From monolingual mindset to plurilingual ethos: challenging perspectives on language(s)

ABSTRACT

Scholars have long critiqued points of view in which monolingual perspectives are seen as normative in research on multilingualism. In relation to this monolingual orientation, however, in which monolingualism is perceived to be the implicit norm, less work has been dedicated to methodological challenges. As disciplinary perspectives on language in Linguistics and related fields move further towards the de-emphasizing and deconstructing of the boundaries between named linguistic varieties, this paper addresses some issues that come up in operationalizing this in our research. It does so through self-reflexivity, from a primarily Sociolinguistic and Applied Linguistic point of view, by addressing monolingual perspectives found in data on a project on multilingual practices in India. It focuses more narrowly on the ethnographic field notes from the research context. With discourse analysis, these are looked at through the lens of the monolingual orientation, with a particular focus on language ideologies and the compartmentalization of named-language varieties. Ideologies in the data are discussed in relation to their bearing on methodology in transcribing, coding, and the framing of research questions. This paper explores the tensions between evolving theoretical perspectives and on-the-ground research practice, concluding by proposing questions for reflection.

KEYWORDS:

• Monolingual orientation
• language ideologies
• multilingual practices
• ethnography
• India

:

• 

Critiques of monolingual normativity in perspectives on multilingualism have been present in academia for some time, but methodological challenges related to it have received little scholarly attention. Taking Sociolinguistics and Applied Linguistics as a point of departure, this paper deals with methodological issues through self-reflexivity and a focus on language ideologies related to the compartmentalization of named-language varieties. It does so as a means to consider how the perspectives researchers hold on language(s) and multilingualism as objects of study can color the lenses through which we look.

The wider ethnographic project in India, from which this discussion is drawn, is situated within interactional sociolinguistics and deals with multilingual practices through the constructs of style, stance and generational identities in a context of social change. However, the plurilingual ethos at the research site invites reflections on named-language categories as constructs, and this paper focuses more narrowly on the project field notes. They are examined through the perspectives of the monolingual orientation (Canagarajah 2013) and the plurilingual ethos (Khubchandani 1997, 1998), with a particular focus on language ideologies and the compartmentalization (Khubchandani 1983, 1993, 69) of named-language varieties. Language ideologies in the data are discussed in relation to their bearing on three areas: transcribing, coding, and the framing of research questions, and the analysis reveals how underlying ideologies of language (a monolingual orientation, in this case) can infuse research practice.² It thus apprehends some of the methodological issues related to the monolingual orientation, as encountered in a PhD research project in India carried out by the author of this paper.³

Following Irvine (1989), Woolard (2021) defines language ideologies as ‘morally and politically loaded representations of the nature, structure, and use of languages in a social world’. She further highlights that ‘[i]deological representations of language(s) are enacted by ordinary community members as well as official institutions and elites, including academic scholars’ (1). It is this latter category of academic perspectives on which the present article is focused, as it explores ways that scholars view language(s). Because these perspectives are not easily separable from our research practice, the paper underlines the importance of self-reflexivity in light of evolving views on how we understand language and named-language varieties with category labels (such as Gujarati, Hindi and English) in multilingual research. The paper moves from observations at the operational level of transcribing and coding, to noticing that these stem from wider perspectives on language(s) that underpin the framing of research questions. It then concludes by proposing questions for reflection.

Table 1. Questions for reflection.

Conceptualization of language: How do I conceptualize language in relation to the research project on which I am reflecting? Moyer discusses four perspectives on language in research on bi-/multilingualism: ‘language as form and structure; language as competence and tacit knowledge; language as production and perception; and language as social action and practice’ (2008: 19). This forms the basis of this question and informs those that follow.

Language and languages: In what ways do I draw distinctions between what I mean by ‘language’ in general and what I mean by named-language categories (discrete enumerated entities we call ‘languages’)? To what extent do I think about named-language varieties as separate entities? To what degree does my view entail assumptions of language(s) as more or less static or dynamic? How do I think about variation within and across named-language varieties?

The situated nature of language: How contextualized (or not) do I conceive language(s) to be? In what ways do I think of language(s) as situated within, or apart from contexts of use? To what degree do I consider language(s) to be separable (or not) from those who ‘use’ it/them? To what research end? To what degree do assumptions about the contextualized nature of language play an important role in my work?

Categories in research: Given the difficulties in doing away with named-language category labels entirely in work on multilingualism, what assumptions underlie those that I do use, and how can I make those assumptions explicit? What category labels do I use and how sharp do I see the category boundaries to be? To what extent do I explicitly interrogate these categories in order to recognize their limitations, boundaries, and parameters?

Research questions: What am I foregrounding in relation to my research aims? Am I foregrounding named-language varieties? Am I foregrounding particular linguistic practices (in which named-language varieties may or may not be salient)? How do that which is foregrounded and the other elements of my project sit in relation to one another? Do my research questions and the terminology I adopt in them reflect how I view language, ideas that are central to my work, and what I want to foreground?

Reflexivity

Reflexivity, as an essential part of the ethnographic endeavor, delves into the subjectivity of the research process and the links between the researcher, research process and object(s) of research. Acknowledging the subjective nature of these extends into methodological and epistemological as well as analytical concerns. As a practice and process championed by sociologist Bourdieu, the laying-bare of subjectivities includes self-reflexivity (Bourdieu 2007). Patiño-Santos takes ‘methodological reflexivity’ to include ‘reflexive practices stemming from a variety of perspectives, whether they be epistemological considerations or the researcher’s questioning of his or her own socio-political, cultural, ideological or other forms of personal subjectivity’. (2020, 213). As a self-reflexive work relating to methodological concerns, and based on ethnographic field notes from participant observation, this paper aligns with Blommaert & Dong’s view that ‘[t]here is no way in which knowledge of language can be separated from the situatedness of the object at a variety of levels, ranging from microscopic to macroscopic levels of ‘context’ and involving, reflexively, the acts of knowledge production by ethnographers themselves’ (2010, 10).

Monolingual and plurilingual perspectives

A large and diverse body of scholarly work has made reference to issues addressing monolingual normativity. This work spans fields and sub-fields related to language and Linguistics, including Sociolinguistics, Applied Linguistics, Linguistic Anthropology, Language Policy, Bi-/multilingual Education, SLA, TESOL and others. The main focus here is on broadly sociolinguistic and applied linguistic perspectives.

It is acknowledged here that the term multilingualism itself has been contested, in particular on the basis that it may be seen as simply a ‘pluralization of monolingualism’ (Canagarajah and Liyanage 2012, 22; Makoni and Pennycook 2007, 50). It is used in this paper within the context of South Asia, following Canagarajah and Liyanage (2012), Canagarajah (2009, 2013), Mohanty (2019), and Khubchandani (1983, 1997, 1998, 2004), in line with the way these scholars conceptualize the plurality of individuals, communities, and societies in the region. A substantive distinction between multilingualism and plurilingualism is not made here. The translanguaging approach (García and Li Wei 2013) differs in some ways from conceptions of plurilingualism (García and Otheguy 2020). However, the present article has at its foundation many of the principles upon which the translanguaging approach also relies, namely the non-compartmentalization of linguistic varieties; the foregrounding of linguistic repertoires and integrated competencies; and a focus on contextualization, processes and practices. The translanguaging approach has been influential mainly within the context of language education in diasporic contexts, but its uptake as a more general theory of language (Li Wei 2018) outside this area has been less so. For these reasons, the term itself is not adopted here.

A monolingual orientation

While it is widely recognized that the large majority of people in the world can be considered to know multiple linguistic varieties, perspectives still persist in which knowing only one named-language variety is perceived to be the norm. Canagarajah and Liyanage (2012, 60) contend that ‘the dominant constructs in linguistics are founded on monolingual norms and practices’. Monolingual perspectives stem from language ideologies (Kroskrity 2010; Schieffelin, Woolard, and Kroskrity 1998; Woolard, 2021) relating to the way that language and the discipline of linguistics have been tied up with European colonial projects (Bauman and Briggs 2003; Calvet 1974; Errington 2001, 2008; Pennycook 1998; Pennycook and Makoni 2020) and with ideas of the nation state. Emanating from ideologies related to the ‘Herderian triad’ in which ‘language = community = place’ (Canagarajah 2013, 20), monolingual normativity not only contributes to the reification of the ‘native speaker’ and different forms of discrimination (Barratt 2018; Lippi-Green 2012), but also is rooted in ‘ontological Eurocentrism’ (Pennycook and Makoni 2020, 73).

Barratt (2018, para. 1) defines monolingual bias as ‘the viewpoint that people who speak only one language, that is, monolinguals, are the norm and that bilinguals and multilinguals are exceptions to that norm’. It is also referred to as a monolingual orientation (Canagarajah 2013), monolingual perspective (Cook and Li Wei 2016), and monolingual mindset (Clyne 2008; Slaughter and Hajek 2015).

Another aspect of the monolingual orientation is the conceptualization of language as a discrete system that is somehow separable from its users. This in turn, is associated with the idea that languages are compartmentalized from one another, and induces the thinking that linguistic forms can only ‘belong’ to one named-language variety. Auer argues that in practice, bi-/multilinguals do not speak like ‘two-fold monolinguals’, and that the perception of difference between language varieties is a language ideological matter rather than an inherent property of the language variety itself (Auer 2007, 323). An important thread running through the monolingual orientation is that of thinking about language(s) in terms of compartmentalization and separation, which contrast with the concepts of contextualization, process and practice.

It has been pointed out that while there is much scholarly work about multilingualism, there is little on the subject of monolingualism (Romaine 1995), thus implying this to be the ‘norm’. Early on, Kachru (1994) critiqued theories in language acquisition and TESOL that ignore cultural contexts across the world in which multilingual acquisition is the norm. Likewise, Sridhar highlights the marginalization of multilingual non-Western contexts in second language acquisition theories, pointing out that ‘current theories are powerfully constrained by Western cultural premises’ (1994, 800). Makoni and Pennycook (2007) and Pennycook and Makoni (2020) echo this viewpoint across a broad swathe of disciplinary knowledge within Applied Linguistics. Some early efforts to decolonize ‘English’ also operated through monolingual ideologies, such as the move from ‘English’ in India, as a monolithic object associated with colonial power, to Kachru’s (1976) ‘Indian English’.⁴ It relies upon ideologies of the compartmentalization of linguistic varieties at the same time as it works to break down the colonial ‘ownership’ of English.

As mentioned previously, European colonialism and its legacy contribute to monolingual normativity. Canagarajah traces the evolution of the monolingual orientation from around the eighteenth century in Western Europe, through movements that ‘display a positivistic orientation to reality’, including romanticism, the enlightenment and modernity, industrialization, nation-state formation, structuralism, colonization and imperialism (2013, 19–20). This evolution relied upon languages being ‘invented’ (Makoni and Pennycook 2007), that is to say, constructed as discrete, bounded entities and given category labels such as ‘English’. Languages are thus ‘socio-historic formations' (Agha 2007, 219). The construction process took place through efforts to standardize and ‘purify’ languages, as well as codify them in different forms, which is ‘often an arbitrary process of grouping diverse semiotic resources together, attaching a label to name them, and claiming ownership of them’ (Canagarajah 2013, 20–21).

European nation-states thus relied upon views about the links between language, community and place, and applied them to particular territories under their control, standardizing a national language and often resulting in the suppression of other languages within the nation’s geographic boundaries, as Calvet (1974) describes in the case of France. Similarly, languages ‘owned’ by colonial powers were valorized by the colonizers as superior to others, and these systems of value were used within colonial power relations to minoritize particular language varieties and the colonized peoples who spoke them (Calvet 1974; Pennycook 1998; Pennycook and Makoni 2020).

A plurilingual tradition

Khubchandani describes a plurilingual ethos in the South Asian context, highlighting that ‘in a plurilingual milieu, language boundaries remain fuzzy and fluid, and a verbal repertoire gets blended across well-knit language systems; speakers in such a situation are hardly aware of operating across language boundaries’, suggesting that this milieu is ‘characterized by a continuum between oral tradition and literate culture’ (2004, 48). In addition, sociolinguistic variation is characterized as the appropriation of ‘deviations as the norm’ (1997, 94). These early ideas have been recognized and expanded upon by later scholars. Canagarajah (2009) and Canagarajah and Liyanage (2012) describe a ‘plurilingual tradition’ that has existed in South Asia since pre-colonial times, and underscore that such traditions also exist in many other parts of the world. Canagarajah argues that rather than seeing communities as having or not having such traditions, it is a matter of whether these practices are acknowledged and promoted (2009, 6). He outlines aspects of the plurilingual tradition that contrast in many ways with how communicative practices are conceptualized within European or other ‘Western’ traditions, as discussed below.

Repertoire and practices

Taking Canagarajah’s discussions of the plurilingual tradition in South Asia (2009) as the starting point, the next paragraphs focus on three areas within the plurilingual ethos that contrast with the monolingual orientation.

The first aspect regards the foregrounding of the linguistic repertoire rather than separate competencies in different varieties. ‘Proficiency in languages is not conceptualized individually, with separate competencies developed individually. What is emphasized is the repertoire – the way different languages constitute an integrated competence’ (Canagarajah 2009, 6). The foregrounding of the repertoire permits linguistic resources to be viewed as dynamic, rather than fixed attributes of a community or individual; akin to heteroglossia (Bailey 2007, 2012; Bakhtin 1981; Blackledge and Creese 2014).

The second aspect of the plurilingual tradition highlighted here relates to interpretations of proficiency, in that ‘[e]qual or advanced proficiency is not expected in all the languages' (Canagarajah 2009, 6). Mohanty characterizes multilingualism in ways consistent with this in terms of his experiences growing up in India,

moving naturally and spontaneously between people and languages, unconcerned by any boundaries and infringement. I did not have to bother about my own inadequacies in the languages I encountered, nor did I have to count the languages I knew or did not know. Levels of my competence in languages around me did not have to be judged. The binaries between knowing or not knowing the language and the borders between them did not matter. What mattered is that I could move between the languages without any self-consciousness and, at the same time, with a sense of transient completeness … living with languages was unmarked and natural. (2019, 1)

While fluid practices unconcerned with proficiency levels are highlighted here, this is not to say that others do not experience feelings of inadequacy as regards their linguistic repertoires, or have their competence judged, as did Hindi-medium-educated women interviewed by Sandhu (2016), some of whom felt discriminated against in job interviews and marriage matching. Because they were not educated in English-medium schools, they were perceived to lack the desired fluency and, implicitly, the socio-economic capital associated with ‘English speakerhood’ (Highet and Del Percio 2021). Therefore, Mohanty’s experience (as an educated, English-knowing person) reflects not only local linguistic practices, but also the experience of those whose multilingual ways of speaking are less stigmatised than others’.

Mohanty’s experiences in India are contrasted with life in the USA and Canada, where he was exposed to patterns in which the societies’ many languages ‘remained isolated and confined to specific groups of speakers, without any sign of sharing the same communicative space’ and were ‘marked’ as different (2019, 12). Also noted was the common pattern of shift across generations from immigrant languages to monolingualism in the dominant language, and a ‘pressure to conform to its normative use’ (12).

The third major aspect of the plurilingual tradition discussed here involves foregrounding language as a local social practice (Pennycook 2010). ‘Language competence is not treated in isolation but as a form of social practice and intercultural competence (…) (t)here is a recognition that speakers develop plurilingual competence themselves (intuitively and through social practice) more than through schools or formal means’ (Canagarajah 2009, 6). Therefore, emphasis is placed on the local nature of ways language is contextually embedded and acquired.

Woven into the above aspects is the underlying idea that on different levels there are fluid boundaries between named-language varieties. Plurilingual practices challenge ‘foundational constructs in linguistics’ (Canagarajah and Liyanage 2012, 52), including the perceived separateness of linguistic varieties.

Suppression of plurilingual practices

In contrast to the plurilingual tradition, ‘modernist linguistic constructs that value the separate identity of languages have resulted in a suppression of plurilingual practices’ (Canagarajah and Liyanage 2012, 51). Ideologies about the separateness of linguistic varieties stem in part from ideologies of linguistic purism linked to nation states, as well as associated state and standardization policies. Canagarajah and Liyanage argue that this may be part of the reason why practices they describe as plurilingual have been ‘suppressed to a greater degree in the West’, and argue in favour of the need for greater research on plurilingual practices in communities of both kinds (2012, 51).

It is against the above background that the next section considers language ideological positionings found in the researcher’s field notes data from India. The focus in particular is the ideology that named-language categories are distinct, bounded entities. Secondly, related to the first, are ideologies of language-script correspondence.

Data and analysis – language ideologies in field notes

In discussing new strands of research on multilingualism, Martin-Jones, Blackledge and Creese suggest that the focus has shifted towards empirical work that is interpretive, ethnographic, and carried out in diverse sociolinguistic contexts (2012: 1). The data under discussion below is from one such project.

Considered here are extracts from over 46,000 words of field notes, part of a sociolinguistic ethnography on multilingual practices in Western India. The project investigates multilingual practices in the homes of four families through theories of style, stance and generational identities, which grew out of the experiences of living in a town in Gujarat state for four years prior to the start of the research project. While the overall project includes recorded audio data, interviews and field notes, the extracts considered here are from the field notes.

Rather than being indicative of the character of the field notes as a whole, the extracts analysed below are exceptional glimpses of the researcher’s self-reflexivity in the form of reflections on observations made over seven months at the research site. Having separated reflections from other observations (Curdt-Christiansen 2020), the extracts were selected from the reflections on the basis that they point to a monolingual orientation, and provide food for thought as to how such positionings may be implicit in research on multilingualism. The data extracts are analysed through discourse analysis (Gee 2014) in terms of language ideologies, through the lens of ‘figured worlds’ (Gee 2014: 89; Holland et al. 1998). This approach points the analytical gaze onto ‘a picture of a simplified world that captures what is taken to be typical or normal’ (Gee 2014: 89), and therefore permits a look into the language ideological underpinnings of the viewpoint from which the researcher is situated. In the present case, monolingual perspectives on language feature as ‘normal’ in the researcher’s figured world.

Khubchandani describes linguistic plurality in the South Asian region as ‘traditionally based upon the complementary use of more than one language and more than one writing system for the same language in one “space”’(1997, 96, emphasis in original). Flexible linguistic practices (Creese and Blackledge 2011) in India are frequently found in both written and spoken language. Instances of what might be categorized as ‘English’ words written in ‘Gujarati’ or ‘Hindi’ script are common in the linguistic landscape of Gujarat, and have been academically investigated in other parts of India (Kathpalia and Ong 2015; LaDousa 2020).

In the early stages of the project, this complementary use of languages and writing systems caught the researcher’s interest. One example of this is , a transliteration of cash deposit machine, written in Gujarati script at the top of a cash machine at a well-known Indian bank. Also observed was a flyer written in Gujarati, advertising a restaurant’s menu in which the Hindi word for a vegetable dish is written (sabji), in Gujarati script. The flyer doesn’t use the Gujarati word referring to a vegetable dish, ostensibly to avoid implying Gujarati style cuisine, rather than the North Indian cuisine the restaurant advertises. Below, other extracts of this kind are discussed in greater depth.

At a doctor’s office

The first extract is slightly different than the two short examples above, in that there are two different signs for the same bathroom written in two different scripts. The signs were observed in a doctor’s office waiting room in a medium-sized town. A doorway off the waiting room had a sign above it with TOILET written in capital letters. To the left of the doorway was another sign that said , a transliteration of the English word bathroom.

The field notes say, ‘so when you translate “toilet” into Gujarati, you get ‘bathroom’ but just written in Gujarati script!’ The exclamation point highlights the researcher’s surprise. Expecting ‘translation’ but finding ‘transliteration’ points to an assumption that two signs for the same thing would be written in two different named-language varieties if written in two different scripts. This comment reveals ideologies of a one-to-one mapping of language onto script, with the decoupling of these as noteworthy, unconventional, or marked. Other examples noted from the same waiting room include the fees for different services at the medical practice, written in Gujarati script and including words which could be considered to be part of both named-language varieties. For example, the cost of opening a new medical case, a new case fee: (navo kes fi), combines the Gujarati word for new (navo) with case fee, written in Gujarati script.

In the researcher’s figured world, two different words signposting the way to the doctor’s office bathroom would be in two different languages because they were written in two different scripts assumed to correspond to those named-language varieties. The word TOILET in Roman script (or ‘English’, in this case) matched expectations in the sense that the script and the word written in that script were perceived by the researcher to be aligned (both ‘English’), whereas the word written in ‘Gujarati’ (Devanagari-derived) script was not perceived to be aligned with the script – the script being ‘Gujarati’ and the word perceived as ‘English’. This perceived lack of alignment is underpinned by an ideology that the ‘norm’ is a one-to-one correspondence between named-language variety and script.

This extract may thus be interpreted as evidence of a particular aspect of monolingual thinking, in the form of language ideologies that English is written in a particular script and Gujarati is written in another, and that these remain distinct from one another. The reflections in the field notes therefore point as much to the monolingual orientation of the researcher, as they do to the plurilingual ethos of the research context. The language ideological assumption of a one-to-one language-script correspondence operates in contrast to what the researcher observed.

Initials

The second extract also relates to the compartmentalization of language varieties, but this time regarding the pronunciation of letter names in the alphabets of Gujarati and English, and the way the initials of people’s names are written. On a book written in Gujarati, the author’s first and middle names were abbreviated in a way that appeared to cause consternation to the researcher. They appeared to her to be ‘English-ified’.

She noticed that on a work of fiction written in Gujarati, the author had written his initials in Gujarati script, using the English pronunciation of the first letters of his first and middle names. (IPA notations have been inserted into the data to assist the readers of this paper.) She writes,

For some reason people insist on writing their initials with the English pronunciation of the letters, even if they are Gujarati letters. So if the guy’s name is Dr. I. K. Vijaliwala, in Gujarati ( …) why doesn’t he write it [i] [kə], which are the names of the Gujarati letters, instead of [aı]. [keı]. (…) [W]hy do they use the English pronunciation of the letters they are abbreviating the names of, when the names and the script being used are both Gujarati?

Whereas the previous extract relates to lexical items, this one relates to the pronunciation of letter names, specifically, how initials are pronounced and transliterated. The researcher perceives the pronunciation of the letter names to correspond to the alphabet she perceives the letters to ‘belong’ to. The use of the phrases ‘for some reason’, ‘people insist on’, ‘why’ and ‘why doesn’t’ demonstrate the researcher’s subjectivities related to the abbreviation practices being observed and the expectation that each named-language variety and script used ‘should’ align. In this case, the observation is that the ‘English’ letter names are transliterated into ‘Gujarati’ script, rather than the ‘Gujarati’ letter names being used to represent the pronunciation of the names of the letters in the Gujarati alphabet. Thus, the expected, or imagined, scenario could be summarized as an ideology of language in which pronunciation-of-letter-names = named-language-variety = script. It is instead met with ‘composite’ (Khubchandani 1997) practices on the ground.

A later part of this extract includes metacommentary about the reflections. ‘The fact that I am noticing and pointing out all these supposedly illogical things must mean that I am coming from a perspective where there is a lot more compartmentalization, and I still have a lot to learn about Indian multilingualism (where compartmentalization is less.) In exploring this I am coming face to face with my own preconceived ideas about multilingualism being more compartmentalized into heterogeneous parts than it is in reality here. Mixing of any and all kinds is not considered weird or illogical (or ‘marked’). It is all homogenized/ unmarked. I have the feeling that what is considered weird and illogical (marked) here is extreme compartmentalization’.

The reference to ‘logic’ can be interpreted as part of a figured world, as ‘a socially and culturally constructed realm of interpretation’ (Holland et al. 1998, 52). At the same time as the implied compartmentalization of Gujarati and English, the so-called logic is also underpinned by language ideologies in which language/community/place meets language/pronunciation/script. When applied to the above example, in the researcher’s figured world, this would create an essentialized ideology of language in which the Gujarati language is used by Gujarati people in Gujarat to write the pronunciation of the initials of their names in Gujarati script.

Another language-ideological equation also emerges: named-language variety = script/writing system. This relates to ideologies of standardization in which (in the case at hand) orthographic conventions are applied to particular spoken language varieties. Milroy & Milroy propose that standardization be understood as a historical process always underway (1985, 22). In that vein, processes of change are the norm, rather than the assumption of fixity, as in an imagined fixed mapping of ‘Gujarati’ script onto ‘Gujarati’ pronunciations of the names of letters of the alphabet. Milroy & Milroy suggest that it is more appropriate to think about standardization as an ideology of language in the abstract sense. They view standard language ‘as an idea in the mind rather than a reality – a set of abstract norms to which actual usage may conform to a greater or lesser extent’ (1985, 23). Above, the researcher found that the practices she observed did not conform to the abstract set of norms she held in her figured world. The language practices observed are something more composite and dynamic than the researcher’s prescriptivist thinking implied. The practices themselves are not foregrounded in her thinking. Instead, there is a focus on compartmentalized named-language varieties rather than plurilingual practices, and a limited focus on the conventions of use surrounding these practices.

The researcher’s awareness of this problematic logic exists at the same time as she employs it, highlighting the complexity of language ideologies and reflexive processes in research. Practices around orthography are not neutral, and the ‘process of representing the sounds of language in written form is thus an activity deeply grounded in frameworks of value’ (Schieffelin and Doucet 1998, 285). Fluidity and flexibility in orthography, rather than strict one-to-one correspondences, may be locally valued and practiced in India (LaDousa 2020). However, the field notes suggest that observations of these practices challenged the researcher’s original expectations which looked to different frameworks of value related to monolingual standard language ideologies.

Named-language varieties vs. practices

The third extract of the field notes discussed in this paper is a meta-reflection on compartmentalization. Written months later than the previous ones, this extract begins to critique the monolingual orientation, or at least the compartmentalization of named-language varieties, and in that sense, goes beyond simply developing an awareness of it.

This extract mentions that one of the things

that I keep banging my head up against time and time again is the realization that the way the field of linguistics teaches us to think is that it is normal to be speaking one language at a time. (…)This compartmentalization is part of a certain reality that we are taught, and that is socially salient to us as speakers, as students of linguistics, and as academics. I am not saying anything new here. (…) Is it possible that the more I try to understand Indian-style multilingual practices the more I run into pre-conceived ideas from US, French, and UK academic systems? What are some of the limitations (?) in thinking that have been at the heart of the discipline of linguistics from the very beginning?

The researcher designates herself as a Western-trained student, an academic-in-training and layperson at the same time, simultaneously being within the field of (socio)linguistics, knowing languages, and being a student of languages. A realization appears to dawn that the problem of the compartmentalization of language varieties seems intractable in one sense, in that to a certain degree it is built into the academic field. However, on the other hand, these meta-reflections, and the engagement with the scholarly literature that would have made them possible, demonstrate an opening up of epistemological and ontological awareness through fieldwork in a research site where a plurilingual ethos predominates. This awareness leads to the idea that while conceiving of named-language varieties as discrete objects is socially salient in some ways, it is also not necessary that they should be the only or the primary units of analysis.

In addition, this kind of awareness highlights the value of ethnographic fieldwork and emic perspectives. Having trained in the US and France in teaching English and French, as well as sociolinguistics, the researcher spent 15 years working as an English language teacher at a high school in Guinea, a language centre in Morocco, and in higher education in France and the UAE, in addition to living in India (her husband’s ancestral home) before doctoral studies in the UK. She engaged in language study formally or informally in all of these places. The reflexive extracts analysed above highlight monolingual orientations in her figured world. As Makoni and Pennycook argue, ‘the concept of language and indeed the ‘metadiscursive regimes’ used to describe languages are firmly located in Western linguistic and cultural suppositions’ (2007, 27). This reminds us that much may be absent from our conceptualizations of language.

Discussion – methodological implications and reflection questions

Writing from an avowedly interdisciplinary perspective, Gramling situates his critique of monolingualism outside any one discipline precisely because ‘disciplines go to great lengths to maintain plausible deniability about our own monolingualist forms and formats’ (2016, 40). Given this reminder of our own blind spots, this section attempts to address how we might begin to apprehend them. It reflects upon ways that language ideologies can infuse different aspects of methodology in transcribing, coding, and the framing of research questions.

Transcribing and coding

The transcription of recorded data has been recognized as partial, political, and laden with researcher positionings (Bucholtz 2000; Kalocsányiová and Shatnawi 2022; Ochs 1979; Vakser 2017). ‘Embedded in the details of transcription are indications of purpose, audience, and the position of the transcriber toward the text’ (Bucholtz 2000, 1440). A recording, an audio representation, captures some aspects of an original multi-modal event. To transcribe this is to move from a time-bound auditory representation of spoken language to a linear, visual, written representation of the recording. Transliterating, or not doing so, also requires choices emanating from one’s perspective.

In transcribing (and transliterating) multilingual speech, we are in effect coding varieties of language during the process, because the way an item is written reveals the language-variety category in which this item is perceived to fit, and the potential readership. The imprecise nature of boundaries between named-language categories comes into play, as in the following example. Four possibilities are presented here: (1) (2) (3) ḍokṭar; (4) doctor. Selecting option 1, in Gujarati script, the transcription might be directed towards a potential readership that is Gujarati literate. It also foregrounds the ‘foreign-ness’ of the vowel sound (and by extension, the word) by the use of the second letter on which the matra vowel symbol has been reversed, indicating the vowel sound is [ɔ], (rather than [o] as it would be in option 2). In option 2, the matra vowel symbol is in its conventional orientation, potentially indicating a ‘local-ness’, rather than a ‘foreign-ness’. This might attribute ‘doctor’ to the named-language category ‘Gujarati’ to a greater degree than option 1 would. Options 3 and 4 use a Roman script and thus pre-suppose a readership that may not be literate in Gujarati script but may read languages written in Roman script. These transliterations would make the transcription more accessible to members of the large Gujarati diaspora not literate in Gujarati but in other languages, and to others who read languages written in the Roman alphabet. Option 3, transliterated, implies belonging of the word to the category ‘Gujarati’ through orthographic choices with diacritics that differ from the ‘English’ spelling of option 4. Thus, in this decontextualized example, attribution to named-language variety categories can be a matter of perceived degree, though is still confined by the artifice of a binary choice of ‘Gujarati’ or ‘English’ (Gardner-Chloros 2009, 187–188). Named-language categories reflect a focus on these language systems as analytically foregrounded, but even when the focus is elsewhere, these categories (and what we choose to put in them) contain traces of how we view language.

Choices in transcription and transliteration bear the marks of the foundations upon which the choices are made. Jaffe describes transcription practices as ‘ideologically and analytically motivated forms of representation’ (2012, 222). Choice of script and transliteration are thus also non-neutral choices reflective of the researcher’s perspectives.

Research questions

Although the main aim or ‘overall question’ and the ‘specific research questions’ (Flick 2014, 147) are generally formulated early on, in an ethnographic project such as the one under discussion here, where fieldwork is an important part of the project, the specific research question(s) may be evaluated and reformulated. This can take place after selecting the research design and method, and after the data has been collected (2014, 147). In the present project, changes were made in the formulation of the specific questions after the fieldwork and data collection. Reflections on the plurilingual ethos made it clear that some terminology in the questions would better suit the research aims if oriented more towards practices rather than named-language varieties, although these would still remain a part of the analysis. While early iterations of the research questions referred to ‘Gujarati-English codeswitching’, later iterations referred to theoretical constructs of ‘style’, ‘stance’ and ‘identity’.

Thus, removing named-language categories from the wording of research questions made space for foregrounding discursive practices. This represented a shift in researcher positioning that came not only from scholarly literature but from the data itself. In the four families under study, met through friendship networks, members of the parents’ generation were asked if or how the way their children spoke was different from the way their own generation did. The researcher expected that they might say something about how members of the younger generation use more English or mix in more English when speaking Gujarati (and some people did say this later in the interview when asked more pointedly about that.) However, in fact, many parents said that their children spoke more frankly or openly than they had been allowed to speak to their own parents. The fact that the researcher expected the participants’ answers to comprise something related to the named language varieties of Gujarati and English indicated that she was more focused on these, but the participants were more focused on discursive practices as generationally different. This realization, in conjunction with stancetaking practices observed in the conversational data, helped initiate a shift towards the foregrounding of discursive practices in the project and the research questions. Therefore, while the named-language categories, as socially meaningful categories within the research context, did not disappear from the project, a shift in focus occurred, moving away from languages as discrete systems towards the foregrounding of local multilingual practices. This, then, also resulted in the subsequent modification of the analytical framework, which ensued as much from the data itself as from the theoretical perspectives that influenced the research questions.

Questions for reflection

The monolingual orientations discussed in this paper relate primarily to the idea of languages as discrete, clearly bounded systems, which has been extensively deconstructed from a theoretical perspective (Makoni and Pennycook 2007). This section proposes questions for reflection so as to move the discussion further into the methodological arena, in terms of how ideas that have largely remained in the realm of theory can speak to researchers’ practices. Rather than a model or framework, it is hoped that these questions in Table 1 will be seen as a starting point for such reflections.

Observations of a plurilingual ethos led the researcher to question categories and assumptions. She confronted ontological differences between her disciplinary thinking and the ethos of the research site. Thus, ‘the ways people understand their worlds challenges the ways the categories of nature, culture or language may be understood’ (Pennycook and Makoni 2020, 73). Indeed, the ideas about language(s) upon which our disciplines and sub-disciplines are founded require questioning as much as any other category (Agha 2007, 218).

Irvine suggests that the idea of ‘ideological work’ is preferable to ‘ideologies’ in that it avoids the possibility of language ideologies being seen as immutable and static, when in fact they are ‘grounded in one’s socially-positioned point of view, and crucially, in the conditions of the historical moment’ (2019, 68). Given this, questions for reflection and awareness-raising may be considered part of a researcher’s own ideological work, in terms of self-reflexivity and engaging with one’s assumptions and preconceived notions as well as with one’s own research practices (Patiño-Santos 2020, 213). The questions above are intended as a starting point, to reflect upon what a researcher is foregrounding, what one is making prominent, and what assumptions that are taken for granted may underpin that.

Conclusion

The constructs of named-language varieties and associated ideologies of compartmentalization direct our thinking in particular ways. Contrasts between a monolingual orientation and a plurilingual ethos could be framed as ontological, and in practice, differences then emerge relating to category boundaries and what becomes foregrounded in research. To foreground language practices and processes may mean placing other things, such as named-language systems, in the background. While named-language varieties can be understood to be constructs that are European epistemological inventions (Makoni and Pennycook 2007; Pennycook and Makoni 2020), these constructs have been made salient to the layperson and academic alike over time. Just as their theoretical deconstruction continues to open up welcome new perspectives (Kusters and Sahasrabude 2018), the dismantling of these constructs also presents new challenges which remain under-explored from a methodological point of view. One such challenge dealt with here is researchers’ awareness of their own monolingual orientations. Indeed, Andrews et al. argue that researcher preparation ‘would benefit from being informed by a translingual mindset’ (2019, 77). Further scholarly work is needed, because as Pennycook and Makoni point out, ‘the straitjacket of monolingual thought is not so easily thrown off’ (2020, 42).

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 I wish to thank Dr. Rajendrasinh Jadeja for his work on the translations. I am also grateful for comments and suggestions from two anonymous reviewers, and the editors of this special issue. I alone am responsible for any remaining flaws.

2 An early iteration of some of these ideas was presented at the British Association for Applied Linguistics (BAAL) conference at University of Leeds, UK, 2017.

3 The research project received ethical approval in April 2014 from Birkbeck, University of London, Department of Applied Linguistics and Communication (now Department of Languages, Cultures and Applied Linguistics).

4 I thank an anonymous reviewer for pointing this out to me.