The multilingual turn as a critical movement in education: assumptions, challenges and a need for reflection

2.1 Language

I identified four strands of developments in the literature: cognitive, integrated, socio-critical and multimodal understandings of language, which I will discuss in this order.

Structuralists understand language as discrete monolingual and separate structures that can be studied (Saussure 1966). Based on cognitive understandings of language, grammar is seen as a system that can be processed by humans (Chomsky 1959; Pienemann 2008; Towell and Hawkins 1994) or as a pattern that humans build in their mind (cf. connectionism, emergentism), and based on usage of language (Ellis 2006; Tomasello 2003). In these views, language is conceptualised largely mono- or intralingually as a stable representational system that exists independent of the people who use it.

Literature suggests a new integrated/crosslingual way of looking at language started around the 1980s, based on the much cited, and at the time visionary, interdependence hypothesis (Cummins 1979), arguing that bilinguals have an underlying integrated language proficiency rather than separate monolingual competencies. This hypothesis, which has since been accepted based on neuroscience and psychology research, confirmed, for example, that multilinguals have an integrated multilingual lexicon (Kroll et al. 2013; Lowie et al. 2014) from which they draw for all communication, placing language in the minds of users. In sociolinguistics this is echoed by languages understood as integrated in the form of a language repertoire (Blommaert and Backus 2011). Complexity theorists understand language “as a complex adaptive system, which emerges bottom-up from interactions of multiple agents in speech communities” and thus as a “dynamic set of patterns emerging from use” (Larsen-Freeman 2011:49) that constitutes a “system [that] is in constant flux” (50). This understanding posits that languages are integrated in the mind, dynamic and constructed through social interaction.

The notion of bottom-up and more socially and critically informed understandings of language, as conceived by sociocultural and post-structural theorists, have had a series of followers, as language is seen as a tool for communication (Halliday 1985), as a dynamic tool for social action (Atkinson 2011a: 146) and mediation (Lantolf 2011). Thus language is not seen as separate from people, but constructed through a local practice (Pennycook 2010) through which “relationships and identities are defined, negotiated, and resisted” (Norton and McKinney 2011: 77), as well as a practice that occurs in contact zones (Pratt 1991) where languages meet. Thus languages are no longer seen as belonging to just one place but are seen as translocal and deterritorialised (Blommaert 2010). Whereas previously languages were thought to ‘belong’ to the people who speak them as native speakers (Wright 2004), more recent user-based understandings of language reject the idea of ownership based on birth and propose dynamic and developmental language models, where language ownership can be acquired later in life (Canagarajah 2013; Jenkins 2006). Thus, from a critical perspective, languages and literacy are not neutral but inherently political and a site of struggle (Sarroub and Quadros 2014).

A further movement led to the understanding of language, as part of a multimodal repertoire, as an “intricate web of social meaning woven from grammar, intonation, gaze, gesture, head movement, bodily orientation, and additional semiotic resources” (Atkinson 2011b: 147), and as socially constructed based on ideology and power as “a web of interlinked socio-political and historical factors that shape one’s identity and voice” (Kumaravadivelu 2005: 72). Dynamic systems theory makes a useful contribution by conceptualising languages as crosslingual systems, including dialects, genres, registers and discourses, as “being patterned and structured while being open to change and further pluralization” (Canagarajah 2013: 31).

In sum, an alternative understanding has developed over the last 35 years or so, which views languages not to exist separately from one another, nor separately from the people who use them, nor from the wider social and political framework, but as an integrated, crosslingual, dynamic and multimodal semiotic system, as translocal and mobile resources owned by those who use them to engage in practice in a contact zone.

2.2 The learners

Early behaviourists conceptualised learners as empty vessels or trainable through external stimuli (Skinner 1957). The cognitive turn, importantly, reconceptualised learners as people who have a cognitive and creative capacity to process environmentally available input. In this view, the learner’s mind and the architecture of the brain constituted the main interest as part of a learning theory (Myles and Mitchell 2004). More socially-informed conceptualisations of the learner (e. g. Block 2003; Firth and Wagner 1997), see learners as social actors who construct meaning and new knowledge in social interaction with others, based on Vygotsky’s (1978) ideas (Lantolf and Appel 1994), in interaction with the world (Norton and McKinney 2011), as agents who “play an active role in language development” (Larsen-Freeman 2011: 55) and in turn shape and are shaped by their environment (Atkinson 2011a; Larsen-Freeman 2011). As will be seen below, the notions of social practice, agency and environment are important to understand the multilingual turn.

The idea that individual differences (e. g. motivation, ability) can be viewed as “stable and monolithic learner traits”, which is associated with cognitive perspectives, has been criticised (Larsen-Freeman 2011). This led to a major departure from behaviourist/cognitive understandings in two ways: 1) towards learner heterogeneity and variability, emphasising “[t]he unique local particularities of the person as self-reflective intentional agent, inherently part of and shaping his or her own context” (Larsen-Freeman 2011, based on Ushioda), and towards affordances and relationships. Based on this, researchers became interested in “how a person understands his or her relationship to the world, how that relationship is constructed across time and space, and how the person understands possibilities for the future” which are associated with identity (Norton and McKinney 2011: 73), based on post-structural and critical conceptualisations. In a review of learner conceptualisations over time, Kibler and Valdés (2016) showed that learner categorisations are not neutral, and the way we talk about learners matters. Thus, identity development is also seen as a site of struggle, as it is socially negotiated and “constantly changing across time and space” (Norton and McKinney 2011: 75).

More recently, theoretical pluralists have proposed socio-cognitive (Atkinson 2011a; Larsen-Freeman 2007; Zuengler and Miller 2006) and ecological (Pennycook 2004; van Lier 2004) understandings, pointing towards reconciliation of theoretical perspectives. The latter sees the learners as “ecological organisms” who “make decisions to deploy language resources to realize transactional, interpersonal, educative, self-expressive, etc. goals and the multiple dimensions of self and identity, affective states, and social face” (Larsen-Freeman 2011: 58), and who “depend on [and shape] their environment to survive” (Atkinson 2011a: 143). Thus, according to Atkinson, mind, body and the world are interactively connected. To conclude this section, I argue that crosslingual alternative views of learners have developed over time.

2.3 Language learning

Perspectives on learning are based on classical, behaviourist, cognitivist and social conceptualisations. Power and identity approaches have broadened the field, proposing more holistic and complex, critical and post-structuralist understandings.

The classical method was based on the understanding that languages need to be studied through the analysis and translation of texts (Cook and Singleton 2014), modelled on the study of Latin or ancient Greek in Europe. Relevant pedagogic methods include the grammar-translation (see Howatt and Widdowson 2004), contrastive-analysis and error-analysis (Gass 2013) approaches that are associated with an emphasis on the correct production of structures and form, based on detailed, crosslingual comparison between languages. These approaches may seem outdated to some, but they are arguably crosslingual and empower the bilingual teacher, as an expert bilingual role model.

Behaviourist views of second or additional language learning have parallels with training animals through environmental stimuli (Skinner 1957). Pedagogic implications of this are that “practice makes perfect” through reinforcement, repetition, imitation and reward (Myles and Mitchell 2004: 31), which was the foundation of the largely monolingually conceived audiolingual approach (Cook and Singleton 2014). Such approaches to learning are what Kumaravadivelu (2005: 90) describes as language-centred approaches, which “treat learning as a linear, additive process” that can be planned and structured, and that is to a certain degree predictable. The task of the teacher, in this view, is “to introduce one discrete linguistic item at the time, and help the learner practice it” (2005: 90).

Cognitivist or psychological views of second language learning are based on the information processing metaphor, suggesting that learners process, develop patterns and make sense of input (Davies and Elder 2004; Krashen 1985) that is provided by the environment. In this view, learning is associated with learning-centeredness, in which learning is seen as unpredictable, and which “requires the creation of conditions in which learners engage in meaningful activities in class”, and with learner-centeredness “based on the learner’s real-life language use in social interaction or for academic study” (Kumaravadivelu 2005: 92). Pedagogies related to this understanding of learning include communicative language teaching (Howatt and Widdowson 2004), immersion education (Lasagabaster and Sierra 2007), task-based learning and teaching (Ellis 2003), and other concept-based teaching methods. This approach is usually based on monolingual and intralingual pedagogies (Kumaravadivelu 2005), modelled on first language acquisition. It is often associated with the native-speaker aim for second language learners, and with the idea that native speakers are best placed as language models and teachers. This preference of native speakers is a practice that is particularly wide-spread in the world of English language teaching globally.

Socio-cultural and post-structuralist understandings assume that language learning is based on co-construction of meaning, focussing “on the construction of interpersonal interactions where participants actively and dynamically negotiate not just textual meaning, but also their social relationships” (Kumaravadivelu 2005: 70). In this conceptualisation of learning, it is not just the teacher that facilitates learning but also peer-learners. Socio-cultural theory, based on Vygotsky (1978), emphasises support/strategy instruction (scaffolding) so that the learners can move towards autonomous learning (self-regulation) and fulfil their potential with the help of others (zone of proximal development) (Lantolf 2011). This conceptualisation assumes that learners have different learning trajectories and backgrounds, and thus learn in different and unpredictable ways based on socio-constructivism. The role of the teacher is therefore to assess individual needs, provide scaffolding and provide formative feedback or dynamic assessment. As this perspective is open to the use of other languages, such as the first language, to support additional language learning (Moore 2013), I argue that this pedagogy could be described as potentially but perhaps not explicitly crosslingual. Based on their understanding of language as a social and political construct, post-structuralists (e. g. Duff and Talmy 2011; Norton and McKinney 2011) and critical educationalists (e. g. Kumaravadivelu 2005) go a step further in that they not only understand learning based on interaction, but also on the power-relations inherent in relationships of learning. Teachers must therefore not only provide learners with suitable content and conditions for learning (context), but also think about identity formation and social transformation, under the umbrella of empowering education (Kumaravadivelu 2005).

More recently, “ecological accounts of learning” that are interested in the intersection between individuals and social interaction perspectives (Larsen-Freeman 2011: 66) have found increased attention, claming that “mind, body, and world function integratively in second language acquisition” (Atkinson 2011a: 143). This intersection approach, combining different perspectives, or “holistic approach” (Larsen-Freeman 2011: 67) to language development, requires locally relevant pedagogies (Kumaravadivelu 2005), positing that teachers should have an awareness that “language, its use, and its acquisition are mutually constitutive” (Larsen-Freeman 2011: 49). In addition to this, language socialisation theorists recommend that “greater attention [is paid] to L1, L2, L3, etc.” and the roles of these in particular contexts (Duff and Talmy 2011: 111).

Crosslingual pedagogic approaches are not a recent phenomenon. Some are clearly based on the cognitive paradigm: grammar-translation (Howatt and Widdowson 2004), contrastive analysis (Gass 2013), intercomprehension (Carrasco Perea 2010); and some guided by socio-constructivist, critical or plural paradigms: plurilingualism ^[1] (Council of Europe 2001), third-language education (Cenoz 2012), use of L1 in L2 education (Cook 2001; Scott & de la Fuente 2008), content and language integrated learning (Mehisto et al. 2008; Meyer et al. 2015), translanguaging (García & Leiva 2012; Williams 2000) and translingual practice (Canagarajah 2013).

2.4 Summary of the literature

In this section I have confirmed that new crosslingual understandings of the three concepts have developed prior to the multilingual turn. Some theorists continue their work guided by cognitive and largely intralingual perspectives (e.g. Gass 2013), while others now work in a more critical multilingual or crosslingual paradigm (e. g. Canagarajah 2013; Cenoz 2012). Furthermore, it shows that crosslingual understandings of languages and learners are a more recent phenomenon, while crosslingual understandings of learning – and teaching – may not be new, but they have changed over time. More recent crosslingual understandings consider languages as dynamic in the way that they shape and are shaped by language users, and that languages are integrated in mind, body and world, and do not exist separate from these. They further consider language use and learning to be based on social practice. In the present study, I will flesh out such cross-lingual understandings by an in-depth analysis of the first two books (Conteh and Meier 2014; May 2014a) explicitly associated with the English phrase the multilingual turn.

4.1 Conceptualisation of language

Interesting findings in this section report the way authors conceptualise/describe languages in different ways, how they all relate language to status and/or power, see it as a resource, and as conceptualised multilingually. The main themes identified (Table 1) will serve as headings, and the numbers in the table indicate in how many chapters each dimension was mentioned. This format is also applied to the following sections.

4.2 Conceptualisation of the learners

Important findings discussed in this section are: labels used to describe learners; learner identities; learners as unique and complex; as agents; as language users and multilingual practitioners; as shaped by society; and as participants in a multilingual world. Additionally, Anderson and Chung (2014) argue that the way we understand the learner needs to be reconceptualised.

4.3 Conceptualisation of language learning

All authors subscribe to the critical paradigm, which, however, does not stop them from drawing on a range of perspectives, in their moves towards multilingual understandings of learning (Table 3).

4.4 Challenges related to the multilingual turn

Some authors have explicitly referred to ‘challenges’, which is an inductively identified theme. In six of the chapters, authors expressed concerns about widespread monolingual myths, hampering wider application, and five observed a lack of support for practitioners in schools. These findings resonate with Weber’s (2014: 186) argument that there are “major pedagogical and attitudinal obstacles” in the way of implementing the more flexible multilingual education he proposes. Based on this, I developped two inductive themes to deepen our understanding of both the attitudinal and the pedagogic challenge.

5.1 Conceptualisations associated with the multilingual turn

One of the contributions of this article is that it makes visible the ontological foundations of the multilingual turn and links this to antecedent literature. At the same time this offers an answer to the research question, as follows.

In the chapters languages are conceptualised as integrated in mind, body and world, and above all as transnational mobile resources for learning and teaching, and as a multilingual social practice. These largely consistent findings confirm developments observed in the field towards an integrated view of languages, as previously proposed by relevant authors (Canagarajah 2013; Jenkins 2006; Atkinson 2011b; Kumaravadivelu 2005). Language thus understood bridges linguistic and educational disciplines as described above.

Related to the above, learners are constructed as multilingual social practitioners and agents with dynamic and complex biographies and identities who exist in a multilingual ecosystem. These findings reflect research interests developed in recent years on user-based linguistics (Ortega 2014), identity (Norton and McKinney 2011; Freeman 1994); language socialisation (Duff and Talmy 2011); and ecological perspectives (van Lier 2000). In addition, user-based linguistics is linked to the idea of ownership of language, steering away from native-speaker, monolingual, and standard language norms (Jenkins 2006; Canagarajah 2013), as well as from territorial views of language (Blommaert 2010). However, there is some inconsistency in the chapters in that language backgrounds are sometimes used to categorise learners, as part of a certain community (e. g. minority or majority), while others steer clear of such categorisations. Essentialised representations or labels are seen as problematic (Sarroub and Quadros 2014; Kibler and Valdes 2016) as this can reify certain deficit identities. Thus the possibility view of emergent bi/multilinguals may be more useful, or the notion of developing linguistic repertoires.

Learning as theorised in the chapters is consistently based on the idea of empowerment and learning as a multilingual practice, based on critical perspectives, possibly, at least in parts, based on editor bias but nevertheless endorsed by the 30 authors in the two books, as well as previous literature (e. g. Atkinson 2011b; Cenoz 2012; Cummins 1979, 2007). Besides this, there is wide-spread agreement that socio-cultural theory is a useful perspective to understand learning. Indeed the latter is seen as a constituent or precursor of the multilingual turn (May 2014b). However, there is no unanimity, and authors in the sample argue that cognitive perspectives are either not useful, or that they need to be revised to accommodate the multilingual turn. Learning could therefore be understood based on theoretical pluralism. This enables researchers, who associate themselves with the multilingual turn, to assume different ontological positions, and draw on revised forms of behaviourism and cognitivism, as well as socio-cultural theory, complexity and ecological perspectives, guided above all by critical understandings. This is not necessarily a contradiction; on the contrary the multilingual turn could be described as inclusive, proposing or at least tolerating, plurality, unlike the previous cognitive and social turns that tended to reject previous understandings of learning. Thus, the multilingual turn may be less divisive along theoretical, but perhaps more divisive along ideological lines, such as between those who welcome and accept more plural societies, and those who may see these as a threat. In the current climate of uncertainties, such as climate change, economic downturns, political changes, refugee crises and terror threats, multilingualism may be seen by some as a further destabilising force, and be one of the reasons for the perpetuation of monolingual and standard language norms by stakeholders in schools and society. To this end the thematic decomposition analysis enabled me to uncover the ontological foundations, the political and ideological nature of the multilingual turn and associated societal challenges.

5.2 A framework for critical reflection

Above, I have shown that there is an over-arching critical, and arguably ideological, theoretical perspective associated with the multilingual turn, which questions widely held assumptions about language, learners and learning, as well as the nature and location of knowledge, social relationships and power; all concepts associated with critical pedagogy (see McLaren 2013; Norton and Toohey 2004).

Teachers, learners, parents and schools may make linguistic decisions that are shaped by widely accepted monolingual and standard language norms and prevailing ideologies. Marxists might refer to this as false consciousness, suggesting unreflected acceptance of unequal power relations. In this article, I show that different ontological stances, or understandings of language, learners and learning can thus lead teachers and schools to perpetuate, resist or challenge current norms. Whatever ontological and ideological positions educators assume, conscious reflection of these concepts may contribute to understanding the consequences of these. Based on this, I invite (student) teachers, teacher educators and researchers to reflect with their students, colleagues and/or participants on their own local practices at societal, school, curricular and individual level and query their prevailing assumptions, associated consequences, such as participation, power and knowledge, and potential alternatives. The framework in Table 6, which is informed by this study and related literature, is designed to inform such a reflection. The left-hand side of the table is loosely guided by traditional, monolingual assumptions, and the right-hand side is guided by critical, multilingual assumptions. However, I argue that there is no right or wrong, as most linguistic choices are compromises based on complex understandings that might encompass both sides. However, being conscious of these assumptions is of critical importance as this may lead to different expectations, actions and power distributions in classrooms and society more widely.

This framework could act as a starting point to critically reflect on a particular situation, using the concepts shown in Table 6, and examine our understandings and attitudes of the potentially restrictive or emancipating effect that monolingual and multilingual approaches may have at different levels, to enable all stakeholders to engage in judicious, sensitive and conscious linguistic practices to jointly make sense of the increasingly multilingual world, and perhaps weaken the monolingual myths in the long term. This would require, however, a reconceptualisation of teachers as multilingual role models, who make visible their linguistic repertoires, including standard and non-standard language varieties. Having said this, I would like to add a note of caution: a multilingual approach to learning and teaching does not advocate a laissez-faire attitude, where learners and teachers use any languages they like, as this may neither be useful for learning nor empower learners (Moore 2013).

The ability to engage with others to make sense of the world and to solve problems in a creative and multilingual way is bound to be an important skill necessary to tackle global and local problems, and may offer an empowering experience, especially if the entire language repertoire, and a more holistic identity, is validated and considered a useful tool for learning and being. From this it follows that any development of relevant multilingual pedagogies ought to be based on joint reflections and revised understandings, and, as Pennycook (2004) points out, to consciousness raising, which is the basis of social action and change. In sum, this article shows that the multilingual turn forms part of a wider critical and transdisciplinary movement in education. While it is relatively well theorised, its translation into practice is challenging, due to widely held monolingually informed attitudes. Multi-agency work is now required to develop greater consciousness in mainstream practice. The framework (Table 6) is designed to contribute to such a process, and I would be very pleased to hear in what way and to what effect readers use this framework in their varied practices.