Research Methods in Applied Linguistics

This article argues that translanguaging itself is a methodology oﬀering a new conceptual framework that promotes a number of important analytical shifts: shift away from language as abstract codes to meaning-and sense-making; attend to a wider range of multi-semiotic resources whilst refusing to privilege particular modes and methods of meaning-making over others; approach translanguaging as an expansively integrated experience. This methodological framework prompts us to ask diﬀerent research questions and to ﬁnd alternative ways of thinking and talking about data and methods of data collection and analysis. The article also reiterates my motivations for Moment Analysis that argues for spur-of-the-moment actions to be treated as signiﬁcant data points in understanding the rhythm and meaning of social life.


Jerry
forum piece is timely and its implications more far-reaching than it may appear.For me, the call for translanguaging research methodologies is a call to transform the way we think about and talk about doing applied linguistics research, to reject abyssal thinking ( García et al., 2021 ), and to decolonise methodology.With regard to the specific topic of translanguaging, Lee's piece reflects growing frustrations (yes, plural) from different camps.First, despite repeated attempts by some to brush it off as nothing different or new, translanguaging shows no sign of losing any discourse space.Then, there is the frustration of those who like the term translanguaging but see it as primarily a descriptive label for language mixing practices (e.g., in classrooms, everyday interaction, digital media).They feel that the leading proponents of translanguaging have failed to offer any precise "method " of either collecting or analysing translanguaging data.For instance, people talk about translanguaging as transcending boundaries not only between named languages but also between language and other semiotic systems, yet researchers still use transcription conventions and procedures that fail to capture the dynamism and multimodality of their data; and advocates of translanguaging pedagogies have failed to show convincing evidence that the pupils learn the target language better in second or foreign language classes if they are allowed to use more of their L1s (but see Li & Garcia, 2022 , for a discussion of why translanguaging is not about the use of L1 in language teaching and learning).Lee's discussion also reflects the frustrations of translanguaging theorists, for want of a better term, for whom translanguaging is already a methodological perspective; it urges people to do precisely what Lee is calling for, i.e., to find a way of doing applied linguistics research and a language to talk about it (another sense of translanguaging) that is radically different from the conventional approaches and methods.
In this short commentary, I elaborate on some of the premises of translanguaging methodology -I take methodology as the conceptual framework for how to do research and methods as specific strategies, tools and techniques of collecting and analysing evidence.Methodology determines the way research questions are asked, which in turn determines what evidence is needed to answer the questions in specific ways, and what methods may be appropriate to gather the evidence and analyse it.In this sense, the title of this https://doi.org/10.commentary should be Translanguaging as Methodology .But I am paraphrasing Chen Kuan-hsing's book title, Asia as Method: Towards Deimperialisation , to index the decolonising agenda of the translanguaging project ( García et al, 2021 ).
As I explained in Li (2014) , I see the objective of doing applied linguistics as understanding the experience of being human, and I see languaging -i.e., meaning-and sense-making through orchestrating "an assemblage of diverse material, biological, semiotic and cognitive properties and capacities…in real-time and across a diversity of timescales " ( Thibault, 2017 , p. 82) -not language, as the heart of human sociality (see also Halliday, 1985 ). Translanguaging focuses firmly on the lived experiences of the vast majority of the human race (i.e., bilinguals and multilinguals), those who have been socialised into knowing and using patterns of linguistic structures, bodily movements, and symbolic practices that scholars identify and assign to differently named languages and cultures.Translanguaging scholars regard named languages as political and ideological constructs.The human brain does not store different named languages in different parts of it; there is no switch that manages the on-and-off of named languages; and the so-called cognitive representation and processing of different named languages -certain distinctive features -are consequences of socialisation and lived experiences.And it is important to point out ( Li, 2016( Li, , 2018 ) ) that human beings have an innate drive to go beyond the boundaries between named languages set by linguists, as well as the boundaries between language and other semiotic resources that they have access to in making meaning and making sense.
From this methodological perspective, then, the focus of applied linguistics research, especially of bilingualism and multilingualism research, should be on how humans orchestrate the diverse range of what Thibault (2017) calls material, biological, semiotic, and cognitive properties and capacities in real-time as well as across different timescales, transcending the artificial and ideological boundaries between named languages and between language and other semiotic means of meaning-and sense-making.This necessitates a number of important methodological shifts: 1. Shift attention away from language as abstract codes to focus on meaning-and sense-making through tran s-languag-ing.It is neither I-language nor E-languages that should be the focus of applied linguistics research, in contrast to theoretical and formal linguistics, but languaging and indeed translanguaging as action and situated meaning-making.In fact, Halliday argued some time ago that language as a complex dynamic social semiotic system offers a "meaning potential, " and linguistics is the study of how people exchange meanings by languaging " (1985).2. Attend to a wider range of multi-semiotic resources whilst refusing to privilege particular modes and methods of meaningmaking over others.To take the orchestration metaphor a little further: it is not quantity but quality that matters in music making -percussionists are just as important as string players in meaning-making in music.Different orchestrations lead to different meanings.In making sense of the meaning of a piece of orchestral music, one would not be listening to one instrument only or one instrument at a time.The same applies to translanguaging; a seemingly slight change in facial expression, body posture, or font size and style can change the meaning of a message just as significantly as a change in the choice of named languages.Of course, the same piece of music could have different arrangements and each performance can have its own specific on-the-spot effect.Translanguaging as an analytical lens focuses on the performative nature of languaging behaviour and is grounded in the doing of sense-and meaning-making.3. Approach translanguaging as an expansively integrated experience.Analytically, the focus must not be on one means of meaningand sense-making (linguistic or otherwise) only or one at a time, but the assemblage and orchestration of diverse means.We therefore need a maximalist approach that focuses on the simultaneous management of multiple means of meaning-and sensemaking, and includes the role played by such things as feeling, experience, history, memory, subjectivity, as well as ideology and power in (trans)languaging.
It follows that the kinds of research questions one asks under this methodological framework may be quite different from what have been asked so far within conventional methodological paradigms.For example, personally I am interested in: • How do the political entities of named languages get cognitively represented?And what constitutes "cognitive representation "? • Why is access to the diverse range of properties and capacities for translanguaging not equal both synchronically across different individuals, communities, and cultures and diachronically across different historical times?• What is the role of one-off incidence or spur-of-the-moment action in meaning-and sense-making?I recognise that these questions are still pretty conventional, in the sense that they are connected to established research paradigms in psycholinguistics, sociolinguistics, and linguistic pragmatics, even though they may not have been asked in these specific ways.What I believe Lee wants us to do is a fundamental epistemological paradigm shift to delegitimise the methodological status quo.A frustration I have experienced is academic journal reviewers' and editors' insistence that a research paper must have a methods section, sometimes erroneously termed as a "methodological section, " where the author should explain explicitly how many participants were involved and how many hours of data were collected.I have found it deeply unsatisfactory to have to declare for my own papers that "This is a critical essay, not a report on a single piece of empirical research.Nevertheless, we draw data from …" ( Li & Zhu, 2021 , p. 739) and "This critical essay is not intended to report any empirical project as such.The two cases we describe below are exemplars of …" ( Li & Garcia, 2022 , p. 2).It is not that I could not provide the precise numbers -that is the easiest part -but the kind of research I am engaged in does not typically have a clear start and end date, and neither can it be represented by such binaries as "researcher/researching " and the "researched " (see further, Li et al., 2020 ).In fact, if the analytical focus is shifted to the assemblage and orchestration of diverse meaning-making means, we will also need to consider radically different formats for publishing research, as some of the outlets for visual communication and multimodalities research are already experimenting.
Credit where credit is due: the "white listening subject " ( Flores & Rosa, 2015 ) excels in selective listening; it does not matter how many times we repeat that named languages exist as political and ideological constructs ( Li, 2022 ), they insist on accusing us of denying the existence of named languages.Naming languages as if they were structurally and psychologically discrete entities is intrinsically connected with the colonising one-nation-state-one-language mindset, exactly what translanguaging aims to challenge.When faced with so-called mixed language data, i.e., discourses that have elements from different named languages, one is also faced with a methodological choice: to identify and separate the elements into different named languages first or to focus on meaningmaking through the assemblage of diverse elements, with the latter constituting a translanguaging approach.
Nevertheless, my first substantive article on translanguaging, Li (2011) , did come with a specific proposal for an alternative method, namely, Moment Analysis (see also TK Lee, 2022 ).The motivation for Moment Analysis for me was entirely personal.I was trained in the quantitative variationist sociolinguistics paradigm and subsequently worked in a speech and language pathology department for many years with colleagues who use predominantly clinic-or laboratory-based experimental designs and methods.As my primary interest was the lived experiences of bilingual and multilingual individuals, I felt a need for "a paradigm shift, away from frequency and regularity oriented, pattern-seeking approaches to a focus on spontaneous, impromptu, and momentary actions and performances of the individual " (p.1224).To be clear, I do not see anything wrong in seeking patterns in individual behaviours.But behavioural patterns for me are long-term outcomes of original, momentary actions, which become recognised, adopted, and repeated by the same and/or other individuals.I was particularly interested in the spontaneous creativity bilinguals show through combining elements from different named languages and from other semiotic systems to create novel expressions in social interaction, either face-to-face or digitally ( Li, 2020 ;Li et al., 2020 ;Li & Zhu, 2019, 2021 ).Moment Analysis helps me to focus on what prompted a specific action (e.g., creating a novel expression by tranßcripting; Li & Zhu, 2019 ) at a specific moment in time and the consequences of the action including the re-actions by other people.A moment in Moment Analysis, as explained in Li and Zhu (2013 , p. 523), is identifiable with two key characteristics.
First, it is mundane yet noticeable by both the participants of social interaction and by the analyst.It is mundane in the sense that it occurs naturally in conversational interaction and is not some "critical incident " that would change the life of the individual entirely.Yet, a moment that is worth analysing must be noticeable by the participants as well as the analyst because of its creativity.Thus, a creative moment is often commented on directly by the participants either immediately or at some later point in the interaction or marked by pause for thought.Such a moment is also noticeable by the analyst whose job is to "make sense of the participants trying to make sense of their world " (Smith and Osborn 2015).Secondly, the noticeability of the moment means that a moment worth analysing has procedural consequentiality, a concept typically associated with Conversation Analysis ( Schegloff, 1991 ) where the analyst is concerned with if and how the context or the setting of social interaction has any consequences for the shape, form, trajectory, content, or character of the interaction that the parties conduct.
To investigate these moments, one can use a variety of strategies, tools and techniques: participant observation within the framework of ethnography, linguistic landscaping, walking methods ( O'Neill & Roberts, 2019 ), etc.I have emphasised the importance of metalanguaging data, i.e., commentaries on the speaker's language practices as lived experience, as they aid Interpretative Phenomenological Analysis that I am interested in doing, i.e., ''the participants are trying to make sense of their world; the researcher is trying to make sense of the participants trying to make sense of their world'' ( Smith & Osborn, 2015 , p. 25).But in practical terms, the specific methods that I have used in doing Moment Analysis are Looking, Listening, Talking, and Thinking, or LLTT ( Li & Zhu, 2013 , p. 520) -all the things I normally do in everyday life, but with a more analytical focus because I have some specific research questions to answer.
In this day and age of big data and crisis of reproducibility in science, no doubt some people will want to brush off data collected through such methods as unsystematic and anecdotal.But it is precisely systematicity and reproducibility that I seek to challenge in doing applied linguistics.As I said above, I am interested in understanding the human experience in the everyday sense.That may well be regarded by some as a non-scientific topic to start with, and that is absolutely fine by me because I am not looking for frequencies and regularities in everyday life.I am more interested in how human beings problem-solve the chaos in their daily lives by assembling and orchestrating multiple resources they have access to, i.e., translanguaging.Lived experiences are personal and time-and context-sensitive and cannot be reproduced by others in a different setting at a different time.Feelings and emotions that are so crucial in personal experiences get lost in big data and the obsession with reproducibility.We must avoid objectification of personal experiences in applied linguistics research.
Our everyday life is spontaneous, full of spur-of-the-moment actions.Moment Analysis calls for more attention to be paid to them as they can fundamentally change our courses of actions and patterns of behaviour.The descriptions of these momentary actions and events may result in what people call anecdotes.They are nonetheless significant data points, often beyond the metrics of pre-determined criteria for data collection.We all grow up with anecdotes; we learn from anecdotes, and we live by anecdotes.Anecdotes matter in our everyday life, and therefore should matter in discipline such as applied linguistics that deal with human beings' lived experiences.The epistemological stance of the translanguaging project is that it is based on the observer-analyst's subjective understanding and interpretation of what they have observed in everyday social life ( Li, 2018 ).
It seems only appropriate for me to end this commentary with an anecdote.One of the reviewers of our manifesto ( García et al., 2021 ) asked if we could italicise the Spanish and other non-English expressions and provide English translations in the final manuscript.But identifying non-English expressions and assigning them to other named languages proved to be rather complicated for us authors.The lead author, Ofelia, could identify only three Spanish expressions that might cause some difficulty in understanding by people who do not read Spanish.Other contributors found a few more, but there was a discussion as to whether any of them really needed translation because some were quite commonplace in everyday conversation amongst the Americans.Our collective decision was to make sure that the text was comprehensible to an international readership, but we resolutely did not want to italicise the so-called non-English words.And our reason is explained very vividly by the American writer Daniel José Older in this short video clip: https://www.youtube.com/watch?v = 24gCI3Ur7FM , an excellent illustration of what Translanguaging as Method entails.