Sage Journals: Discover world-class research

Abstract

Qualitative research projects are often conducted across languages and cultures. Several specific decisions need to be made in such multilingual projects, such as deciding how to communicate with participants, whether to translate and how to navigate transcription. However, researchers do not often articulate these decision-making processes, thereby conveying a sense that such decisions were made without tensions. Issues of power come into play in linguistic decision making which may impact the relationship between the researcher and participants and, thus, the data collected. This paper explores linguistic reflexivity through reflections on our decision-making processes in our multilingual research projects. Foregrounding our positionalities, we highlight how researchers’ linguistic profiles inevitably impact their linguistic decisions and the ways in which such decisions foster linguistic reflexivity. Underpinned by a decolonial lens, our linguistic reflexivities prompted analytical, methodological, ethical and epistemological concerns and choices which enabled us to forge more meaningful connections with participants in their places. In articulating our decisions, we encourage researchers to begin their research with linguistic intentionality and think critically about the usage and exclusion of language throughout the key stages of research.

Keywords

decoloniality linguistic reflexivity positionality multilingual research cross-cultural research

Introduction

Researching multilingually involves “using, or accounting for the use of, more than one language” in research (Holmes et al., 2016; cited in Rolland et al., 2023). Cross-cultural, multilingual research involves applying research methodologies and values generated in one language to another (Regmi et al., 2010). This research approach has become increasingly popular as researchers traverse cultures and contexts between and within boundaries and mobilities. In doing so, researchers may engage with participants with whom they share no language (and) or whose culture and context they little understand while, in keeping with the dominant academic practice, report their research findings in English (Goitom, 2019), perpetuating a racialised language ideology of colonial language superiority (McKinney & Molate, 2022). In such research, it still often remains the case that researchers offer little or no account of how they engaged with participants during data collection. In particular, limited or no reflexivity is practised around the influence of their linguistic choices (intentional or otherwise) on the methodological and analytical decisions made at different points of the research process. Even where researchers engage with participants from their own cultures or with whom they share one or more linguistic profiles, as is the case with the co-authors of this article, the need for linguistic reflexivity remains and is critical for research whose analysis seeks to “be true to the voices of all the participants and reflect an understanding of the topic that is shared by the researcher and participants alike,” as argued by Indigenous scholar Shawn Wilson (Wilson, 2008, p. 101).

The implications of linguistic reflexivity extend beyond the analytical and the methodological towards the epistemological and political (Temple & Young, 2004), the ethical (Pereira et al., 2009) and the relational, the latter which, as Wilson (2008) suggests, requires reflecting and building on the relationship or connection between research participants, research ideas (or topics) and researchers themselves. Alongside culture and context, language is fundamental to this relationship and is one of the tools through which researchers make connections between themselves, their participants (and their contexts or places) and their research to generate data and analysis which meets not just conventional quality criteria such as validity and reliability, but also relational ones such as trustworthiness, authenticity, or credibility.

In this view, language is not an abstract conception but an “embodied relation between communicating subjects” (Chapman, 2023, p. 36), created through social interaction (Veronelli, 2015). This relational, place-based view of language in research thus inheres an embodied pluriversality in which the researcher is connected with others within the context of the research (Dunford, 2017), and their research is informed by this connection. Data, and by extension, analysis are a product of these relationships and connections, and are not a ‘disembodied universality’, disconnected from participants and place (Dunford, 2017, p. 386). Moreover, such a relational, place-based view of language is fundamental to exploring the linguistic relations of power (Veronelli, 2015) inherent in multilingual research. This requires attention to whether, and the extent to which, the languages employed or excluded in multilingual research reinscribe coloniality of power through language by their hierarchisation of the expressive tools available to (or employed by) researchers and their participants, and the colonially prescribed epistemic value of these tools (Veronelli, 2015). Attention to how the language(s) employed during research connects us to people and place (or disconnects us from these), and to whom it gives or removes power, therefore requires attention to the linguistic decisions (and linguistic acts) in cross-cultural multilingual research, particularly during data generation, transcription, analysis, and reporting. Such decisions are enabled by linguistic reflexivity and require attention to when and how we translate from or make sense of wor(l)ds expressed in one language to another.

This paper explores these issues from the perspective of two scholars who conducted qualitative research in their countries of origin with participants of a shared, overarching cultural group with whom they also shared a non-English linguistic profile. Scholars with similar positionalities have interrogated issues of reflexivity and the broader ethics of doing research in ‘home’ contexts (e.g., (Siwale, 2015). However, this paper focuses on language as an essential element of critical, ethical cross-cultural and multilingual research in the context of a primarily qualitatively oriented research methodology. The paper begins with a review of relevant literature around the notions of reflexivity, translation, and linguistic reflexivity, as well as the literature around linguistic decision-making during the fieldwork, transcription, analysis and reporting stages of research. It then introduces the co-authors before providing a reflexive account of the linguistic decisions in the context of their cross-cultural, multilingual research and the ways in which these enabled connections with participants in place or revealed power dynamics. The paper concludes with the discussion of how researchers’ may forge meaningful connections and transfer power to their participants by engaging in linguistic reflexivity from the beginning of their research projects.

Reflexivity, Translation, and Linguistic Reflexivity

Reflexivity

Reflexivity refers to the personal views and positionality through which a researcher perceives gendered, classed, aged, racial/ethnic, and other ways of seeing and feeling in the world (Erikson, 2017) as well as the ways in which they are perceived (Oyinloye, 2021). In some cases, some of a researcher’s positional markers, e.g., culture, language and nationality, may be shared with those of their participants, and much has been written on the notions of ‘insiderness’ and ‘outsiderness’ in cross-cultural research, including their fluidity and the spectrum in between (Carling et al., 2014; Cormier, 2018; Lizzi & Leon, 2016; McNess et al., 2015; Nowicka & Cieslik, 2014; Sultana, 2007). Particularly with cross-cultural research, researchers’ positionalities, including how they are perceived, and how researchers navigate them, have methodological, analytical and, ultimately, epistemological consequences. Notably, our reference to positionality transcends performative positionality statements or declarations (Gani & Khan, 2024) towards a meaningful interrogation which seeks to minimise hierarchical power dynamics between researchers and their participants.

Moreover, cross-cultural researchers who identify, to whatever extent, as ‘insiders’ in relation to the language(s) of their research participants, i.e., as linguistic insiders (Cormier, 2018) and therefore conduct at least some aspect of their research in one such shared language, grapple with various tensions across the research cycle. Among these are the tensions between the linguistic decisions, actions, and perceptions of researchers and participants (Rolland et al., 2023) and those inherent in the process of translating researcher-participant interactions into the defacto language of academic publication during the analytical and/or reporting phases of research. Such tensions occur, in part, as researchers and participants navigate their linguistic power dynamics, while influencing the connectedness between the two.

Cross-cultural research often involves translation (and translators), and the translation process in studies where researchers themselves are not linguistic insiders and are thus limitedly involved in the process has been widely discussed (Regmi et al., 2010; Temple & Young, 2004). However, the “mediation and intervention” (Qun & Carey, 2023, p. 5) conducted by bi- or multilingual researchers who conduct some or all of the translation of their research tools or data has received less attention. Multilingual actions, particularly within cross-cultural research are not typically made explicit, and limited practical guidance exists for researchers on how to consider the interactions and intersections of researchers’ and participants’ linguistic repertoires and their influence on relationality and power within the research process (Rolland et al., 2023).

Translating Multilingual Data

An overarching understanding within cross-cultural, multilingual research is that the translation of qualitative data, and tools, cannot be undertaken with a promise of “neutrality, objectivity, and honesty” (Shklarov, 2007, p. 530). As such, scholars are increasingly reflexive about the translation and interpretation of bi- and multilingual data. For instance, translation is no longer considered a routine logical or technical issue, but rather an intentional and thoughtful task which may significantly impact the research process, outcomes and ethics (Shklarov, 2007, p. 530). Others have highlighted concerns about accuracy, trustworthiness, ethics and presentation and how these arise at different points within the research and translation process (Greer, 2003; Halai, 2007). Thoughtful, reflexive translation requires conscious linguistic choices at different stages of the research process which interrogate how such choices enable relationality and manage power, particularly between the researcher and their participants. Such choices include what language(s) to use to generate data, what data to translate, how and by whom (or what), and at what point to translate. The temporality of translation, e.g., considerations around when to introduce translation into the research process, is an important decision which requires greater explicit discussion than what is typically carried out (Halai, 2007; Santos Jr et al., 2015). For Temple and Young (2004), this is not just a methodological decision, but a political and ontological one, such as the dialogic, visibilising effect of late translations into English. For Cortazzi et al. (2011), critical linguistic (and, as such, translation) decisions are required at the pre- and during interviews stages, and the stages of transcription, translation, analysis, and reporting (writing up).

Given the limited discussion on these considerations around translation (Cortazzi et al., 2011; Douglas & Craig, 2007), relatively limited guidance exists to support researchers’ linguistic decision-making, particularly around the critical questions that should be asked during research design, and revisited during research conduct, analysis, and reporting. Such guidance should move beyond considerations around financial resources and time (Chen & Boore, 2010; Lopez et al., 2008; Santos Jr et al., 2015), important though these are, to include those around power (Cormier, 2018) and the extent to which linguistic decisions foster relationality through connectedness and the transmission of not only discourses, but also cultures & experiences (Qun & Carey, 2023, p. 5).

Linguistic Reflexivity

A researcher’s linguistic positionality explains the researcher’s linguistic identity or profile in relation to participants and the research process, towards deepening understanding of the influences of the former on the latter. Doing this is likely to uncover tensions not only within the data generation process, but also within the data generated. For Cormier (2018), researchers’ critical understanding of their linguistic positionality may be amongst the ways to understand such tensions, particularly those inherent within the translation and interpretation of cross-cultural, multilingual data generated by researchers who identify to whatever extent, as linguistic insiders. Linguistic reflexivity, however, moves beyond merely understanding to interrogate how this positionality influences linguistic decisions within the processes of collecting, translating, and interpreting data, as well as writing up (Oyinloye, 2021). This requires reflexivity not only about the languages used by all those involved in the research process, but also the languages not used, or excluded, by different stakeholders and at different phases of the research process (Rolland et al., 2023).

Linguistic reflexivity also necessitates a consideration of and response to the issue of power, and scholars have underlined the implications of linguistic choices on the researcher-participant power dynamic during data collection and other parts of the research process (Cormier, 2018; Rolland et al., 2023). For Cormier (2018), a researcher’s linguistic capabilities, particularly in a dominant language, becomes a form of power which extends the language of interaction towards the language of analysis, translation, and reporting. However, limited studies have explored how researchers’ insecurities around their linguistic proficiency may shift the researcher-participant linguistic power dynamic. Fundamentally, linguistic reflexivity offers opportunity for a participant-centred linguistic approach which shifts the linguistic centre from the researcher’s language towards participants’ while enabling a more meaningful connection between participants, the ideas explored in the research, and the researcher. Linguistic reflexivity is therefore critical to enabling a trustworthy transmission of participants’ worlds in a manner which affirms those worlds, including their complexities and nuances.

In the next section, we review the limited literature around linguistic choices in cross-cultural, multilingual research drawing on the stages identified by Cortazzi et al. (2011): pre- and during data collection (here discussed together as ‘fieldwork’); transcription; and analysis and reporting. Unlike Cortazzi et al. (2011) who consider translation a distinct stage, we include it here as a cross-cutting element of the linguistic decision-making process in cross-cultural research.

Linguistic Decisions in Cross-Cultural Research

Fieldwork

We consider fieldwork to include the decisions taken during the data collection period as well those taken beforehand during the research design phase (pre-fieldwork). During pre-fieldwork, linguistic decisions focus on developing data collection tools which are authentically translated to ensure participants have clarity about what the research is about and what they are consenting to (Brislin, 1970; Sinaiko & Brislin, 1973). This means ensuring that surveys, questionnaires and interview questions are devoid of colloquialisms, translated into the target language where necessary, and checked for grammatical translation errors (Brislin, 1970; Sinaiko & Brislin, 1973). This process may require written linguistic capabilities beyond the researcher’s own capabilities, particularly where additional documents such as information sheets are required to be translated.

Translation moreover comes into play during fieldwork in considering who engages with participants on the one hand; and where required, the role of the interpreter on the other. For instance, Cormier (2018) discusses the possibility of collecting data, e.g., interviews, in the source or participant language with an interpreter whose target language words are then transcribed; or conducting an interview entirely in the source language and thereafter translating the ensuing transcript. Moreover, she highlights the need to be attuned to participants’ code-switching and seek for elaborations where participants share multiple linguistic profiles with researchers and therefore assume researchers’ implicit understandings of their words. Mediating the relationship between the researcher and participants, other scholars have highlighted the mediatory role played by interpreters who are unable or unwilling (for fear of causing offense) to convey the researcher’s intended information to participants (Shamim & Qureshi, 2013). In general, where researchers collect data themselves, limited literature exists around the linguistic decisions they make prior to and during data collection, particularly where linguistic capabilities vary.

Transcription and Analysis

A transcript is a written version of interviews or other field data. Cross-cultural multilingual interview of fieldnotes data collected in audio or audio and video forms are typically transformed, i.e., transcribed, into a written form in preparation for deeper analysis. Given English remains the predominant language of peer reviewed academic outputs, transcripts are mostly written in English thereby requiring the translation of the non-English source data. Scholars have considered various timelines for the translation of such data (Cortazzi et al., 2011; Santos Jr et al., 2015). The earliest phase in the translation timeline refers to translating at the time of transcription and doing back translations, i.e. returning to the original interview audio to check translations for authenticity. To avoid mistranslations which can be carried forward and compound subsequent phases of data analysis, some scholars suggest researchers delay translation or use the source language for as long as they can (Santos Jr et al., 2015; Van Nes et al., 2010). As already highlighted, for other scholars, this is not just an analytical act, but a political one as early translation may be seen as collusion with the invisibility of certain languages (Temple & Young, 2004).

Linguistic decision making within transcription also includes whether or not to translate into the language of analysis (where source data has been collected in a different language); and where the decision has been taken to translate, who will translate (whether to employ translators or not), how and to what extent. For instance, Sutrisno et al. (2014) discuss differences between lexical (i.e., word for word) and conceptual (i.e., idea, sentence-level) translation; and single, back, and parallel translation methods. Where word for word translation may not be possible as there may not be a direct equivalent of a source language word, researchers may seek conceptual equivalence where they try to identify a word or phrase that is not only comparable but also familiar to the intended readers (Qun & Carey, 2023, p. 4). This process involves linguistic reflexivity as researchers must carefully consider the perspectives of participants in the source language and then translate it as closely as possible to a different language. Single translation involves translating from source to target language; back translation involves single translation followed by translation, by a different translator, of the translated data back into the source language; while parallel translation involves two simultaneous translations followed by comparison of the translations. Analysing data in multilingual research thus requires a back and forth between data in different languages, which demands researchers engage in “continuous acts of interlinguistic translation” (Santos Jr et al., 2015, p. 142; Tarozzi, 2013).

Reporting

Regardless of how researchers decide to translate their data and the timelines they follow, given the dominance of English as the language of publication, most researchers will eventually need to include translated English excerpts in their research outputs, particularly for peer reviewed academic publications. Where data include participants’ own words in English (as spoken by participants for example), where English is not participants’ first language, some researchers may choose to ‘clean up’ such words to sound less awkward, or to render the words as if spoken by a native English speaker. While this is not translation per se, it constitutes part of the often-invisible linguistic work researchers undertake when they render multilingual data into English. Including source excerpts of data generated in another language makes visible this largely invisible translation act (Cormier, 2018). Moreover, it enables an interrogation of the authenticity of the quotations used, i.e., the extent to which these represent participants’ voices and may be considered direct quotes, and the extent to which the translated excerpt captures the subtleties of meaning embedded in the source excerpt (Halai, 2007, p. 351). Cormier (2018) highlights four ways of presenting translated qualitative data or excerpts: only translated data, with note that data is translated; all source quotes, followed by translated quotes; translations directly following individual speakers’ words in source language; and source language directly following translated words. Others, like Qun and Carey (2023) have included key source phrases in brackets within translated text while including additional explanations for conceptually translated text.

The next section highlights the linguistic decisions made during our research, including how they influenced our connectedness with participants and places, and helped uncover implicit power dynamics. To deepen understanding of these decisions, we begin the section with a discussion of our (and our participants’) positional markers and the context of our studies. We then examine the linguistic decisions we made prior to and during our fieldwork as we reflected on our linguistic positionalities and how these may shape our research. Thereafter, we discuss our consideration of language during transcription, analysis, and reporting, particularly how our connectedness with participants influenced the linguistic decisions we made during these stages.

Linguistic Decision-Making in Practice: Reflections from Our Research

Linguistic Positionality – Hamid

Due to Pakistan’s colonial history and its intersection with my educational background and linguistic repertoires, I am simultaneously a linguistic insider and an outsider in relation to my research participants. A former British colony which gained independence in 1947, the country continues to grapple with a language war between its many regional languages and its two national languages: Urdu and English. I am an insider because I am fluent and can communicate in Urdu since I have spent most of my [educational] life in Pakistan, including my undergraduate education. However, I am an outsider because my socioeconomic status enabled me to study in private schools where English was the medium of instruction (as was also the case for my undergraduate degree). I am more comfortable writing and thinking in English than in Urdu, and my higher competence in English played a role as I navigated the task of sharing my participants’ voices as authentically as possible.

My research aims to understand lower socioeconomic students’ experiences of higher education in Pakistan and whether English plays a role in their chances for social mobility. Prior to this, I worked at an A level equivalent college owned by an NGO, catering to students from low socioeconomic backgrounds. The alumni of this college now constitute the participants of my research, and I use the college as the case study. Having worked at the college makes me a somewhat familiar face to some of my participants, because I could understand their experiences to a certain extent. However, my participants studied in Urdu-medium schools before transitioning to this English-medium college, and they did the local exams whereas I did the GCSEs. Due to my socioeconomic status, English medium education, and opportunity to pursue a postgraduate research degree in the UK (including the ability to return to Pakistan for fieldwork), participants typically perceived me more as an outsider than an insider.

Later in the paper, I discuss how I navigated my positionality as both an insider and outsider. Moreover, the various decisions taken during the fieldwork, e.g., around language and fieldwork, as discussed in the next sections of the paper, were heavily influenced by my positionality. These decisions were informed by my own reflexivity as well as the in-depth discussions I had with my research supervisors at every stage of the fieldwork.

Linguistic Positionality – Oyinloye

My research explored parents’ perspectives and practices around their children’s formal public schooling, as well as teachers’ perceptions of these, in rural Muslim Yorùbá communities in North central Nigeria. The country is a former British colony, gaining independence in 1960; thus, the government public schooling system is traceable to colonial administration schools and influenced by its missionary counterparts. The study comprised a two-staged ethnographic fieldwork which used partly structured interviews with parents, teachers and children, and observations of children’s pre- and after school activities within their communities. It was therefore necessary to speak the local language which I did given my Yorùbá cultural heritage, despite having emigrated from Nigeria as a child. Although I positioned myself as a linguistic insider, I was nevertheless anxious about my linguistic competence due to the characteristic depth of language in rural communities, particularly among older persons such as my core participants, and my unfamiliarity with colloquialisms and other situated ways of speaking in the study context which were vastly different to the urban spaces where I lived as a child.

As such, throughout my study, I positioned myself as a linguistic insider but a partial cultural insider, though participants’ recognition of these positionalities shifted relative to the participant and context of the interaction. For instance, because of my visible and linguistic attributes, participants initially perceived me as a complete cultural insider, albeit from a different part of their region. However, these perceptions waned the more of myself I revealed during the fieldwork. Positional markers such as my then place of residence (i.e., the United Kingdom) and level of education (advanced postgraduate degree) shifted perceptions of me as an insider and in some ways influenced our interactions. For instance, as a ‘highly’ educated non-member of the community who travelled overseas to come to ‘research’ with a notebook and pen perpetually in hand, some participants initially perceived an expectation of formality during our initial interactions. These, like other perceptions, waned during the fieldwork the more participants and I interacted, and they realised I had no such expectations. Notably, my diasporicness intersects with my linguistic positionality in that while I position myself a linguistic insider, my cognitive language has been English from a very young age due to the English colonial origin public schooling system in Nigeria. The extent to which this renders tenuous my self-positioned linguistic insiderness in Yorùbá is something that will be later explored in this paper. Like Hamid, the decisions taken during fieldwork were influenced by my own reflexivity; albeit, unlike her, these were further informed by my discussions with my local contacts and experiences with participants themselves.

Language and Fieldwork – Hamid

When preparing for my fieldwork, I wrote all the documents in English due to administrative (the need for feedback from supervisors and to submit all the documents for the ethical review in English) and personal reasons (higher written fluency in English). Once my consent forms, participant information sheets and interview prompts were finalised and approved, I translated them into Urdu. I then sent the translated documents alongside their original English versions to another native Urdu speaker to further check for accuracy and linguistic correspondence, as this was a requirement from the institution’s ethics board.

When I began fieldwork, I offered my participants the consent forms and participant information sheets in both languages. However, they all requested the English versions. At the time of the interviews, the participants had either spent two years at the college or had started university and so they were all receiving English-medium education and thus felt more comfortable reading the forms in English. It may also be because they wanted to be seen as being proficient in English, although from my experience at the college, I knew that students typically developed fluency in reading faster than writing, with speaking being the most challenging.

Alongside exercising linguistic reflexivity, I applied my contextual and other cultural knowledge during my interviews to ensure that interviews not only centred participants but also served as a space to build trust and encourage rapport. For example, for all my interviews, the conversations began organically in Urdu as pleasantries were exchanged. Given my suspicion that my outsiderness may inadvertently induce pressure to speak in English, I decided not to ask participants what language they wished to use for the interview. This was also an effort on my part to disrupt power dynamics. However, nearly all participants mixed in English words as they spoke. Some used English more than others but all the interviews were ‘bilingual’. It is a very common practice to mix English and Urdu in speech in Pakistan and to continually move back and forth between the two languages. Even with participants who spoke English more than Urdu, I tried to speak predominantly in Urdu to foreground my insiderness. Moreover, I wanted to avoid scenarios where I use an English word which my participants may not be familiar with. Not only would this hinder their understanding of the questions and therefore their answers, it could have also made them uncomfortable. Being conscious of my power status as an outsider and a researcher, I sought to use language to balance this power dynamic. Since I was conducting research as a sole researcher with fluency in both languages, I did not require interpreters during fieldwork.

Language and Fieldwork – Oyinloye

While planning my fieldwork, like Hamid, I applied contextual and cultural knowledge and linguistic reflexivity to build trust, encourage rapport and disrupt power dynamics. I adopted multiple approaches for the fieldwork documents and data collection process. Information and consent forms were all written in English. The documents were not translated to Yorùbá given my knowledge of the predominance of oracy in my research communities, the time and resources required to accurately translate the documents due to my own limited Yorùbá writing capabilities, and my uncertainty about parents’ Yorùbá literacy capabilities. However, I planned to enumerate information and consent forms in Yorùbá, with parents providing oral consent. Given my linguistic capabilities in the local language, I planned to conduct all interactions with parents and children in Yorùbá, and for these groups, my fieldwork linguistic approach transpired as planned.

For teachers, given English was the official language and medium of instruction at the upper primary levels, I presumed they would be able to read the information documents and sign their own consent forms. For the same reason, I planned to interact with teachers in English. Upon arrival, I immediately realised that for richness and breadth, interactions with teachers would similarly need to occur in Yorùbá. Except two head teachers who read their own information sheets (one of whom read it aloud in my presence), I enumerated the information sheets in Yorùbá for teachers. However, all teachers read and signed their individual consent forms in English, with clarifications offered as requested, and I returned a photocopy of the signed form to them.

As with parents and children, interacting in the language of their familiar not just during interviews but also during general conversations deepened my ability to connect with participants in their place. It also disrupted the researcher-participant power dynamic because participants’ Yorùbá linguistic capabilities were superior to mine. While I would have preferred to have had parity of linguistic competence, I ultimately embraced my linguistic subordination as a way of ceding power to participants. All interactions and interviews therefore occurred in Yorùbá, and all audio recordings were in Yorùbá. Other data included handwritten, typed (mobile phone), and voiced / narrated fieldnotes in English, with specific Yorùbá terms and phrases interspersed; field (voice) notes narrated in English; and field reflection notes written in English.

Transcription and Analysis – Hamid

As previously mentioned, my fluency in both languages allowed me to conduct fieldwork on my own without the support of an interpreter. Likewise, I did not feel the need to work with translators when I began transcribing my interview data. When I first began transcribing my interviews, I did not stop to think about the impact that language and translation would have on the conversations I shared with my participants. Given the requirement to write my thesis in English, and my first supervisor’s non-comprehension of Urdu, I mentally translated the bilingual recordings and then transcribed them directly to English. Because of my inexperience with translation, I also ‘tidied up’ participants words to erase grammatical mistakes in sentence structures in order to make the transcriptions more easily readable for the English reader.

After I shared the first three transcripts with my supervisors, they noted that the transcripts sounded very fluent in English, and questioned how the transcripts would read if back translated into Urdu. The issue of authenticity was, moreover, raised by my second supervisor, a British-Pakistani fluent in Urdu and familiar with the educational landscape of Pakistan. Their interrogations engendered a valuable discussion where I also shared the challenges I faced when I translated the recordings directly into English transcripts. I had annotated these first three transcripts to allow my supervisors to assess my analytical process. However, translating the interviews proved complicated as several phrases exist in Urdu which hold cultural meaning and cannot be directly translated into English. I struggled to translate such phrases and came up with the best possible translations. I therefore annotated the transcripts because the English phrases could not capture these cultural meanings. After several in-depth discussions with my supervisors and reiterations of the transcriptions, we decided that that since my second supervisor could understand Urdu, I would transcribe the interviews into Urdu, without translating them into English and my second supervisor would read through them to give me feedback. However, if I also wanted a more in-depth discussion with my first supervisor, I would then also translate some sections or quotes into English.

Since the interviews were bilingual, I also decided that I would not ‘tidy up’ my participants’ English, and I would transcribe the way they spoke. My transcripts are therefore bilingual, combining Urdu with English. However, due to my own linguistic limitations around writing correctly and typing in Urdu, I transcribed in Roman Urdu, which uses English language characters and does not require a special keyboard. Because I sought to remain true to my participants’ words for as long as I could, I continued to use my Roman Urdu transcripts to analyse my data. I used a thematic analytical approach and coded my data manually and my excerpts are drawn from my Roman Urdu transcripts.

Transcription and Analysis – Oyinloye

I had two stages of fieldwork, thus, transcription occurred, to some extent, in the two stages. During the more extensive first stage, I began to transcribe the audio recordings with the aid of a local (paid) translator who helped transcribe recordings into Yorùbá in a notebook, with the intention that I would later translate them into English. During this process, the translator and I listened to the recordings together to ensure shared understanding, particularly for some less audible parts. Delaying translation in this way was an attempt to continue dialogue in the source language. However, I stopped after a few weeks into the process due to the physical and mental exhaustion of almost daily fieldwork and weekend (as well as sometimes weekday) translation.

At the end of the first stage, after I returned from fieldwork, I simultaneously translated and transcribed the audio recordings into English; and transcribed the English field voice notes. Given the qualitative coding analytical method I sought to use would be carried out in English given my own limitations in ‘proper’ written Yorùbá, this time I did not first physically or digitally translate into Yorùbá, but rather first mentally translated into Yorùbá and thereafter, digitally transcribed them into English, a process referred to as ‘double rendering’ by Nikander (2008). To maintain connection to and make visible the source language, the transcription process was highly detailed. It transpired over the seven months prior to the second follow-up stage of the fieldwork and included intonations and filler Yorùbá words. Notably, as part of an intentional linguistically reflexive act, the transcripts reflected a combination of both lexical (literal / direct) – word-for-word – and conceptual – equivalents of ideas at the sentence level – translations (Cormier, 2018), and sought to retain participants’ cultural tones. Moreover, during transcription (and translation), I occasionally consulted the local Yorùbá contacts who helped clarify meanings of specific terms or phrases encountered interview recordings or my own audio field notes. I also kept a document with notes about salient Yorùbá words, phrases and proverbs, all written in Yorùbá alongside their English interpretations.

Unlike the first stage of fieldwork, there was no transcription during the second, shorter stage of fieldwork. However, the post fieldwork transcription process unfolded in the same way as the first: I mentally translated into Yorùbá, then digitally translated into English, while taking note of salient Yorùbá terms, phrases and proverbs and clarifying understanding with local contacts where required. I analysed the data using the capability approach and thematic analysis (Braun & Clarke, 2006). The first step of the latter is transcription, hitherto discussed, while subsequent steps include the generation of codes – words or short phrases that capture the essence of a portion of data (Vanover et al., 2021) – and the identification of themes – patterned responses or meanings aggregated from codes (Braun & Clarke, 2006). To do this, I coded the digitised English language transcripts in the qualitative analysis software Nvivo and used word processing software to aggregate codes and identify themes. As such, my thematic analytical process involved only English.

Reporting – Hamid

As discussed above in the section of Transcription and Analysis, the final decision I made with my supervisors was to not translate my recorded interviews into English transcripts. Using Roman Urdu, I transcribed all my interviews into Urdu and used these Urdu transcripts to code my data for thematic analysis. There was constant back and forth as I used my Roman Urdu transcripts and developed themes and codes in English for the purpose of data analysis. I juggled both languages simultaneously, which was representative of the interviews themselves where my participants easily slipped between the two languages to express themselves.

However, once I started writing my findings chapters, I faced the challenge of translation again. In not wanting to exclude my participants’ words completely, I was not in favour of presenting their words in English, and I was not in the favour of (again) ‘tidying up’ their speech. However, given the requirement to report in English, I decided to translate my participants’ quotes as closely as I could and provide original transcripts in the appendix so that Urdu-speaking readers may refer to the original conversations. When discussing the quotes, I noted instances where participants used English words, especially when these words had significance. Here is an example from my thesis:

I am coming from that background where even in our families and stuff anyone who has really good English, we say they are really intelligent.

Like all the interviews with my participants, Aiman’s interview was conducted bilingually. In the above quote, she used the word ‘background’ in English so that word has not been translated. The word ‘background’ which Aiman (pseudonym) used was significant because she was referring to coming from a background where people automatically assume that fluency in English means intelligence. It was therefore important to mention that this word was not a translation but quoted as spoken by the participant. There were various such instances where participants used English words which had contextual significance, especially in relation to the conversation. In the appendix, Aiman’s quote is presented in roman Urdu as follows:

Because like I am coming from that background toh even hamare background even in our families and stuff jis ki english bohot achi hai hum kehte hain yeh intelligent hai. So even meri english chalein ab behtar hai abhi unko kaisa pata kay may intelligent hun.

Reporting – Oyinloye

Part of the thematic analytical method I used consisted of the selection of extracts to illustrate the themes identified, ensuring relevance to the study’s research questions and the literature. Thus, to illustrate specific themes, I selected extracts from the existing English transcripts. Moreover, to illuminate salient terms and add emphasis, my extracts often included Yorùbá words, written with their accentuation and followed in parentheses by the English interpretation. Sometimes, the English term was written first followed by the Yorùbá term in parentheses. Where English terms were used by participants – typically by teachers – they were retained in the extracts and accompanied by single quotation marks.

To retain the sense of the excerpts in the source language, my transcriptions typically occurred in a lexical-conceptual spectrum, and sometime veered towards the lexical. Where this occurred and it was evident that the English excerpts may not be immediately perceptible to a ‘native’ English speaker, rather than use complete conceptual translations, I retained the more lexical translation and included anglicised conceptual equivalents in parentheses. For instance, in the following excerpt by a teacher who described the ubiquity of informal apprenticeships in one of the communities, “Because another one might read now (i.e., go to school) …”, ‘go to school’ is the conceptual equivalent of the more lexical ‘might read now’. To my white British supervisors, the lexical translations which required conceptual equivalents read awkwardly, and they questioned their inclusion in the extracts and in my thesis. However, I defended their inclusion as a reflexive and a political act, one which sought not to conceal that the quoted interactions had not originally occurred in English. On one hand, this transferred power from the researcher to the reader to assess the accuracy of my translation and thus the trustworthiness of my research. On the other hand, as an act of critical linguistic reflexivity, it made visible participants’ linguistic worlds and laid bare the connection between myself as a researcher, my research ideas, and participants’ sense making of those ideas.

In other instances, additional words were used in square parentheses to further grammatical sense-making in English. As already alluded, some extracts also included common spoken fillers words such as ‘o’ and expressions such as ‘uhhnn’, ‘ehnn’ or ‘ah!’ which deepened the authenticity of participants’ words. The following illustrates an extract which has some of these elements (the translated words of a parent who sought to explain the differences between formal and elementary Islamic schooling):

…. If a child doesn’t come to ilé kéwú (elementary Islamic school) in three months, …where s/he’s left her tírà (lesson book) …is where s/he’ll find it…. That of school is not like that o! … You see that of school, the child who doesn’t come this week has touched the ground (i.e., fallen behind)! When it’s next week, when s/he comes on Monday today, it’s what the others are doing today is what s/he’ll be doing o…

Finally, to accord linguistic and epistemological space to participants more generally, I included a glossary of the key Yorùbá terms used in the thesis either in quotes or as explanations; and I included, as a precursor to each findings chapter, a Yorùbá proverb which surmised the essence of the chapter.

Discussion and Conclusion

The above reflections highlight how decisions about language played an important role at key stages in our research, from the fieldwork to transcription and analysis as well as reporting. During each of these stages, our decisions were underpinned by our linguistic reflexivity which deepened our understanding of the potential of language to either build trust or create barriers while shaping the power dynamics between ourselves and our research participants.

Our reflection demonstrates the concerns with which we grappled during our decision-making, concerns which ranged from the analytical and the methodological, to the political, ethical and the epistemological. Methodologically, for example, despite our stronger linguistic competencies in English, our positionalities informed our decisions to dialogue with participants in their preferred language. For Oyinloye, this was most participants’ only known language, and one in which they possessed a stronger linguistic competence while for Hamid, it was languages in which participants possessed either a strong or a weaker profile as they did not always choose to converse in their first language. Particularly notable is Hamid’s decision not to have a conversation about language choice, even in a situation of multiple linguistic possibilities. Evidently, these decisions not only enabled participants to more richly transmit their worlds (Qun & Carey, 2023) with all its tensions and contradictions, e.g., code-switching urban young Pakistanis, it also served a decolonial aim. That is, it enabled us to alter the linguistic and intellectual relations of power (Veronelli, 2015) whereby we shifted the linguistic and overall power dynamic from us (the researchers) to them (our participants).

Linguistic decisions during analysis were similarly crucial, as it was important for us to remain and thereby, analyse, in the source language for as long and as much as we could (Santos Jr et al., 2015). For Oyinloye, this meant extensive periods of mental dialogue with the source language during the extended transcription process, and English-cum-hybrid transcripts which included Yorùbá filler terms, key phrase retention in Yorùbá, lexical-cum-conceptual translations, and notes of important Yorùbá proverbs, all of which deepened the analysis. For Hamid, this meant transcription into Roman Urdu, and analysis of these transcripts, a departure from the typical practice of transcribing into and analysing in English. As Pereira (2021) notes, discussions of linguistic incommensurability ought to move beyond an assumption of loss, e.g., of meaning during the translation process. Instead, they should move towards an appreciation of the opportunities created, e.g., for novel or deeper conceptual and methodological insights as well as trustworthiness when the incommensurable enables an interrogation of the linguistic meaning-making process.

In our work, the political was decolonial, with both of our supervisors proving essential to the ways in which we employed linguistic reflexivity towards more political (and also, decolonial) aims, to render visible often invisible source languages (Temple & Young, 2004). For Oyinloye, this occurred at the reporting stage where her white British supervisors questioned the awkwardness of the English excerpts in her thesis. Rather than dissuade her, their intervention resulted in an explicit account of the linguistic reflexive, political and decolonial rationale underpinning this act. For Hamid, it ensued during transcription where her white and British-Pakistani supervisors questioned the smoothness of her English transcripts, resulting in a linguistically reflexive move to render her transcripts entirely in Roman Urdu. As can be deduced, these examples of the visibilising, i.e., inclusionary, work carried out at these latter stages sustained both authors’ prior reflexive praxis of including participants’ languages where and to the degree feasible, e.g., translation of fieldwork documents and consent forms; conduct of interviews in participants’ languages (there this was chosen or where there was no alternative); transcription into source language; presentation of quotes in source and English languages; etc. These scenarios across the various stages of the research underscore the need for researchers, supervisors and other collaborators to continually and collectively examine linguistic decisions as well as assumptions to deepen linguistic reflexivity towards greater linguistic justice.

Evident with Oyinloye’s account is the limit of linguistic reflexivity and the importance of a fair assessment of one’s own linguistic capabilities. Forcing linguistically reflexive practice reinscribes rather than resolves issues of linguistic relations of power and suggests performativity rather than authenticity. As such, we encourage researchers to practice linguistic reflexivity in relation to their linguistic competence while being transparent about the linguistic decisions they make and why. Similarly, with Hamid’s account, there were limits to linguistic capability in transcribing the interviews in the Urdu script. Transcriptions were therefore typed in Roman Urdu, which uses the English alphabet to type Urdu transcripts.

As the reflections demonstrate, attending to the ways in which we presented participants’ worlds through their words serves ethical and, connectedly, epistemological functions (Pereira, 2024). On the one hand, our choices of conceptual equivalents and inclusion of proverbs positioned participants as ‘rational’ rather than ‘simple’ communicators who possess a language which expresses knowledge (Veronelli, 2015). On the other hand, our disinclination to smooth out ‘awkward’ English excerpts renders explicit the challenges of source language to English translations and makes visible participants’ own voices. Speaking for (and writing about) others, as we do through research, is a political act (Alcoff, 1991; Back & Solomos, 1993; Wilkinson & Kitzinger, 1996). Moreover, as Pereira et al. (2009) suggest, representing participants in languages they have not spoken in brings with it ethical questions, including how to grapple with language differences and translation. At the least, we suggest that this requires transparency and explicitness about the linguistic decisions we consider and eventually make at key stages of our research.

Fundamentally, our linguistic reflexivity prompted analytical, methodological, ethical and epistemological concerns and choices which enabled us to forge more meaningful connections with participants in place. This reflexivity was underpinned by our positionalities, in turn constituted by our knowledge of participants’ places, among other elements. Researchers’ knowledge of participants’ places (which we argue is accessible even by researchers not from such places), particularly the intersections between language and culture, can thus meaningfully contribute to a relational, placed-based, linguistically reflexive praxis. As we have demonstrated, such praxis ensures that research is informed by researcher(s)’ knowledge of and connection with others in the context of the research; individually or collectively interrogates the linguistic decision-making process and attends to issues of power inherent in those decisions; seeks to visibilise participants’ languages through different stages of the research; and ultimately, approaches linguistic reflexivity with an authentic rather than performative ethos.

While relationality is critical in any multilingual research, it is perhaps even more critical in cross-cultural multilingual research where issues of language intersect with historical, social and economic ones, as is the case in the two contexts discussed in this paper. A linguistically reflexive praxis therefore offers the potential to disrupt coloniality (McKinney, 2020; Veronelli, 2015) by enabling us as researchers to use our words as counter-hegemonic, self-liberatory tools (hooks, 1994). As our research demonstrates, a relational, place-based praxis is grounded in participants’ places, a view which transcends geography to include cultural, epistemological and the philosophical dimensions of place. This view also traverses geography given researchers continue their dialogue with participants’ worlds long after they may have exited the latter’s geographic contexts. Notably, our research has also shown that linguistic decisions are not only made by researchers. Bi- or multilingual participants may also make linguistic decisions, influenced by their perceptions of researchers’ positionalities or their own motivations (see also Cormier, 2018).

Inadequate discussions or complete omissions of researchers’ linguistic approaches (or lack thereof) abound within reported methodological accounts. The in-depth reflections in the paper may be used by researchers, particularly those conducting cross-cultural, multilingual research, to consider the ways in which they may engage with language, moving beyond solely technical and logistical fieldwork concerns to also address issues of power and representation toward forging an authentic connection with participants in place. To aid this process, we suggest some questions in the table below which may be used as a guide to integrate a linguistically reflexive praxis. Although we conducted research as individual researchers, these questions may also be considered in the context of research teams. Moreover, we have included a consideration around dissemination, as an extension of the reporting stage.

Overarching questions to consider
To what extent do the epistemological assumptions of the research enable a linguistically reflexive methodological approach?To what extent has a linguistically reflexive approach (e.g., a relational, place-based approach) been considered, articulated and, where relevant, included in the institutional ethics application process? How does the approach make visible participants’ voices, worlds and deepen trustworthiness?Does the approach developed consider researcher(s)’ positionalities (including linguistic capabilities), as well as the time and resources available?What linguistic norms or practices will the research align or contend with in the context of the research?
Research stage	Questions to consider
Data collection (and pre-fieldwork)	Will fieldwork documents require translation, or can they be enumerated in participants’ language(s)? Which specific documents will need to be translated, for whom, and why?Based on participant demographics and researchers’ (or contacts’) knowledge of participants’ culture and language, what language(s) may be used to interact with participants, and who will carry out these interactions? Where an interpreter is used, do the researcher(s) and interpreter(s) have a shared understanding of linguistic reflexivity and its application in the research?Where interpretation or translation will be used at any point during this stage, whether the research is carried out by an individual or a team, have the required time, resources and researcher support been considered?Where participants are bi or multilingual, can they be given a choice of the language(s) in which they may interact with the researcher?How do the pre-fieldwork linguistic decisions affect the research design and fieldwork plan?
Transcription & analysis	Will the researcher(s) conduct the transcription (and translation, where required) – if not, how may this impact the transcript generated, and the analytical process?Where translation of transcripts is required, at what point does it occur and why; and in what ways are participants’ words or worlds included or made visible in the transcription process? How does this affect attempts to make participants’ voices visible in the research outputs?Will the analysis be conducted in the source interview language(s)? If yes or if not, how may this impact the analytical process and therefore findings?Will the transcripts (translated or source) be shared with or reviewed by persons other than the researcher(s)? If so, how does this review process contribute to linguistic meaning-making?
Reporting	Does the research report include a sufficient account of the researcher(s)’ linguistic decision-making processes across the key stages of the research (this includes the use of interpretation and translation by others)?If different parts or forms of data will exist in different languages, how will the researcher bring these together analytically (this refers less to unifying the ‘language’ but analysing consistently)?Has thought been given to the forms of written or non-written outputs, and the languages in which they will be produced? Have participants and other relevant stakeholders been consulted in these decisions?

We hope that our paper encourages researchers to begin their research with linguistic intentionality, so that they may think critically about the use and exclusion of language (Rolland et al., 2023) throughout the key stages of their research.

Footnotes

Acknowledgements

The authors would like to thank the reviewers for their constructive feedback.

ORCID iD

Amal Hamid

Ethical Considerations

The study upon which this paper is based received ethical approval from the authors’ respective institutions. Reference for ethics:Author 1: 2022-14175-24312,Author 2: HREC/3021/Oyinloye.

Funding

The authors received no financial support for the research,authorship,and/or publication of this article.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research,authorship,and/or publication of this article.

References

Alcoff

(1991). The problem of speaking for others. Cultural Critique, 20, 5–32. https://doi.org/10.2307/1354221

Back

Solomos

(1993). Doing research, writing politics: The dilemmas of political intervention in research on racism: Doing research, writing politics. Economy and Society, 22(2), 178–199. https://doi.org/10.1080/03085149300000010

Braun

Clarke

(2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3(2), 77–101. https://doi.org/10.1191/1478088706qp063oa

Brislin

R. W.

(1970). Back-translation for cross-cultural research. Journal of Cross-Cultural Psychology, 1(3), 185–216. https://doi.org/10.1177/135910457000100301

Carling

Erdal

M. B.

Ezzati

(2014). Beyond the insider–outsider divide in migration research. Migration Studies, 2(1), 36–54. https://doi.org/10.1093/migration/mnt022

Chapman

(2023). Revocalising human geography: Decolonial language geographies beyond the nation-state. Progress in Human Geography, 47(1), 24–42. https://doi.org/10.1177/03091325221131852

Chen

H. Y.

Boore

J. R.

(2010). Translation and back‐translation in qualitative nursing research: Methodological review. Journal of Clinical Nursing, 19(1-2), 234–239. https://doi.org/10.1111/j.1365-2702.2009.02896.x

Cormier

(2018). The language variable in educational research: An exploration of researcher positionality, translation, and interpretation. International Journal of Research & Method in Education, 41(3), 328–341. https://doi.org/10.1080/1743727x.2017.1307335

Cortazzi

Pilcher

Jin

(2011). Language choices and ‘blind shadows’: Investigating interviews with Chinese participants. Qualitative Research, 11(5), 505–535. https://doi.org/10.1177/1468794111413225

10.

Douglas

S. P.

Craig

C. S.

(2007). Collaborative and iterative translation: An alternative approach to back translation. Journal of International Marketing, 15(1), 30–43. https://doi.org/10.1509/jimk.15.1.030

11.

Dunford

(2017). Toward a decolonial global ethics. Journal of Global Ethics, 13(3), 380–397. https://doi.org/10.1080/17449626.2017.1373140

12.

Erikson

(2017). A history of qualitative inquiry in social and educational research. In Denzin

Lincoln

(Eds.), The SAGE Handbook of Qualitative Research (5th edn., pp. 36–65). Thousand Oaks, CA: SAGE Publications, Inc.

13.

Gani

J. K.

Khan

R. M.

(2024). Positionality statements as a function of coloniality: Interrogating reflexive methodologies. International Studies Quarterly, 68(2), sqae038. https://doi.org/10.1093/isq/sqae038

14.

Goitom

(2019). Multilingual research: Reflections on translating qualitative data. British Journal of Social Work, 50(2), 548–564. https://doi.org/10.1093/bjsw/bcz162

15.

Greer

(2003). Transcription approaches to multilingual discourse analysis. In: Conference proceedings of the 1997 symposium Jalt Pan-SIG conference, Kyoto, Japan.

16.

Halai

(2007). Making use of bilingual interview data: Some experiences from the field. Qualitative Report, 12(3), 344. https://doi.org/10.46743/2160-3715/2007.1621

17.

Lizzi

O. M.

Leon

(2016). English as a medium of instruction in postcolonial contexts: Moving the debate forward. Comparative Education, 52(3), 277–280. https://doi.org/10.1080/03050068.2016.1185251

18.

Lopez

G. I.

Figueroa

Connor

S. E.

Maliski

S. L.

(2008). Translation barriers in conducting qualitative research with Spanish speakers. Qualitative Health Research, 18(12), 1729–1737. https://doi.org/10.1177/1049732308325857

19.

McKinney

(2020). Decoloniality and language in education: Transgressing language boundaries in South Africa. In The Dynamics of Language and Inequality in Education: Social and Symbolic Boundaries in the Global South, 115.

20.

McKinney

Molate

2022. Coloniality & family language policy in an African Multilingual Family. In Wright

Higgins

(Eds). Diversifying Family Language Policy. London: Bloomsbury. https://www.bloomsbury.com/uk/diversifying-family-language-policy-9781350189904 Coloniality and Family Language Policy in an African Multilingual Family . Available from: https://www.researchgate.net/publication/357497124_Coloniality_and_Family_Language_Policy_in_an_African_Multilingual_Family[accessed [accessed Sep 09 2025].

21.

McNess

Arthur

Crossley

(2015). ‘Ethnographic dazzle’and the construction of the ‘Other’: Revisiting dimensions of insider and outsider research for international and comparative education. Compare: A Journal of Comparative and International Education, 45(2), 295–316. https://doi.org/10.1080/03057925.2013.854616

22.

Nikander

(2008). Working with transcripts and translated data. Qualitative research in psychology, 5(3), 225–231.

23.

Nowicka

Cieslik

(2014). Beyond methodological nationalism in insider research with migrants. Migration Studies, 2(1), 1–15. https://doi.org/10.1093/migration/mnt024

24.

Oyinloye

(2021). Towards an Ọmọlúàbí code of research ethics: Applying a situated, participant-centred virtue ethics framework to fieldwork with disadvantaged populations in diverse cultural settings. Research Ethics, 17(4), 401–422. https://doi.org/10.1177/17470161211010863

25.

Pereira

M. d. M.

(2021, June 9). Translating the field: Qualitative research across languages [The Ethnography Atelier]. YouTube. Retrieved May 09, 2024, from https://www.youtube.com/watch?v=BaTjFCoeYZc

26.

Pereira

L. S. M.

(2024). Decolonizing the internationalization of higher education in the global south. Routledge.

27.

Pereira

M. d. M.

Scharff

Marhia

(2009). Interrogating language difference and translation in social science research: Towards a critical and interdisciplinary approach. Graduate Journal of Social Science, 6(3), 1–12. https://eprints.whiterose.ac.uk/id/eprint/43440/2/0906.3a01-Final.pdf.

28.

Qun

Carey

(2023). Translating interviews, interpreting lives: Bi-lingual research analysis informing less westernised views of international student mobility. Qualitative Research, 24(2), 324–343. https://doi.org/10.1177/14687941221149588

29.

Regmi

Naidoo

Pilkington

(2010). Understanding the processes of translation and transliteration in qualitative research. International Journal of Qualitative Methods, 9(1), 16–26. https://doi.org/10.1177/160940691000900103

30.

Rolland

King

H. M.

Lorette

(2023). Methodological implications of participant and researcher multilingualism: Making language dynamics visible. Journal of Multilingual and Multicultural Development, 44(8), 645–656. https://doi.org/10.1080/01434632.2023.2224774

31.

Santos Jr

H. P.

Black

A. M.

Sandelowski

(2015). Timing of translation in cross-language qualitative research. Qualitative Health Research, 25(1), 134–144. https://doi.org/10.1177/1049732314549603

32.

Shamim

Qureshi

(2013). Informed consent in educational research in the South: Tensions and accommodations. Compare: A Journal of Comparative and International Education, 43(4), 464–482. https://doi.org/10.1080/03057925.2013.797729

33.

Shklarov

(2007). Double vision uncertainty: The bilingual researcher and the ethics of cross-language research. Qualitative Health Research, 17(4), 529–538. https://doi.org/10.1177/1049732306298263

34.

Sinaiko

H. W.

Brislin

R. W.

(1973). Evaluating language translations: Experiments on three assessment methods. Journal of Applied Psychology, 57(3), 328–334. https://doi.org/10.1037/h0034677

35.

Siwale

(2015). Why did I not prepare for this? The politics of negotiating fieldwork access, identity, and methodology in researching microfinance institutions. Sage Open, 5(2), Article 2158244015587560. https://doi.org/10.1177/2158244015587560

36.

Sultana

(2007). Reflexivity, positionality and participatory ethics: Negotiating fieldwork dilemmas in international research. ACME: An international journal for critical geographies, 6(3), 374–385. https://doi.org/10.14288/acme.v6i3.786

37.

Sutrisno

Nguyen

N. T.

Tangen

(2014). Incorporating translation in qualitative studies: Two case studies in education. International Journal of Qualitative Studies in Education, 27(10), 1337–1353. https://doi.org/10.1080/09518398.2013.837211

38.

Tarozzi

(2013). Translating and doing grounded theory methodology. Intercultural mediation as an analytic resource. Forum Qualitative Sozialforschung/Forum: Qualitative Social Research, 14(2), https://doi.org/10.17169/fqs-14.2.1429

39.

Temple

Young

(2004). Qualitative research and translation dilemmas. Qualitative Research, 4(2), 161–178. https://doi.org/10.1177/1468794104044430

40.

Van Nes

Abma

Jonsson

Deeg

(2010). Language differences in qualitative research: Is meaning lost in translation? European Journal of Ageing, 7(4), 313–316. https://doi.org/10.1007/s10433-010-0168-y

41.

Vanover

Mihas

Saldaña

(2021). Analyzing and interpreting qualitative research: After the interview. Sage Publications.

42.

Veronelli

G. A.

(2015). The coloniality of language: Race, expressivity, power, and the darker side of modernity. Wagadu: A Journal of Transnational Women’s & Gender Studies, 13(1), 5. https://digitalcommons.cortland.edu/wagadu/vol13/iss1/5.

43.

Wilkinson

Kitzinger

(1996). Representing the other: A feminism & psychology reader. Sage.

44.

Wilson

S. a.

Linguistic Reflexivity in Multilingual Research

Abstract

Keywords

Introduction

Reflexivity, Translation, and Linguistic Reflexivity

Reflexivity

Translating Multilingual Data

Linguistic Reflexivity

Linguistic Decisions in Cross-Cultural Research

Fieldwork

Transcription and Analysis

Reporting

Linguistic Decision-Making in Practice: Reflections from Our Research

Linguistic Positionality – Hamid

Linguistic Positionality – Oyinloye

Language and Fieldwork – Hamid

Language and Fieldwork – Oyinloye

Transcription and Analysis – Hamid

Transcription and Analysis – Oyinloye

Reporting – Hamid

Reporting – Oyinloye

Discussion and Conclusion

Footnotes

Acknowledgements

ORCID iD

Ethical Considerations

Funding

Declaration of Conflicting Interests

References