dabiyoon (packets) in Jaunsari and karkhaano (factories) in Hindi/Urdu (or Hindustani) are less frequently used in contemporary everyday conversations in urban, middle-class contexts. In the last utterance, the mother makes a move to Hindi, probably to resonate with the child’s choice of language. Such translanguaging practices are a part of their regular interactions. In other words, their multilingual identities are embedded in such utterances.
It is important to emphasise that the data presented here seems to be ‘code-switching’ or ‘code-mixing’ by another name. In fact, it is a different interpretation of the same empirical data that distinguishes theories of code-switching and mixing from translanguaging. While in the former set of theories, one could conclude that these words represent instances of lexical borrowings across three discrete languages; from a translanguaging perspective, borrowings must be interpreted as instances that blur the boundaries among the languages in a rule-governed manner. Challenging the dual-mental grammar hypothesis which forms the basis of theories of code-switching, translanguaging presupposes a unitary, integrated mental-grammar (Cenoz & Gorter, 2015; Otheguy et al., 2015; 2018). Canagarajah (2011), for instance, contends that multilingual competence does not include separate competencies for each language that speakers know, but rather ‘a multicompetence that functions symbiotically for the different languages in one’s repertoire integrated systems’ (p. 401). In all such definitions, the term has been used to refer to the skilful use of linguistic features available in the ‘whole’ of one’s multilingual repertoire. Thus, when viewed in linguistic terms, ‘language’ collapses into a set of idiolects (each individual’s unique, personal language) such that no two idiolects are ever the same – even among siblings raised in the same household. Makoni and Pennycook (2007, cited in Garcia, 2009) argue that named languages are not a priori natural systems; rather, they are sociocultural and political inventions that have historically benefitted monolingual communities, while invalidating the linguistic practices of multilinguals.
A number of scholars view translanguaging as a means to an emancipatory end—as a way to ensure engagement, inclusivity, linguistic rights, democracy and pluralisation voices in society. Garcia (2011, p. 3) suggests that ‘by making use of flexible language practices, translanguaging releases ways of speaking that are often very much controlled and silenced. When new voices are released, histories of subjugation are brought forth, building a future of equity and social justice’. Translanguaging, therefore, draws from hetroglossic ideologies to recognise and validate the voices of multilinguals that are often suppressed under dominant monolingual ideologies. The notion of heteroglossia rejects structuralist views of language, ‘which conceives language as a static entity attached to a particular (national) speech community’ (Ke & Lin, 2017, pp. 34–35).
Otheguy et al. (2015, p. 295) posit that
The difference between monolinguals and bilinguals is that monolinguals are allowed to deploy all or most of their lexical and structural repertoire mostly freely, whereas bilinguals can only do so in the safety of environments that are sheltered from the perspective power of named languages.
In the realm of education, monoglossic ideologies often translate into ‘one language only’ pedagogies where the use of fluid language practices by learners is equated with ‘linguistic corruption’ by the teachers. Use of languages other than the target language by the teacher is interpreted as an act of ‘diluting’ of the input being provided to the students. Such pedagogies jeopardise the multilingual identities of students and teachers and forces them to wear the garb of monolingual speakers. Translanguaging challenges such views on language education and sponsors a multilingual perspective whereby, instead of suppressing a part of one’s linguistic repertoire, the whole set of linguistic capabilities of students and teachers is valued and utilised.
In summary, it can be stated that the analysis of various definitions of translanguaging shows that the term is being constantly broadened since it was first coined. It has been used to refer to a number of multilingual phenomena including pedagogic strategies (Williams, 1996), discursive practices (Garcia, 2011), the ability of multilingual speakers to fluidly move among languages (Canagarajah, 2011), to the deployment of a speaker’s full linguistic repertoire within a conversation and as a means to disrupt language-based hierarchies and address historical subjugation of multilinguals (Garcia, 2011).
Translanguaging in Indian Classrooms
While a significant number of studies have been conducted on pedagogical translanguaging (e.g., Prilutskaya, 2021 for a review of literature), only a very few studies have investigated the nature, role and implications of translanguaging-based approaches to teaching and learning in Indian classrooms. Most of these studies discuss the negative connotations that are attached to fluid language practices in the context. This could be related directly to the hegemony of monoglossic ideologies and a tacit rejection of multilingualism within the education system.
Pallavi (2021) examined classroom conversations to understand the nature of translanguaging practices which multilingual students and teachers used in the context of government-run primary schools in Delhi. The analysis revealed multiple ways in which multilingual students and teachers crossed boundaries that are typically drawn between so-called ‘separate’ languages to fluidly engage in translanguaging during the process of teaching and learning. The study also looked at how teachers’ perspectives regarding fluid language practices impacted the kind of multilinguality that they permitted in their classrooms. Consider the following example quoted in the study.
(The teacher asked the students to write translations of Sanskrit words in the textbook.)
[Plain text- Hindi; Bold- English; Italics- Sanskrit] T: chashkah. ab aap dekhenge chashkah. kahan hai? upar se dusra shabd hai. usko S: gilaas (Glass) S1: T: nahi (forcefully)! uske baad agla shabd hai brihad (No (forcefully)! After that the next word is big.) S2: brihad (Big) T: pahle usko S: bada. (Big) T: ye dekhiyega. samajh me aa raha hai? (Look at this. Can you understand?) S:
Readers are again reminded of the fact that these data could be interpreted as an instance of code-mixing. It is in the interpretation of what is happening in the multilingual person’s brain that the two paradigms differ, with the translanguaging paradigm suggesting that there is a unified linguistic repertoire in the speaker’s mind. The example shows two things; first, that speakers naturally and fluidly used words from different named languages in their interactions; and second, that while translanguaging between Hindi and Sanskrit was accepted by the teacher, translanguaging between English and other languages (Hindi and Sanskrit) was discouraged, especially in written work. The data collected from interview during the study showed that the teacher followed a strictly monoglossic perspective (supporting-nationalist ideologies) and viewed English as a foreign language that was ‘intruding’ into Indian languages. Furthermore, while he allowed some flexibility in oral conversations, the teacher talked about the primacy of maintaining the purity of languages especially in written mode during his interview. His responses indicated that instances of translanguaging while writing were equated with language corruption and penalised in his classes.
Anderson and Lightfoot (2018) have likewise reported that teachers associate negative connotations with fluid language use in Indian schools. Even though most of the teachers used ‘other languages’ while teaching English, they did so with a sense of guilt. They were either discouraged from or prohibited by their institutions to use any other language than English, while teaching English. They argued that
greater acceptance and inclusion of the natural translingual practices reported in local communities will help learners to develop their linguistic resources holistically and humanely, this latter factor being particularly important at primary level and in rural communities, where learners are likely to be most disadvantaged and vulnerable to dropping out of school. (ibid., p. 15)
However, the study did not establish any corelation between the use of translanguaging and achievement in learning (or school drop-outs) through data.
Evidence from a study conducted with Grade 8 students in government schools in Andhra Pradesh shows that there were three major advantages to permitting students to engage in translanguaging practices (among Hindi, Telegu and English) while composing essays in English (Sathuvalli & Chimirala, 2017). First, it allowed students to develop a positive view of other languages while paving the way to build the target language; second, it encouraged students to move from teacher controlled to ‘self-initiation’ stages during learning activities; and third, it built meta-awareness about linguistic resources that were available to support learning. Bisai’s and Smriti’s (2019) study on translanguaging and collaborative learning with Grade 8 students of a state-run school in Paschim Medinipur district of West Bengal, India, similarly showed that interactions that are done using translanguaging as a resource allowed students to meaningfully engage in various collaborative tasks. Furthermore, it assisted the linguistic minority students ‘to bring their linguistic repertoire into the classroom, validate their identity and enable them to make sense of their world’ (p. 7). It was, therefore, also linked to democracy, inclusivity and engagement in this study.
In summary, it can be stated that although the sparse examples from the Indian context indicate that translanguaging is used extensively by teachers and students in their conversations as a language practice, and that pedagogical translanguaging relates positively with learning outcomes in the Indian context, theoretically there is a dearth of the literature that look at classroom processes from a translanguaging based approach. In addition, translanguaging is often rejected in classrooms due to the hegemony of monoglossic ideologies. Much more research needs to be conducted to establish the role that translanguaging plays in challenging linguistic hierarchies, or in developing linguistic capabilities.
Translanguaging: A Critique
Otheguy et al. (2015) offer two reasons for a widespread acceptance of translanguaging. First, problematising named languages would free us to understand the implications of translanguaging for the education and assessment of bi/multilinguals. Second, the widespread acceptance of translanguaging would help to disrupt the socially constructed language hierarchies that are responsible for the suppression of many languages globally, especially in schools and classrooms. Each of these needs to be considered carefully—should we dismantle named languages? And would this create fundamentally transformative conditions for equity in society? We turn to a consideration of each of these in the following sections in order to argue against both.
Named Languages
We argue that the foundational thesis of translanguaging is a socially well-intentioned, but conceptually problematic one. The insight that actual individual language use does not always correspond with distinct named languages has been stretched to dichotomise the ‘linguistic’ from the ‘cultural’. This calls the fundamentally cultural nature of language into question. From a Vygotskian perspective, there is nothing that is purely ‘linguistic’ in the human repertoire after the age of approximately 18 months when children begin word naming (Vygotsky, 1962). From this viewpoint, it could be argued that the psychological nature of human beings is an aggregate of social formations and relations transferred within, which then become functions of the personality and dynamic parts of its structure. That is, what is conventionally viewed as ‘internal’ (or ‘linguistic’ in this case), originates externally in social relations and interactions, and is gradually internalised. Vygotsky (1962) maintained that the way in which reality is generalised and reflected in a word is in itself both conceptual and cultural, with both sound and meaning of the word being indivisible parts of a whole. For example, in many Indian languages, the word for the maternal grandparent is different from that for the paternal grandparent. The need for this distinction possibly arose from sociohistorical/cultural arrangements of family life and the different roles assigned to each of these grandparents in a child’s life. Whether a Malayalam speaking child refers to their maternal grandfather as muthashan or ammachan or, more idiostylistically as mushi does not take away from the fact that the cultural is already and always present in the very manner in which the category of the maternal grandparent has been conceptualised in that culture. Therefore, attempting to separate out the lexical/grammatical aspects of language from the cultural is an endeavor that is on theoretically very shaky grounds. It should be emphasised that from a Vygotskian/Marxist approach, the ‘cultural’ would also include sociohistorical, economic and political formations.
We draw upon Searle (2006) to further question the critique of named languages mounted by translanguaging theorists. Searle points out that very little in social reality has any ontological objectivity. While physical objects like mountains and mosses might have some existence independent of the observer, elections, money, marriage and so on are aspects of social reality that are entirely ontologically relative. Yet, social reality can be ontologically relative, even while being epistemically objective. For example, the idea that a piece of paper represents a specific currency of a particular denomination worth an exchange value of a certain amount has no ontological objectivity. However, a social idea such as this would have epistemic objectivity if it fulfilled certain conditions, such as collective intentionality, assignment of specific functions, and constitutive rules, norms and procedures for regulating that aspect of social reality.
If we were to evaluate named languages by this set of criteria, it becomes immediately apparent that as an aspect of social reality, named languages have no ontological objectivity. In that, the translanguaging theorists are correct: ‘Kannada’ or ‘English’ or ‘Hindi’ have no discrete existences of their own apart from the meanings given to them by observers. Still, the lack of ontological objectivity does not mean that named languages are entirely arbitrary impositions that hold no social validity. Perhaps, it is better to think of them as collective fictions that exist on the basis of shared intentionality, assignment of specific functions in society, as well as the existence of rules and norms that govern their usage. Like all social fictions, they inherently hold the potential to be useful as well as oppressive, and often, both simultaneously.
That considerable heterogeneity exists among members of speech communities has been acknowledged for several decades (LeFebvre, 1974). Labov (1966) maintained that linguistic variation exists in the performance of the members of every linguistic community, despite all members sharing the same basic grammar. In contrast, other scholars have proposed the existence of more than a single grammar within a linguistic community, such that the variations of speech communities could be described along a scale with each point of the scale representing a unique grammar (Stewart, 1964, cited in LeFebvre, 1974). The intent of this article is not to settle the debate in terms of whether overlapping linguistic variations are accounted for by a single grammatical system, or multiple; but to point out that the presence of what translanguaging theorists refer to as ‘idiolects’ has been acknowledged by other terms in the literature, most notably, as linguistic variation in speech communities. It is true that most of the research in this latter tradition has been conducted on sociolects (speech variations demonstrated by groups of people); however, this does not theoretically preclude the possibility that variability could trickle down to every last member of the linguistic community having their own idiosyncratic linguistic usage patterns.
What, then, has been added to the existing discourse? It appears that the thrust of the conceptual contribution of the translanguaging theorists is their underlining and elaboration of the idea that named languages are not bounded, autonomous, impermeable systems, but are external impositions that do not necessarily correspond to the internality of how multilingual speakers experience or use their integrated linguistic repertoire. We have pointed out in at the beginning of this section that Vygotskian/Marxist psychology would question the dichotomising and polarisation of unities such as internal-external, linguistic-cultural, individual-social and so on. From this perspective, transformation cannot reductively target certain dimensions of materiality and exclude from the critical gaze certain other structures that produce and maintain discrimination. Neither would this argument hold up to philosophical scrutiny in that most of social reality has no ontological objectivity; and if that were to be the gold standard of acceptance, then much of social life would need to be stripped away to expose the fluid, dynamic phenomena that form the basis of social reality. As Searle (2006), pointed out:
My dog can see a person carry a ball across a line; but what he cannot see is the person scoring a touchdown.. The physical facts that the dog sees and the physical facts that I see are exactly the same. What I have, that the dog does not have, is the capacity to represent those facts in a certain way, to represent them as existing at the higher level … what is special about human beings is that they have the capacity, which the dog does not have, to see and think an institutional reality, but that is impossible just on the basis of the sheer physical facts because there is nothing in the physical facts to give the semantic content either to the thought or to the perception. (pp. 19–20)
Translanguaging: A Transformative Pedagogy?
Canagarajah (in De Costa et al., 2017) points out that the prefix trans in translanguaging refers to pedagogies that not only transcend individual languages, or standard semiotic resources, but is also seen in the word, transformative. Pedagogies based on this paradigm could potentially contribute to cultural and linguistic decolonisation and challenge prejudice and discrimination based solely on one’s language use. De Costa et al. (2017) describe three principles that underpin translanguaging pedagogies: (a) a broadened understanding of the semiotic resources that multilingual students bring into the classroom; (b) student repertoire expansion beyond engaging with canonical texts and forms in the classroom and (c) inviting students to critically reflect on their own linguistic histories and repertoires and to use their multilingual resources in the classroom.
This is, of course, intuitively appealing to most progressive educators working in multilingual contexts. Permitting teachers and learners to legitimately use their entire, integrated linguistic repertoire in the classroom seems much more commonsensically appealing than to forbid or to have them surreptitiously use other languages. Examples of translanguaging pedagogies in the literature typically are from work conducted with older (secondary school and undergraduate) students and demonstrate either how ‘naturally’ translanguaging occurs in multilingual classrooms; or, else, how students are helped to develop a critical awareness of the dynamic, contested and material nature of the languages they use. Taken together, it is unclear how these studies add to existent understandings about the plurilingual ethos of Indian classrooms (e.g., Agnihotri, 2007; Khubchandani, 1996).
Furthermore, the studies, taken as a whole suffer from several gaps that are summarised by Canagarajah (2011). A major shortcoming is that many studies conducted in school contexts show translanguaging to be a naturally occurring phenomenon, but do not highlight the conscious use of pedagogical strategies that would promote the development of students’ linguistic repertoires. Canagarajah (2011) notes that at best, many of the studies demonstrate the provision of safe spaces by teachers to use multiple languages in the classroom, as well as some collaborative attempts to enable students to do so. If translanguaging is seen as an intuitive capacity that all multilinguals possess, it might be simplistically assumed that it does not need to be taught, but simply permitted. This leads to a second concern—how can we conceptualise of development within a theoretical framework that does not seem to permit the recognition of ‘errors’ at all? If all attempts at language use by students are viewed as creative, enabling and productive, then it would be impossible to set curricular or pedagogical goals developmentally or to progressively enable students to become more proficient.
Canagarajah also points out that rhetorical considerations must be applied in assessing the effectiveness of translanguaging. Researchers working in the tradition have rarely asked whether translanguaging is appropriate for that context in rhetorical terms. Lisa Delipt, the noted African American educator, has cautioned us that seemingly progressive discourses with emancipatory intent might marginalise the already marginalised by denying them explicit access to ‘codes of power’ (Delpit, 1988). Delpit emphasised that the underprivileged need to be taught these codes explicitly. ‘If you are not already a participant in the culture of power, being told explicitly the rules of that culture makes acquiring power easier’. (ibid., p. 282). From that perspective, while a child from a non-English speaking background might be encouraged to draw upon her linguistic repertoire fluidly, she could also be gradually presented with models of higher power language/register use and explicitly coached on them.
If translanguaging is to be deemed as ‘transformative’ in nature, it is necessary that scholars working within this tradition be asked to provide evidence that simply permitting children to engage in translanguaging in classrooms is sufficient to counter the hegemony of the powerful languages. However, even the data that we have presented in the earlier section of this article clearly demonstrates a hierarchy between the language varieties used during translanguaging – the content words used in the conversation are more likely to be resourced from relatively powerful languages while the function words are in all likelihood resourced from the relatively powerless languages. This division of labour is not at all surprising—it mirrors the current linguistic order. Such an order is antithetical to a ‘transformative approach’ to language education. Discussions of translanguaging that we have accessed do not clearly address how these inherent hierarchies in language varieties can be transformed in the classroom through the use of a translanguaging perspective or pedagogy.
It would appear that the intent of translanguaging pedagogies is more to dismantle essentialising, structuralist accounts of language use, than it is to empower marginalised students to critically access the existing codes of power described eloquently by Delpit. We have alluded in an earlier section to how Vygotskian conceptions of language do not exclude the cultural (including sociohistorical, sociopolitical and socio-economic aspects). In contrast, the translanguaging perspective is somewhat reductive in its conceptualisation of transformation—restricting it to the freedom to understand and use multiple language varieties freely without the restrictions imposed by the presence of named languages. Approaches to social justice that are one-dimensional lose their critical edge; truly transformative approaches would need to take a more nuanced and multi-dimensional approach. For example, from this perspective, it is not possible to consider the fact that since hierarchies do exist in language varieties in societies, there is a real danger in not providing children, especially those from already marginalised linguistic communities, with access to powerful norms and discourses. Lodge refers to this as the ‘access paradox’ (Lodge, 1997, cited in Janks, 2004)—not providing individuals with access to powerful codes further marginalises them; while providing them with the access to these codes further reinstates or reinscribes the codes of power. The complexity of this paradox needs to be acknowledged, and nuance adopted in navigating it.
Another gap appears to be not including the voices and wishes of parents, communities, teachers and students themselves. There are no accounts in the literature of translanguaging pedagogies being arrived at dialogically, or in consultation with multiple stakeholders about their aspirations for children’s language use. Canagarajah has pointed out that we currently don’t know how participants other than the researcher interpret or respond to these translanguaging displays (Canagarajah, 2011, p. 5). In India, for example, parents and prominent representatives of marginalized, especially Dalit communities have been very vocal in their aspiration and demand for English medium education. Dalit activists have argued that keeping English from Dalits is yet another form of structural oppression; and that English, however poorly taught is a key avenue for resistance and transformation of social conditions for Dalits (e.g., Ilaiah, 2011; Chandra Bhan Prasad, cited in Pandey, 2011).
Otheguy et al. (2015) acknowledge that named languages have a real existence in the social world:
We recognize that these notions have had, and continue to have, real and material consequences in the lives of many people, especially in the lives of minoritized people and those who engage with us in advocacy on their behalf and their ways of speaking. (p. 293) To repeat, there is nothing inherently wrong with the categories of named languages. On many occasions and with many interlocutors, it may make sense to talk in those terms. Such talk is perfectly acceptable so long as we remain aware that the named language categories have been ultimately constructed throughout history for social purposes that bear a well-known and well-documented connection to the imposition of political power. (p. 298)
If we were to accept the position that there is nothing inherently wrong with working with named languages and that we need to recognise the real and material consequences of these constructs in people’s lives, then it would be fair to ask where that leaves us in terms of actual curricular and pedagogical practices. Could we then ask about acquiring competence in a given (monolingual) discursive practice even while acknowledging its socioculturally constructed nature? After all, Otheguy et al. (2015) acknowledge that being able to communicate with audiences of different linguistic backgrounds and being able to behave in linguistically acceptable ways in a variety of social settings (including the settings that devalue translanguaging) are valuable skills, especially for older students. At the same time, they assert that:
Assessing the size, development, flexibility, richness, complexity, and agility of deployment of an idiolect must be kept separate from testing the ability to recognize and adhere to politically defined boundaries in the deployment of the idiolect. The former is a true assessment of linguistic proficiency; the latter is an assessment of cultural and political proficiency. (p. 299 [italics emphasis by the authors]).
As pointed out in a previous section, the portrayal of language as having a ‘true’ linguistic content wearing an overcoat of cultural and political content theoretically reverses decades of research and work in sociolinguistics that reveals the intertwined strands of linguistic, social, cultural, historical, political and material content as the ‘true’ nature of language.
Therefore, it is unclear whether the use of translanguaging in and by itself could create transformative conditions for multilingual speakers, especially those from historically underprivileged groups. For one, pedagogically underdeveloped models may be pushed onto students and teachers (Jaspers, 2017), which is especially problematic given the history of rote and transmission teaching in Indian classrooms. Taking away the support of categories that teachers/practitioners understand (named languages) might actually be counter-productive in such contexts, where, in the view of the authors of this article, the priority in terms of multilingual education is to move curricular and pedagogic practices to a point of welcoming and productively utilizing multiple cultural and linguistic resources in the classroom, rather than of contesting the existence of named linguistic categories.
It is inadequate to target the essentialising of named languages as the primary site of transforming the disenfranchisement of the linguistically marginalised. Ramanathan (2005) in her in-depth study of English- and vernacular-medium textbooks and pedagogic practices in Gujarat, has demonstrated the differences that exist in cultural models for teaching languages in classrooms—from the choice and representation of content to different sets of audiences (English-medium and vernacular medium), to the pedagogic practices employed in each. To this, we could add the material conditions in which teaching and learning occur—the presence of knowledgeable teachers, of pedagogic resources other than textbooks, of ventilation, of light, of space, of functional bathrooms—in spaces where different groups of children learn. We need a complex and multidimensional approach to productively challenge the educational marginalisation of the under-privileged; which would include moving beyond schools and classrooms, to include other lager sites of production and reproduction of linguistic inequities in society.
Therefore, in our view, it might be sound to view translanguaging as a pedagogic framework in the classroom for utilising the multiple linguistic resources that children bring; it could also be used for generating critical consciousness among teachers and students related to language use in society. However, it would be irresponsible and counter-productive to leave children unskilled or under-skilled in conventional language use, especially those who belong to already oppressed or marginalised communities. Jaspers (2017, p. 12) cautions that ‘…translanguaging, in spite of its liberating reputation, can equally become a dominating force, a moral imperative that disqualifies other concerns with language as beyond the pale’. Translanguaging has been especially unable to respond to the concerns of minority language maintenance. Jaspers goes on to state that translanguaging theorists ‘naturalize a specific concern (a non-purist one) with language as the only sensible perspective…’ In presenting minority language concerns in similar terms, translanguaging too seems to emerge as the only rational, ideology-free option, and this is one way in which it can become a dominating force (ibid., p. 13). To promote linguistic hybridity as the only reasonable choice is as dangerous as to promote linguistic purity as that (ibid.).
If Not Translanguaging, What Then?
To summarise the arguments thus far, we believe that translanguaging is an important corrective to purist ideas of language use in several ways. It enables us to view multilinguality as a universal human condition, therefore as an asset and a resource, and not as a problem to be solved. While teaching, it permits us to not view the mixing of languages as ‘errors’, but to understand fluidity as naturally language practice among multilinguals. It does not just permit, but encourages teachers and students to fluidly and intentionally move among named languages. It teaches us to value the diversity and hybridity of people’s language practices in classrooms, to provide space for them, and to generate a critical awareness of language use in society.
That said, we believe that there is a need to go beyond a conceptual critique of named languages in order to find more nuanced approaches to addressing multilingualism in Indian schools/classrooms. For example, children’s languages could be welcomed into the classroom in order to learn a named language. Parikh and Menon (2020) reported potential in using Marathi in third- and fourth-grade Warli-speaking tribal children’s learning of English. Even though children were encouraged to use their linguistic repertoire flexibly in conversation and writing (Figure 1[a] and [b]), the discussion-based pedagogy attended intentionally and systematically to the acquisition of key vocabulary in English. While it acknowledged and encouraged children’s attempts to communicate their thoughts translingually, it also simultaneously drew their attention to conventional forms in English through teacher modelling. Thus, it integrated both function and form into language teaching, rejecting neither. Likewise, it was based on a model of developing competence in the English language, without relying on rote and meaningless learning, or, on a rejection of the children’s linguistic repertoires in the classroom.
Within a complex sociolinguistic conceptualisation of language (that unites the linguistic and the cultural/material), it is more germane to speak of developing competencies in specific discourses and discursive practices, than in any ‘true’ linguistic proficiency. Gee (2013), for example, distinguishes between ‘primary’ and ‘secondary’ discourses. Primary discourses are ‘…ways with words, deeds, things, thoughts, and feelings that, as children, we associate with being a “person like us” (like our socialising group)’ (ibid., p. 61). On the other hand, as people go through life, they pick up a variety of secondary discourses, which are ‘… ways with words, deeds, thoughts, feelings, and things that are connected to various “public” institutions beyond the family, institutions such as churches, workplaces, government institutions, and schools’ (ibid., p. 61). When people function within secondary discourses, they are not acting as ‘everyday people’, but as specialists within that discursive domain. Schools function largely in the realm of secondary discourses. It is undoubtedly true that the secondary discourses valued in schools build upon the primary discourses of some children more than of others; but it is also true that all children learn discursive practices in schools that are quite different from those of their everyday settings. Even a fully translanguaging school will require both the renunciation of certain linguistic practices and the expansion of others (Jaspers, 2017).
For building transformative classrooms, it would be important to imagine the kinds of discursive proficiencies children need to function capably in contemporary societies and economies—and to assess proficiencies in terms of both the ‘breadth’ and the ‘depth’ of sociolinguistic practices that children are expected to participate in within and outside of schools (Luke & Freebody, 1999). It would also be important to ask how to provide children—especially those from marginalized backgrounds—with critical access to the discursive practices of the more privileged, and how to enable them to bridge the gap between their primary and secondary discourses. This does not necessarily mean reinscribing the codes of power, although it carries the danger of doing so. Janks (2004) has argued that access to the higher-powered language(s) with critical awareness can provide opportunities for redesign, instead of re-inscription of linguistic hierarchies. This possibility for redesign need not be interpreted as leaving curricular and pedagogic priorities ill-defined or open-ended. Janks (2006) has described a multilingual, multimodal ‘games’ project through which multilingual South African children created representations of local games for Australian peers. Through the span of the project, children used their everyday multilingual repertoire in flexible, teacher and peer-supported ways to acquire new ‘schooled’ genres (e.g., giving procedural instructions in the written form in English); even while they were able to bring their multiple language varieties, lived experiences and out-of-school knowledges to the tasks of multi-modal text production (video, art and written instructions) at hand. They were also able to view themselves as agentic and powerful, as having something of significance to share with children across the world, children who had traditionally occupied more powerful positions in society. Thus, acquiring competence in specific discursive forms happened along with the possibility of redesigning power relationships—with different linguistic registers and language varieties brought into the design of the task, including the more dominant ones; as well as children seeing themselves and their lived realities as having something meaningful to contribute to more traditionally more powerful audiences.
Currently, many states across India are in a hurry to provide access to higher power languages such as English to all children. But, in doing so without a theory of critical access, they err on two counts. First, as Ramanathan (2005) has pointed out English pedagogy in Indian schools is enacted through very different cultural models of how to be literate-in-English. The content and pedagogy provided to well-to-do English medium schools are much more in alignment with the discursive practices of the Westernised elite—thus, reinscribing relations of power which access to English per se may not disrupt. Second, providing access to codes of power without building critical awareness of language histories and use in society could contribute to the re-inscription of relations of power (Janks, 2004)—with English getting more and more entrenched as the language of power, and specific varieties of it as the discursive practices of the powerful.
Mohanty (2019) has cautioned us of the vicious cycle of language disadvantage, where powerful languages (English, Hindi and regional languages) gain even more power over time, while marginalized (ITM—indigenous/tribal/minority/minoritised) languages get progressively disenfranchised as functional domains for their usage shrinks, sustaining what he refers to as the ‘double divide’—between English and regional languages on the one hand; and regional and ITM languages on the other. Simply letting English or other powerful regional languages loose among the pigeons is thus, not advisable; nor, is simply permitting teachers and students to fluidly move within their unified linguistic repertoire going to disrupt the power and prominence of certain linguistic and discursive practices over others. The challenge for Indian education will be to provide children with a critical access to a variety of discursive practices that contains within them the potential for re-design.
Canagarajah (2011), citing Kirsch and Royster (2010, p. 647), cautions us against the ‘three Rs—rescue, recovery and reinscription’ of suppressed communicative practices. To this, he adds a ‘fourth R’—romanticisation—an uncritical orientation towards marginalized rhetorical traditions (Canagarajah, 2011, p. 3). To navigate between creating functional spaces in schools and classrooms that value a range of linguistic resources and practices, even while avoiding romanticization or uncritical acceptance, is a tall order for policy makers, teacher educators, teachers and students. It might be useful to conceptualise curriculum, pedagogy and teacher education in languages as ‘third spaces’ that help with the ‘…reorganization of the past and present for future psychological functions’—what Cole (1985) calls ‘a dialogue with the future’ (Gutiérrez, 2008, p. 157). It is beyond the scope of this article to elaborate on what these third spaces might look like. More concrete articulations would evolve with a multiplicity of empirical investigations and attempts at theorising located in multilingual educational contexts in India. It is our hope that Indian educators engage in such complex, multidimensional, sociolinguistic investigations of curriculum and pedagogy in the ensuing years and decades. Else, rhetorical celebrations of multilingualism in policy documents will remain empty promises unfulfilled by knowledge, culture or practices that support multilingual education on the ground.
Footnotes
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.
Funding
The authors received no financial support for the research, authorship and/or publication of this article.
