Prologue

I accepted her invitation to sit in on her community English classes, learn how she teaches, and take over the instruction when she retired. I was a newly minted English graduate who had returned to my village from the capital city in Sri Lanka looking for teaching opportunities. Though I had a position as an instructor at the local university, University of Jaffna (UJ), I couldn't decline the invitation of the Pandita because I had to show some community responsibility. Besides, she was a community elder and well-respected neighbor. (Pandita is the title given to those who have achieved mastery in traditional Hindu scholarship.)

Her teaching, in a hut next to her house, attended by about four or five neighborhood children, was different from what I had learnt in my university courses. She brought readings on Hindu epics and scriptures translated into English. She sprinkled her lectures with proverbs, axioms, and vignettes from Hindu/Tamil civilization. As she did so, she switched between English and Tamil throughout the instruction. After about the fourth class, I became restless. It seemed to me that the instruction was not structured nor directed towards any grammar learning goals. I also became tired of the preaching on ethics and culture. It appeared as if the class was focused on moral education, not English. During the fifth class, in the middle of instruction, I rose to leave. The Pandita was surprised. She mentioned that she would probably see me in the next class. At this point, it seemed to me that I had to confront her with the truth: ‘No, I won't be coming again. This is not teaching. This is preaching. You don't have a good pedagogy for English instruction. I don't see any point in these classes.’ She was shocked, but still kept her composure. She simply said, ‘I am sorry that you don't see the value of what I am doing here.’

A few months later she passed away, and her English classes were discontinued. I think of that incident with considerable pain these days, with my revised understanding of the field of applied linguistics and emerging realizations in English language teaching. If I had had my current knowledge of the state of the art, I might have interpreted her teaching differently.

2. Homogeneity as the norm

When I became a teacher of English language at UJ in 1984, I was keen to develop my professional identity. Not satisfied with the single course on structural linguistics in my undergraduate education, I started browsing the books in our department's resource room and discovered a field called applied linguistics that fitted the work I was doing. Reading Pit Corder's (Reference Corder1973) Introducing applied linguistics and Lado's (Reference Lado1964) Language teaching: A scientific approach, I got the following picture (partial, of course) of this discipline: that it was an application of linguistics for language teaching purposes – as in ‘linguistics applied’ (LA) rather than the theoretically more plural ‘applied linguistics’ (AL) – as distinguished by Davies & Elder (Reference Davies, Elder, Davies and Elder2004);Footnote ¹ that the linguistics adopted was structuralist in orientation; and that this orientation claimed to be a scientific approach to language, language competence, and language teaching. The upbeat claims of these scholars on the way this scientific discipline would put language teaching on a firmer footing made an impression on me. The teaching methods influenced by this linguistic/scientific orientation I understood to be shaped by systematic and objective research. While not understanding these methods well, I saw that many of the teaching materials exported to our local context from outside claimed to be influenced by these methods.

Gradually, Chomsky and his transformative generative grammar were also seeping into my department resource room, but his linguistics was too esoteric for us and for our classroom purposes. We continued with business as usual, with some new realizations that strengthened our earlier assumptions. We gathered from Chomsky that we all learnt language the same way, wherever we were located and whichever language we spoke; that learning L2 (second language) adopted the same sequence as learning L1 (first language); that grammar was still key to knowing a language; and that native speakers of a language were the authorities in the language and, therefore, models to be emulated. Though this last point introduced a bias that sat uneasily with the other egalitarian orientations, we resolved the tension by saying that we too could aspire to reach native speaker status.

Looking back, I realize how these early foundations of applied linguistics introduced certain ideologies favoring an objective, acontextual, and instrumentalist inquiry. I label them modernist (see Kramsch Reference Kramsch2014 for an articulation of its influence in foreign language teaching). The implications for language competence were clear. We thought of our competence as primarily dependent on grammar, the deep structure of communication, and controlled cognitively. How this grammar was deployed in practice was an insignificant secondary concern that didn't merit our analysis or instruction.

Because we focused so much on cognition and grammar, we suppressed questions about social and cultural differences relating to communicating and teaching. We adopted different strategies for dealing with diversity in practice. In language teaching, we sometimes practiced our homegrown teaching methods, which were whole language and values based, as in the teaching of the Pandita. Though our students were drawn to such approaches, and we ourselves enjoyed adopting them, we didn't acknowledge or theorize them but made sure that scholars from the West didn't get to see them in any of their training visits to our local institutions. We read with awe the newest iteration of teaching methods that came to our institution with our textbooks as they appeared to make the internalization of form more and more systematic and controlled. That these methods and textbooks didn't have a place for our local languages didn't bother us too much. We believed in Lado's (Reference Lado1964) contrastive linguistics, and assumed that our languages would interfere with the learning of English.

A particularly troubling matter was our Sri Lankan English. Though we used it all the time outside the classroom, we tried to keep it away during instruction. Some teachers would mimic the British accent, which they considered superior to American English. Though it was difficult to get students to adopt this English all the time, for set performances such as speech and essay competitions we expected the winners to be exemplars of the British norm. Sri Lankan English didn't have a name those days. We simply called it ‘broken English.’ Since we conceived of grammatical form as abstract, there was only one English. What we taught was the capacity to use English grammar, which we considered adequate to help our students communicate whether locally or abroad.

It was in this context that something happened to disturb my beliefs and practices in a professional development conference in the capital city in 1985. I listened to some British and American scholars talking about recent psycholinguistic approaches to language acquisition, astounding us with new terms and theories. At the very end of the conference, the organizers had scheduled an Indian professor for a plenary, obviously a tokenist move to assuage our Asian sensibility. But Professor Braj Kachru went on the attack and slaughtered what he called the ‘sacred cows’ of applied linguistics and language teaching. After questioning the emulation of native speaker norms, he made a case for why our local varieties of English were not only functional, but also grammatical. He laid out the notion of World Englishes (WE), making a case for the diversity of English as it spreads globally. I remember a British scholar seated next to me trying to sound me out on my response to Kachru. As the applause was dying, he asked me, ‘So what do you think of that talk?’ I said, ‘It is revolutionary. He shows why we should take our local English seriously.’ The scholar responded, ‘It sounds great in theory, but doesn't have much implications for language teaching. So what are you going to do now? Teach Sri Lankan English to your students? Do you even know the grammar of Sri Lankan English? Do you have materials for that? And what will your students do when they travel outside your country? How will they communicate with others?’

3. Different but equal

To appreciate Kachru's (Reference Kachru1986) perspective, I had to turn to sociolinguistics. I started looking for books on sociolinguistics soon after the conference, fascinated by the social connection of language, which the LA orientation to applied linguistics had so far failed to develop. It was when I traveled to the US for graduate education in August 1985 that I had the opportunity to understand the implications of Kachru's perspective to my profession more clearly. I discovered that there were other ways to deal with my local English and teaching styles. Though Labov's (Reference Labov1969) work on the logic of nonstandard English and Hymes’ (Reference Hymes1966) orientation to communicative competence had been already published in the late ’60s and early ’70s, the implications were only now being debated in my graduate level classes. These debates were ushering a relativistic orientation, whereby we appreciated the systematicity of each dialect and discourse, without judging them from privileged norms.

Both Labov and Hymes retheorized language competence to make a space for social context, modifying the dominant abstract and mentalist notion of competence. What Chomsky relegated to performance started receiving more significance now. Labov's orientation to language variation demonstrated that the abstract mental grammar was useless without a social understanding of how that grammar found variable realization in different contexts and communities. Hymes's notion of communicative competence further socialized cognition. It showed that recognizing diversity, functions, and contexts was integral to one's ability to communicate. I understood communicative competence as demonstrating that grammar outside social context is meaningless. Once we consider context, there are additional norms relating to performance that we had to think about. There were conventions about how grammars should be used, and how grammars found their place in texts and talk. In this sense, there were norms beyond the syntactic level. Sociolinguistics thus sparked my interest in pragmatics and discourse analysis. I understood how it is possible for someone to use ‘standard English’ with discourse conventions differing from those of ‘native speaker’ communities. In drawing from fields like anthropology and sociology, sociolinguists also developed a more interdisciplinary applied linguistics orientation to the field.

What were the pedagogical implications for the writing classes I taught as a graduate assistant in the US at that time? To begin with, I realized that the English varieties minority and international students brought to classrooms shouldn't be stigmatized. They had a right to exist in communities and contexts where they were appropriate. Sometimes, I went a step further and devised activities that started from where the students are (i.e., the grammars and discourses they were competent in) and used them as a bridge to transition them to ‘standard English.’ A third possibility was to enable students to shuttle between their community norms and school norms of English in classroom exercises. This way, I conveyed to them that their home dialects were appreciated by instructors and that we could help them develop fluency in code switching. The National Council of Teachers of English statement on ‘students’ right to their own language’ (1974) sought to educate teachers on the logic of less privileged dialects and make spaces for them in the classroom. The relativistic and liberal orientation of the discipline at this time encouraged teachers to help students communicate differently in appropriate contexts according to relevant norms, in what I referred to as a code switching pedagogy.

Under the influence of what I might label a variationist applied linguistics (VAL), I undertook my own doctoral research. Particularly interesting was the way African American and Hispanic scholars addressed issues of power and inequality in language learning. My primary teaching area of English as a second language (ESL) was treated as instrumentalist and apolitical, perhaps under the influence of the LA approach. Gaining familiarity with the writing and scholarship of Black sociolinguists such as Smitherman (Reference Smitherman1974) and Baugh (Reference Baugh1983), I found a way of merging my teaching with social activism. I had already served as a social worker in many inner city neighborhoods (in South Bronx, Los Angeles, and Washington DC), developing an interest in the concerns of the African American community and its communicative forms. I was interested in studying the implications of their language usage for academic literacy. In these ways, I was continuing my trajectory as a border crosser, intending to learn from the African American experience to inform the postcolonial challenges for my Sri Lankan community.

I was also becoming a disciplinary border crosser and made forays into the curious field called ‘composition’ in the US. This field helped me address issues beyond the grammatical and syntactic level, and consider the applied linguistics of writing and rhetoric. In fact, the orientation to minority students in composition was problematic. As academic conventions were treated as normative, some scholars considered the divergent writing patterns of Black students as demonstrating their cognitive and linguistic deficiency (see Farrell Reference Farrell1978). An inability to write in an objective, linear, and controlled manner was considered evidence of their illogical thought patterns. The difference paradigm would suggest that African American students had their own norms of good writing, and that they may differ from those of the academy. I developed my vision of a suitable pedagogy around the metaphor of code switching, as articulated earlier. While accommodating the logic of Black students’ vernacular codes, I wanted to explore how they shuttled between community-based and academic discourses, and the attitudes and motivations that may or may not facilitate this code switching activity. Understanding this practice of shuttling would help me develop more effective pedagogies and transfer them to ESL students who similarly shuttle between languages, dialects, and discourses in their writing.

4. Different and unequal

What transpired from my research, however, made me question the relativistic ‘different but equal’ orientation to competence and language norms. I did find that the African American students brought a different orientation to written discourse, understood how their preferred discourses differed from the academic discourse, and could switch between both as necessary. However, they also gradually resisted my pedagogical approach of treating ‘standard English’ as the norm for academic contexts. It appeared to me that they were leaning towards an alternative option for their writing and pedagogical challenges (see Canagarajah Reference Canagarajah1990).

For my students, the separation of Black English for home contexts and ‘standard English’ for academic contexts suggested inequality. Their ethnic variety was implicitly deemed unsuitable for institutional, high stakes, and gate keeping contexts. Furthermore, they were not satisfied with keeping their own norms away from other contexts of communication. Though they approximated ‘standard English’ in the early stages of the course, I found their writing becoming perfunctory as the course proceeded, when they found the requirement to keep their preferred grammars and discourses away from their academic writing somewhat alienating. They seemed to be silently seeking ways to merge their communicative resources into academic writing.

This teaching experience introduced considerations of power into discussions of difference. The liberal position of VAL ignored why certain varieties were privileged for formal and institutional contexts, with others relegated to informal and home contexts. Acknowledging the equality of all languages and dialects sounded nice and accommodating in theory, but raised questions in social practice. The good intentions and pronouncements of linguists were not adequate to challenge structural and pervasive social inequalities. Despite the well intentioned policies and pedagogies of VAL, prevailing social inequalities spelled that African American Vernacular English would be treated as inferior and speakers of that dialect subject to social discrimination.

It was also clear that my students were concerned with issues of voice. How can they communicate in the academy or other institutional contexts with their values and identities intact, rather than relegating them for domestic and community contexts? This question challenged another key assumption of VAL: i.e., ‘one language = one context.’ The separation of texts and contexts to feature only one code at a time revealed a reductive orientation that failed to accommodate the inherent diversity and fluidity of language. Contexts, codes and communities were compartmentalized and separated. The students seemed to seek ways of merging the resources and values of one context into another.

Based on my teaching and research experience, I considered appropriation (Canagarajah Reference Canagarajah1999) a better way to address power in language diversity. That is, students from multilingual backgrounds will ‘adopt’ standard norms for their own purposes, according to their own values. Also they would insert into the established norms and conventions their own values and codes. I considered texts and talk polyphonous to accommodate such diversity. Theories of those like Bakhtin (Reference Bakhtin1986) had prepared me to consider the inherent hybridity of language and writing. I gradually came to call this approach code meshing, in opposition to the approach of code switching I had started with (Canagarajah Reference Canagarajah2006a). To facilitate this pedagogy, I made students familiar with their own norms and that of dominant contexts, and encouraged them to take ownership over their language products. I motivated them to relate their language learning and use to the development of an empowered identity, orientating communication to their social and material wellbeing. I enabled students to also question the underlying norms and values behind their textbooks and dominant discourses. Students were encouraged to develop a critical awareness of how language norms were constructed and how they can be renegotiated to serve one's own interests. Social practice took precedence over form and cognition.

This shift led me to redefine my understanding of language competence. VAL was still considerably influenced by norms, structures, and cognition, despite its attempt to diversify them in comparison to the more homogeneous approach of structuralist and generative linguistics. Though social context received legitimacy now, it was still secondary to cognition. And though variation received importance, it was secondary to the formulaic norms and rules of grammar. Hymes may not have intended it, but communicative competence became codified into alternate norms of contextual variation that were part of one's competence. For many teachers and scholars, communicative competence translated into formulas like the following: ‘use code a with convention b for context c.’ This formulaic orientation to competence defined the ability to communicate in terms of a formal mastery of relevant norms, ignoring contextual negotiation (see Leung Reference Leung2005). In attempts at codifying communicative competence for language teaching, grammatical competence came first. It was then followed by sociolinguistic, discoursal, and strategic competence (see Canale & Swain Reference Canale and Swain1980).

5. Power in difference

A move to teach in Sri Lanka, after my Ph.D. in applied linguistics in 1990, saw me focus on developing students’ identities in English language learning. Treating their values and identities as resourceful, I encouraged students to examine the values informing their textbooks, texts, and discourses. We worked towards finding ways to merge local Englishes, languages, and values in the learning and using of English, in the fashion of code meshing. I found that some of my colleagues in UJ were already practicing such a critical pedagogy without explicitly theorizing it as such. One of my colleagues had developed with his students a critique of the ideology (which he identified as WASP, or ‘White Anglo-Saxon Protestant’) embedded in the narratives and language structures in one of the American ESL textbooks we used (Raj Reference Raj1982). In encouraging this critique, he made students aware of the way language choices are implicated in identity positions, their own desired identities, and possibilities of developing more empowering multilingual identities.

In making these shifts, I found affinity with colleagues in the school of critical applied linguistics (CAL)Footnote ² who made identity the driving force in language learning (see Peirce Reference Peirce1995; Kramsch Reference Kramsch2000; Pennycook Reference Pennycook2001). In the place of a purely grammatical consideration to acquisition, CAL scholars were making a place for learning as identity development (Peirce Reference Peirce1995). Learners’ investment in the language, motivated by their social and material interests, determined their type of learning. Relating language use for social purposes, learners were prepared to shape grammar according to their own needs and contexts, without a mechanical conformity to purported norms imposed by others. While CAL recognized the reality of power differences, it also carved a place for agency to renegotiate power.

I also began to critique the high place given to native speaker identity, hitherto treated as the target for our students. Even well-intentioned orientations, such as VAL, had assumed the native speaker norm as non-negotiable for formal, institutional, and lingua franca contexts. In my publications, I demonstrated how the dominant orientations to styles of learning, theories of acquisition, and pragmatic conventions treated the native speaker as the norm (Canagarajah Reference Canagarajah1999). I promoted the learning styles, values, and conventions of local communities, in an effort to develop a localized and situated approach to knowledge construction in the field. Particularly enlightening was how the monolingual norms of native speakers failed to accommodate the possibility of hybridity in texts and talk (in the fashion of code meshing) that was an intuitive everyday practice for multilinguals (Canagarajah Reference Canagarajah2006a). As a way of resisting the power of the monolingual West, I glorified the multilingual East.

6. Hybridity

However, my location of teaching, research, and scholarship changed soon and revealed some of the limitations of the CAL position articulated above. I moved to the US at the end of 1994, as the ethnic fighting in Sri Lanka paralyzed social and educational life there. From rural Jaffna, I moved to a megapolis in the City University of New York (CUNY). This change of location from the periphery to the center also changed my locus of thinking and enunciation. After living in a homogeneous Tamil community, I had now moved to a highly heterogeneous location, with people from different languages and cultures. While I had taught mostly linguistically and culturally homogeneous rural Tamil students in Sri Lanka, my classes in CUNY had students from more than a dozen language groups, each student bringing a repertoire of languages. As I taught them and conducted research on their learning styles and interests, I also revised my orientations to difference.

Facilitating my new research and teaching were theoretical paradigms related to postmodernity and globalization, motivating more cosmopolitan orientations to difference. Philosophically, postmodernity might be explained as anti-foundationalist. It encouraged skepticism on the possibility of grounding truth or knowledge in scientific facts or ‘reality.’ Scholars adopting this orientation perceived their thinking and knowledge as mediated by language, values, and social relationships. In other words, they considered all knowledge paradigms as constructs that are socially and linguistically shaped. This orientation led to a relentless critique of binaries and hierarchies that were the stock in trade of academic scholarship. They were deconstructed for the values and interests that undergirded them. While binaries such as center/periphery, native/nonnative, monolingual/multilingual, and West/East had to be critiqued for the way the latter term in each pair served as a foil for the former, reversing the binary in favor of the latter was not empowering either. In reversing the power imbalance, we were unwittingly treating the dominant binaries and hierarchies as our frame of reference. This realization revealed the limitations of my formerly critical approach to teaching and research. In celebrating the grammatical creativity and cognitive complexity of multilinguals, I was justifying their competence in terms of how monolingual competence had been theorized previously. Even when I showed how they deviated from monolinguals, I treated monolingual norms as my frame of reference. In doing so, I was failing to consider the uniqueness of multilingual competence or acknowledging how it might have strengths and limitations in its own way.

A more constructive approach is to deconstruct the whole center/periphery, monolingual/multilingual, and native/nonnative distinction. These binaries were too reductive to capture actual language use and learning practice. All such constructs are essentializing. They postulate generalized values, norms, and characteristics to categorize diverse social groups and experiences. For example, one might argue that the native/nonnative distinction doesn't have meaning as linguistic nativity is itself a fallacy. When people are always in contact with diverse languages and cultures, and possess variable proficiencies in different languages, the notion of a native speaker with a homogenous identity and community is meaningless. I started speaking English, Tamil, and Sinhala around the same time in my childhood. I consider Sri Lankan English native to me. Though I speak more in Tamil, I write more in English. My parents too were bilingual in English and Tamil, and always mixed them in communication. Then what is my native language? Similarly, the students in my classes in New York each spoke diverse languages, and inherited different linguacultural family backgrounds, that it was difficult to categorize them as native or nonnative to English.

In response to such orientations, new scholarly developments emphasized the value of hybridity. They presented the possibility that diverse values and languages can cohere meaningfully to give identity to people, texts, and communities. In the place of these constructs being defined in terms of uniformity, it presented the possibility that mixing and multiplicity could still give them identity and coherence. In some ways, this thinking was a reaction to modernity, which aimed to identify the essence of human experiences (i.e., thus ‘essentializing’) in order to unveil the underlying universal laws that explained their logic of operation. Such modernist impulses had led to fixed, stable, and uniform knowledge constructs in many disciplines, thus essentializing identities and communities, and limiting the orientation to diversity. Postmodern orientations celebrated fluidity, fusion, and flux.

One pedagogical response to these developments was to argue for the possibility that multilingual students had the agency to borrow from any language or cultural tradition for their communicative purposes. They were not restricted by their backgrounds or overwhelmed by dominant codes. Vivian Zamel (Reference Zamel1997) articulated this position well in L2 writing, labeling it a transculturation approach.Footnote ³ L2 pedagogies which assumed that multilingual students are conditioned by their own linguistic and cultural background to think that they will be alienated by those of others, or that textual genres for specific academic or social purposes have their own well circumscribed conventions and norms, Zamel considered essentializing. Freeing texts, languages, and communities for transculturation, her approach encouraged students to communicate without being bound by norms or structures. This orientation provided more space for creativity for L2 students. Excited about this pedagogical possibility, I explored how my CUNY students can draw from all their influences to construct coherent academic texts. In fact, this pedagogical orientation seemed to fit better with what I had earlier developed as the code meshing approach to languages. Whereas I had thought of code meshing as simply a negotiation of one's first language (L1) and second language (L2), with a guarded meshing of personal codes into a preconstructed dominant register and conventions, I now considered code meshing as accommodating even more diverse influences in a less inhibited mix of resources.

I dabbled with models borrowed from other fields, such as complexity theory (Larsen-Freeman Reference Larsen-Freeman and Kramsch2002), to accommodate hybridity in the competence of individual learners. Developed in fields such as physics and philosophy, complexity theory articulated an orientation to learning and cognition that accounted for fluidity and fixity, heterogeneity and system, change and stability, in the development of grammatical competence. It thus explained the learner's grammatical understanding as a constantly changing yet orderly phenomenon, thus correcting the equation of system with uniformity in modernist paradigms. This theory challenged models which posited separate systems for separate languages, thus compartmentalizing the mind of the multilingual. Under the influence of these movements, I researched how multilinguals can communicate with all the rich repertoires in their background. I developed the notion of ‘shuttling between languages’ (Canagarajah Reference Canagarajah2006b) to demonstrate how proficient bilinguals moved between different varieties of English, or their multilingual repertoires, in their speaking and writing. I analyzed how an advanced Sri Lankan academic writer switched between the norms of center/English and local/English and local/Tamil in his publishing to accomplish a three-way switch (Canagarajah Reference Canagarajah2006b). Though he switched in recognition of the dominant norms of each context/community he addressed, he also infused the text with strengths from alternate traditions to provide a critical edge to his writing. In this sense, he was adopting the established genres and norms in his own terms to construct hybrid texts in each context. Thus hybridity was not a multilingual alternative for monolingual academic or community norms. All communication was hybrid.

With such shifts in pedagogy and scholarship, applied linguistics was also becoming more international in orientation. There was a readiness to appreciate research from diverse global locations and languages. With the claim to valid knowledge by modernity challenged, there was also a theoretical exuberance in the field. Other theoretical paradigms that promoted hybridity received uptake in applied linguistics. In many ways, they all shared a suspicion of empirical positivism and leaned towards constructivist orientations to knowledge (e.g., Pennycook Reference Pennycook2001; van Lier Reference van Lier2004; Atkinson Reference Atkinson2014; Lantolf Reference Lantolf and Atkinson2011; Duff Reference Duff, Duff and Hornberger2008). It was beginning to appear that Lantolf's (Reference Lantolf1996) call for a hundred theoretical flowers to bloom in applied linguistics was becoming a reality within a decade of his pronouncement. With the importation of such multidisciplinary theories, the field was becoming even more AL in scope, applying broader interdisciplinary and theoretically sophisticated models to address language problems (Canagarajah Reference Canagarajah2006c).

7. Difference as the norm

That was not the end of my professional and scholarly journey. Swept along by postmodernist developments, I encountered some new questions and unanticipated problems. Though accommodating hybridity seems to be democratic, it is not easily achieved. It also masks subtle power differences that can be ignored. It transpired that this approach can sometimes be used to contain diversity in manageable new structures and norms, rather than honestly engage with the challenges and possibilities in difference. Postmodern notions of hybridity dealt with multilingualism as a product, rather than as a form of practice. In focusing on coherent texts and talk being constructed (albeit displaying hybridity), applied linguists were containing the disturbing but also enriching possibilities in difference. Furthermore, in focusing on texts and talk in a product-oriented fashion, they were treating diversity in a static manner (for the distinction between static diversity and emergent difference, see Bhabha Reference Bhabha1994). However, mobility was becoming an important consideration with globalization and migration, and the possibilities for mobile languages and texts were increasing in the context of digital technology. To address difference as a form of practice, in the context of mobility, scholars have to consider other issues such as the production, circulation, and reception of codes and texts, focusing on strategies of negotiating difference.

A particular case where I encountered some of the limitations of hybridity was in my work as editor of TESOL Quarterly (2005–2009). As part of an initiative for providing greater access to multilingual authors from nontraditional settings, I was engaged in mentoring a young scholar, Ena Lee. When the reviewers found her submission promising but needing substantial revision, I worked with Ena to address reviewer expectations. Whereas Ena wrote a reflective narrative on her development as a colored teacher in teaching English to speakers of other languages (TESOL), negotiating the biases against her to question power differences, the reviewers wanted her to engage with the dominant conventions of academic discourses more directly. Though Ena already adopted a hybrid discourse (merging theorization and citations with her personal reflection and narrative), the reviewers considered her discourse too unconventional for the journal. They wanted her to adopt a more objective and analytical approach, and also provide more explicit justification and signposting for readers on her alternate discourse, as a way of entering into the dominant discourses and genre conventions more cautiously. However, Ena considered these expectations as distorting her voice and interests. Ena sensed that she was being pressured to adopt the preference of the gatekeepers in order to get published. She also realized that the dominant genres of academic writing and publishing still reflected the ideologies of modernity (such as the need for objectivity, rationality, linear logic, and empirical evidence). It appeared that her personal narrative was devalued if it didn't satisfy the established norms or made gestures towards accommodating them. Therefore, eventually, Ena declined publication (Canagarajah & Lee Reference Canagarajah, Lee, Thesen and Cooper2014).

This episode taught me that hybridity did not automatically answer the challenge of addressing difference. There were two types of hybridity at issue in this example. Ena's accomplishment of hybridity was considered unacceptable, and the hybridity preferred by the reviewers and editor enjoyed more clout. The notion that you have to achieve a balanced mix of established and alternate discourses means that the established discourses still have the power to compromise the critical edge of difference. Hybridity might be considered a modest reconfiguration of power to accommodate difference on the terms of the status quo.

From this perspective, we can also understand the limitations of competence models that accommodate language fluidity. Though languages could be mixed, cognition provided stability and control. In complexity models, for example, the paradoxical fixity and fluidity of grammar takes place in cognition. The mind is still in control. The difference is that whereas cognition was homogeneous earlier, it has been diversified now. Furthermore, despite multilingualism, cognition is still important for managing this capacity for difference. The individual who earlier managed monolingual communication is now posited to adopt similar strategies to manage multiple languages. Note also that pre-existing languages are taken for granted, as students are allowed to borrow from them for their hybrid communication and shuttling between languages. Thus, fluidity and change don't affect the system of language. We could go on with business as usual, once difference was given a space in the existing structures of cognition and grammar.

8. Translingual practice

Such developments as mobility, digital communication, and transnational relations of more recent times are perhaps inviting new ways of dealing with difference. They point to the need for new paradigms that question the very definition of community, cognition, meaning, and form. Rather than treating these constructs as bounded, shared, and preconstructed, recent paradigms are treating them as constantly reconstructed in situated interactions. In this process, they are treating difference as emergent, rather than accommodating diversity into predictable structures or forms.

This orientation also encourages us to look at how language resources are mobile, traveling with or without people to come into contact with other languages and communities, generating new grammars and meanings (Blommaert Reference Blommaert2010). It is not that there are no grammars; they evolve in practice. As people continue to use their mix of resources for specific purposes, their resources get patterned into grammatical and discourse conventions. In this sense, meaning and form are constantly reconstructed in situated interactions out of the diverse resources people bring to communication.

This way of looking at languages questions the dominant understanding of multilingualism. When we think of multilingualism as separate languages put together, we are considering their relationship in an additive manner. In keeping the languages separate in multilingualism, we are accommodating diverse languages into the dominant monolingual ideologies and structures. Scholars have drawn attention to this irony by calling such forms of multilingualism ‘separate bilingualism’ (Creese & Blackledge Reference Creese and Blackledge2008) or ‘two solitudes’ (Cummins Reference Cummins, Cummins and Hornberger2008). For this reason, I have been recently using the label translingual, to accommodate the new insights on language as a set of mobile resources (see Canagarajah Reference Canagarajah2013). The translingual orientation encourages us to look at communication as going beyond individual languages. It draws attention to the synergy between codes that come into contact, as they generate new meanings and grammars.

While the ‘trans’ in translingual stands for meaning relationships beyond individual languages, it also perceives communication as going beyond words and accommodating other semiotic systems (such as sound, visuals, graphics, body, and ecology) in creating meaning. The celebration of language as the chief means of communication (perhaps more logical and clear than other modalities and semiotic systems) is another recent modernist bias. The need to consider all the resources at our disposal and consider the material and situated nature of communication becomes important when we consider how interactions take place in contact zones (Pratt Reference Pratt1991). When no languages or values are shared by interlocutors, grammar alone cannot help us account for competence and meaning. People use all the resources at their disposal in the local ecology, such as objects, gestures, and the body, for meaning-making. Block (Reference Block and May2014) has critiqued the ‘lingual bias’ in language acquisition models that focuses solely on grammatical competence, excluding the role of the body and other ecological resources.

The translingual orientation treats competence as a form of practice and not a form of knowledge. In this respect, the focus shifts to what people do in communication and not what they know. Practice also removes the focus from the individual as the agent of meaning and considers communicative success as a social accomplishment. When form and meaning are not shared, interlocutors have to collaborate with each other and adopt cooperative strategies to negotiate meaning. From this perspective, cognition receives qualified significance, as we begin to focus on one's procedural knowledge and not propositional knowledge, to use Byram's (Reference Byram2008) useful distinction. In his treatment, propositional knowledge constitutes norms, facts, and conventions of language; procedural knowledge constitutes practices and emerges from practice. The latter resembles ‘strategic competence’ that Canale & Swain (Reference Canale and Swain1980) identified as one of the four components of communicative competence. Strategic competence refers to the ability to anticipate and repair potential communication breakdown in contexts of variable grammatical proficiency among interlocutors. But listed as the final component after grammatical, sociolinguistic, and discourse competence, it has not received the attention it deserves in applied linguistics circles. We also now perceive cognition as shaped by social practice, with emergent new competence and knowledge based on one's ongoing situated interactions.

My recent research with skilled migrants inspires these revised ways of looking at language and competence (see Canagarajah Reference Canagarajah2013). I have been studying how professionals from less developed communities migrate to more advanced English-speaking countries and manage their linguistic interactions. My subjects state in their narratives of professional communication that each of them adopts their own varieties of English for such interactions. They don't shift to a neutral or common variety, such as British English. One person will speak in Indian English and the other in Nigerian English. How intelligibility works is through both parties opening themselves to learning the norms of the other as they speak. This calls for a different ethic or disposition of communication. Rather than seeking refuge in a uniform shared norm, or their own preferred norms, the interlocutors are disposed to engage with the otherness of the interlocutor. They adopt diverse strategies to make communication succeed. I found subjects using strategies such as confirmation check, repetition, and clarification requests to manage possible communicative breakdown due to norm differences (cf., Kaur Reference Kaur2009; Pitzl Reference Pitzl2010). Other resources such as gestures, setting, and objects also help them understand each other. Eventually, the meanings are deeply intersubjective and co-constructed. As they adjust to each other's preferred norms, values, and identities, both parties may end up adopting an idiosyncratic word, even though it may not be part of the lexicon of native speakers. They will achieve communicative success through features that they thus make normal for their situated interaction (which Firth Reference Firth1996 labels the ‘make-it-normal’ strategy). These co-constructed forms and meanings might become new grammars over time. What my research shows is that grammar and community are co-constructed by the participants through their activity.

In research in classroom contexts, I find that both native and nonnative students demonstrate such a disposition for translingual practice. They engage with texts and codes from diverse backgrounds, deploying creative strategies to negotiate meaning and achieve communicative success, which they seem to have developed in lifelong everyday learning (Canagarajah Reference Canagarajah2013). This is not surprising. As all of us are always situated in contact zones, the orientation to translingual practice and competence is similar for both native and nonnative English speakers.

9. A way forward

What my story of the field suggests is that we have gradually moved from treating sharedness and homogeneity as guaranteeing meaning to addressing difference as the norm. We find that the field has had to make other shifts in its definition of language, community, competence, and learning in order to address difference honestly and fully. I would summarize the shifts in the following way:

• • Language as a set of contextualized resources rather than a preconstructed abstract structure;

• • Communication as accommodating diverse semiotic resources in addition to language;

• • Communities as always liminal and constructed through contact, rather than bound homogeneously;

• • Competence as shaped primarily by procedural knowledge on communicative practices rather than propositional knowledge of grammatical norms; and

• • Learning as developing suitable dispositions for translingual interactions rather than focusing primarily on knowledge or skills.

If we are to engage with difference in this manner, we have to deal with many new questions going forward. Scholars are exploring how we can redefine a multilingual competence for situated negotiation of codes as social practice rather than cognitive capacity (see Block Reference Block and May2014; May Reference May2014). There are tremendous implications for pedagogy as we consider how we can encourage students to engage with diverse semiotic resources in global contact zones for communicative success (see Kramsch Reference Kramsch2014). There are also policy implications as we consider how we can make spaces for diverse language repertoires in the nation state, workplace, and schools (see Duchene & Heller Reference Duchene and Heller2012).

While these efforts continue, the nature of the field is changing. It seems as if the field is moving from LA and AL to something different. We are not applying knowledge from descriptive linguistics to language teaching (LA) or borrowing from diverse disciplines and theories to solve language problems (AL), but positioning ourselves in a more agentive and mediating role. To begin with, we are not in a dependent status, only applying linguistics or other disciplinary insights to our professional concerns. We are theorists who can make our own contribution to other fields. The research I have been doing on skilled migration contributes to scholarship in geography, sociology, and labor relations. It sheds light on how workplace communication takes place in the context of transnational production and development efforts. Since we also know that diverse social domains and experiences are linguistically constructed (with language gaining a constitutive role, and not just a secondary regulative function), applied linguists are in a position to offer insights into broader social processes. To perform this function effectively, we have to position ourselves between language and other social domains to consider the implications for both sides of the connection. Returning to our traditional concern with teaching, we also find that interactions outside the classroom reveal valuable insights about learning (as in the way skilled migrants learn in the workplace) and shape the types of functional and hybrid communication that needs to be addressed by the school. This positioning between language, pedagogy, and diverse social and theoretical domains to mediate knowledge and interactions for multidirectional disciplinary contributions provides applied linguists a more agentive and critical role.

Coda

I now return with new eyes to my experience with Pandita's teaching. I realize that what she was trying to develop among her students are translingual dispositions, not grammatical or communicative skills per se. A suitable language awareness, negotiation strategies, and learning practices would prepare her students to negotiate diverse languages in their social ecology in lifelong learning. It is for this reason that Pandita mixed English and Tamil, and addressed ethical dispositions. She was not focused on developing English competence alone, but a translingual awareness. Coming from a multilingual and multicultural locale, she was attempting to develop in her students a readiness to achieve meaning and communicative success through adopting suitable dispositions. She was also developing this in a manner that the students would be grounded in their local languages, identities, and values. It is possible that Pandita was drawing from traditions of communication, learning, and teaching from local history, where translingual practices have been a fact of life for several generations. Linguistic and social life were arguably more diverse in precolonial times rather than in modernity. There are secrets in treating difference as the norm for communication and learning from precolonial communities that our field has yet to tap into. The hubris that applied linguistics is modern, scientific, and efficient has to be tempered by the realization that successful multilingual communication and teaching have been going on for centuries around the world and that the practices ancient people adopted are not lacking in wisdom.