2. Theoretical issues

There are two main concerns that might be raised regarding language socialization as a theoretical framework that could potentially hinder its application to EFL settings. The first is whether an approach that locates learning in socializing interactions between ‘expert’ and ‘novice’ members of a community is applicable to contexts that lie outside of what has traditionally been considered the ‘target’ (i.e., English-dominant) academic community. However, the ‘target’ community in an ADS study should not be assumed a priori, but determined empirically; that is, the aim should be to investigate what kind of community students in EFL settings are being socialized into. This community may be one governed by Anglo-American practices and norms, but alternatively it may be one in which these practices have been ‘localized,’ one in which English serves as a ‘contact language’ (Canagarajah, Reference Canagarajah2013, p. 56) among users from different language backgrounds, or one characterized by multilingual and/or multimodal practices (e.g., Canagarajah, Reference Canagarajah2013, Reference Canagarajah2018; Rose et al., Reference Rose, McKinley and Galloway2021).

In addition, the term socialization is sometimes taken as implying that ‘the appropriation of target cultural norms is always desirable, virtuous, inevitable, and complete’ (Duff, Reference Duff2002, p. 291; see also Lea & Street, Reference Lea and Street1998). Such assumptions may not always apply, especially among adolescent or adult EAL students who bring to the classroom their expertise in a range of knowledge domains from prior socialization experiences and who may not necessarily aspire to membership in an English-dominant academic community (e.g., Bronson & Watson-Gegeo, Reference Bronson, Watson-Gegeo, Duff and Hornberger2008; Duff, Reference Duff, Marriott, Moore and Spence-Brown2007; Duff & Anderson, Reference Duff, Anderson and Markee2015). However, although early language socialization research tended to focus on the seemingly unproblematic socialization of children into local community values and practices, the field has always recognized that socialization is achieved or resisted through negotiation (e.g., Schieffelin & Ochs, Reference Schieffelin and Ochs1986). More recent research has increasingly highlighted the contingent and multidirectional nature of socialization and emphasized the possibility of diverse outcomes, from full to partial to non-participation in community practices, as well as newcomers’ agency to resist or transform the beliefs and practices into which others endeavor to socialize them (e.g., de León & García-Sánchez, Reference de León and García-Sánchez2021; Duff, Reference Duff, Marriott, Moore and Spence-Brown2007; Ochs & Schieffelin, Reference Ochs, Schieffelin, Duranti, Ochs and Schieffelin2012).

3. Methodological issues

As language socialization has been adopted by researchers in disciplines beyond its origins in linguistic anthropology, a debate has arisen around what constitutes a language socialization study. Those working in the anthropological tradition have prioritized ethnographic research that includes: (a) immersion in the research setting for an extended period in order to develop understandings of local beliefs and practices and to trace changes over time, and (b) close analysis of socializing interactions that have been audio- or video-recorded (e.g., Bronson & Watson-Gegeo, Reference Bronson, Watson-Gegeo, Duff and Hornberger2008; Garrett & Baquedano-López, Reference Garrett and Baquedano-López2002; Kulick & Schieffelin, Reference Kulick, Schieffelin and Duranti2004). However, some applied linguists have argued for a more flexible approach, noting that insistence on ethnography as the sole legitimate methodology may close off potentially fruitful avenues of research where it may not be feasible (Duff, Reference Duff2010; Duff & Anderson, Reference Duff, Anderson and Markee2015). Below, I review methodological options for ADS research based on four-way taxonomy proposed by Bronson and Watson-Gegeo (Reference Bronson, Watson-Gegeo, Duff and Hornberger2008): language socialization as a topic (i.e., studies that are consistent with a focus on ‘the intersection of social life, language use, and language development’ (p. 48) but that do not adopt language socialization as a conceptual or methodological framework), an approach (i.e., studies that locate learning within its social context but may not be longitudinal), a method (i.e., studies that follow language socialization conceptual and methodological principles), and an intervention (i.e., language socialization studies that aim to effect positive change).

ADS researchers adopting language socialization as a method employ ethnography or case study and collect data from multiple sources. Ethnographic ADS research typically involves longitudinal observation and recording in one or more classrooms, schools, writing center tutorials, or other learning spaces and aims to understand the practices that mediate socialization in these communities (e.g., Beiler, Reference Beiler2020; Duff, Reference Duff2002; Friedman, Reference Friedman2019a; Talmy, Reference Talmy2008). Case studies focus on the experiences of one or more individuals, who may be observed across multiple sites or over a period of time (e.g., Anderson, Reference Anderson2021; Fujioka, Reference Fujioka2014; Kobayashi, Reference Kobayashi2016). Recorded interaction in ethnography and case study is typically analyzed at the micro-level in order to enable a detailed and empirically grounded account of socializing interactions (e.g., Duff, Reference Duff2002; Garrett & Baquedano-López, Reference Garrett and Baquedano-López2002). In addition to language, analyses may examine features such as turn-taking as well as other semiotic resources employed by participants, including non-verbal features (e.g., gestures, eye gaze), orientations towards physical objects (e.g., books, blackboards), or the multiple modalities (e.g., words, pictures, music, emojis, etc.) through which meaning is constructed in online environments (e.g., Schreiber, Reference Schreiber2015; Talmy, Reference Talmy2008). Although LS research has traditionally privileged face-to-face interaction, ADS studies using LS as a method have also examined computer-mediated environments (e.g., Hafner & Yu, Reference Hafner and Yu2020; Yim, Reference Yim2011) and written feedback (e.g., Anderson, Reference Anderson2021; Séror, Reference Séror, Byrd Clark and Dervin2014) as sites for socializing interactions. These data may be supplemented by interviews to provide an emic (insider) perspective to complement the researcher's etic (outsider) perspective and collection of relevant artifacts (e.g., syllabi, assignments, textbooks, curricular guidelines, educational policy documents, etc.) to situate socializing interactions within a larger institutional, social, or political context.

ADS research for which language socialization serves as a topic or approach may draw primarily or exclusively from indirect sources such as questionnaires, interviews, narratives, or diaries in which participants describe their current or past socialization experiences (e.g., Kim & Duff, Reference Kim and Duff2012; Okuda & Anderson, Reference Okuda and Anderson2018; Zappa-Hollman & Duff, Reference Zappa-Hollman and Duff2015). These sources can provide valuable information that might otherwise remain inaccessible to researchers; for example, interviews and writing logs enabled Zappa-Hollman and Duff (Reference Zappa-Hollman and Duff2015) to identify the major role played by peer interactions outside of the classroom (which they could not observe directly) in the socialization of Mexican undergraduates at a Canadian university.

The fourth category in Bronson and Watson-Gegeo's (Reference Bronson, Watson-Gegeo, Duff and Hornberger2008) taxonomy, language socialization as an intervention, does not prescribe any particular methodology, but rather encompasses research that addresses educational inequities and gives voices to marginalized research participants. For example, ADS studies have examined issues such as the impact of social identity categories such as race (Talmy, Reference Talmy2010a), social class (Kanno, Reference Kanno2003), gender (Morita, Reference Morita2009), or ‘ESL student’ (Talmy, Reference Talmy2008) on ADS trajectories and outcomes. Other critical ADS research has investigated how EAL students are socialized into academic discourse practices that privilege certain languages over others in multilingual English-medium educational settings (Beiler, Reference Beiler2020).

Consistent with a view of academic discourse as being socially constructed, ADS researchers typically regard research findings as co-constructed between the researcher and participants rather than a set of objective facts, a stance that requires critical reflection on the relationships between the researcher, the researched, and the knowledge being created between them (e.g., Starfield, Reference Starfield and Chapelle2013; Talmy, Reference Talmy and Chapelle2013). For example, there has been increasing awareness that data from sources such as interviews should not be treated as factual reports of participants’ experiences and feelings, but as subjective accounts produced for a specific audience at a specific moment in time for a specific purpose (e.g., Talmy, Reference Talmy2010b). Interviews themselves may also constitute a context for socialization (Friedman, Reference Friedman2019b); for example, asking interviewees to reflect on past experiences may contribute to their reframing of those experiences in ways that affect their ongoing socialization. It is therefore important to consider factors that may have influenced data collection and interpretation, such as relationships and power differentials between researchers and participants, the social identities that interviewers and interviewees display in interviews, and the interactional and social contexts in which the data were produced.

A final methodological issue is documenting the outcomes of socialization, that is, the extent to which students have taken up, resisted, or transformed the language or practices promoted through socializing interactions (e.g., Kulick & Schieffelin, Reference Kulick, Schieffelin and Duranti2004). In longitudinal ADS research, this may involve tracking changes (or documenting the lack thereof) in participants’ language, levels of participation, and/or ability to engage in valued academic practices over time (e.g., Kobayashi, Reference Kobayashi2016). In studies of written academic discourse, analyzing textual changes across multiple drafts can provide evidence of socialization into (or resistance to) the writing practices and norms promoted through feedback or other socializing interactions (e.g., Anderson, Reference Anderson2021). Technologies such as screen capture recordings (Hamel & Séror, Reference Hamel, Séror, Caws and Hamel2016) have also been used in a number of recent studies to document and analyze writing processes and how these may have been influenced by socializing messages (e.g., Beiler, Reference Beiler2020; Séror & Gentil, Reference Séror and Gentil2020). Interviews may be used to ascertain the perspectives of instructors and students on the outcomes of socialization (e.g., Anderson, Reference Anderson2021; Friedman, Reference Friedman2019a).

4. ADS and ‘cultural differences’

Since Kaplan's (Reference Kaplan1966) famous ‘doodle’ article positing the influence of ‘cultural thought patterns’ on ESL students’ academic writing, researchers have debated how ostensibly ‘cultural’ factors shape orientations towards features of academic discourse such as text organization (e.g., Chien, Reference Chien2011; Li, Reference Li2014), source attribution (e.g., Li & Flowerdew, Reference Li, Flowerdew, Pecorari and Shaw2019; Pecorari & Petríc, Reference Pecorari and Petríc2014), critical thinking (e.g., Tian & Low, Reference Tian and Low2011), critical literacy (e.g., Crookes, Reference Crookes2021), and peer feedback (e.g. Hu, Reference Hu, Hyland and Hyland2019; Yu & Lee, Reference Yu and Lee2016). Early discussions around this question drew attention to the fact that academic discourse practices that are valued in Western academia are culturally and ideologically mediated. However, they also tended to make sweeping assertions regarding ‘cultural differences’ between ‘Western’ societies, which were said to emphasize textual ownership, directness, originality, and individualism, and ‘Asian’ societies that supposedly valued collectivism, indirectness, memorization, rote learning, and deference to authority (e.g., Carson & Nelson, Reference Carson and Nelson1996; Ramanathan & Atkinson, Reference Ramanathan and Atkinson1999). These assertions were subsequently criticized as reflecting monolithic, static, and essentialist views of ‘culture’ that ignored individual variation within communities (e.g., disciplinary differences regarding source use in Western academia; see Pecorari & Petríc, Reference Pecorari and Petríc2014) and based on unsupported and possibly inaccurate assumptions about academic discourse practices in non-Western societies (e.g., Belcher, Reference Belcher2014; Kubota, Reference Kubota1999, Reference Kubota2004).

The question of how (or whether) academic discourse conventions and practices differ cross-culturally remains of interest in the field of English language teaching; however, attention has shifted towards a closer investigation of the processes through which orientations to English academic discourse are socialized in EFL settings and a more nuanced consideration of the multiple factors at the micro-, meso-, and macro-levels that shape these experiences. These may include local contextual factors such as teacher workloads and large classes (Tian & Low, Reference Tian and Low2011) or emphasis on preparing students for high-stakes exams (Curdt-Christianson, Reference Curdt-Christianson2010) rather than (or in addition to) ‘cultural’ preferences. Other studies have documented how instructors and students in EFL contexts in Asia have reshaped ‘Western’ practices such as peer review (Tsui & Ng, Reference Tsui and Ng2010) to local expectations and norms.

With its view of classroom practices as embedded within larger institutional and sociopolitical contexts, the ADS framework is well positioned to contribute to and extend this research. Questions that might be addressed include which academic discourse practices (e.g., skills-based vs. critical literacy) are privileged in specific settings; the attitudes, values, practical contingencies, and/or political imperatives that support or constrain these practices; and how teachers and/or students adopt, resist, or transform academic discourse practices that have been imposed from above or outside.

Adopting an ethnographic or case study design for this task would enable researchers to situate the focal practice (peer review, essay organization, etc.) within a larger social environment. For example, in her study at two universities in Mexico, LoCastro (Reference LoCastro, Connor, Nagelhout and Rozycki2008) contextualized a comparative discourse analysis of students’ writing in both English and Spanish with classroom observations and teacher and student interviews in order to understand how students had been socialized into academic writing conventions. Recordings of classroom interaction would allow for a detailed examination of socializing interactions, while collecting data in multiple sites (e.g., EFL classes and content-area classes) would be valuable for investigating possible variation across language, disciplinary, or institutional boundaries. When classroom observation is not feasible, interviews and/or diaries can provide insights into participants’ perspectives on past and/or current socialization experiences; for example, Ene et al. (Reference Ene, McIntosh and Connor2019) and Wette and Furneaux (Reference Wette and Furneaux2018) asked international students to compare their experiences with academic writing in English and their first language and discuss how they negotiated differences between them. Textbooks can also provide insights into the knowledge and practices that are promoted within a given academic community (e.g., Li & Flowerdew, Reference Li, Flowerdew, Pecorari and Shaw2019; Liao & Chen, Reference Liao and Chen2009). However, textbooks by themselves do not socialize students (Curdt-Christiansen, Reference Curdt-Christiansen, Duff and May2017), and textbook analyses should be supplemented with data such as classroom observation and/or interviews in order to understand how textbooks are used and interpreted within a particular setting.

Regardless of the approach taken, researchers should take care not to make facile judgements regarding the ‘cultural’ origins of observed behaviors or expressed attitudes and to keep in mind that notions of ‘cultural difference’ may be ‘socially, politically, and historically produced’ (Kubota, Reference Kubota2004, p. 29). For example, noting some researchers’ tendency to accept at face value interviewees’ culturally-based explanations of prior academic experiences (e.g., that critical thinking is discouraged in Japan in order to ‘create harmony,’ p. 22), Kubota (Reference Kubota1999) cautioned that such accounts should not be taken as ‘objective truths or individual expressions divorced from social, political, and ideological dimensions’ (p. 22), but rather understood as ideologically mediated perspectives that may have been shaped by prior socialization into discourses of cultural difference. She further encouraged researchers to deconstruct such discourses by critically examining the social, historical, and ideological forces that brought them into being.

5. ADS and ‘expert/novice’ identities

Although the idea that ‘experts’ (e.g., instructors) socialize ‘novices’ (e.g., students) has long been central to LS, the field has also recognized the fluidity of expertise across situations or even within a single interaction (e.g., Jacoby & Gonzales, Reference Jacoby and Gonzales1991; Morita, Reference Morita2000), as well as the possibility of bidirectional socialization in which ‘novices’ socialize ‘experts’ (Garrett & Baquedano-López, Reference Garrett and Baquedano-López2002; Ochs & Schieffelin, Reference Ochs, Schieffelin, Duranti, Ochs and Schieffelin2012). It is also acknowledged that expertise is predicated on asymmetries of knowledge that are, in turn, bound up with regimes of power and authority that favor certain kinds of knowledge as ‘ratified’ (Ochs & Schieffelin, Reference Ochs, Schieffelin, Duranti, Ochs and Schieffelin2012, p. 6) and mediated by ideologies that may constrain individual agency to claim expert identities and have those claims validated by others (Duff & Anderson, Reference Duff, Anderson and Markee2015; Garrett & Baquedano-López, Reference Garrett and Baquedano-López2002; Ochs & Schieffelin, Reference Ochs, Schieffelin, Duranti, Ochs and Schieffelin2012). For example, in the field of English language teaching, research has demonstrated how social identity categories such as race or ethnicity (e.g., Von Esch et al., Reference Von Esch, Motha and Kubota2020) and ‘nonnative speaker’ status (e.g., Moussu & Llurda, Reference Moussu and Llurda2008) can call instructors’ expertise into question, thus restricting their ability to serve as socializing agents (e.g., Lee & Schallert, Reference Lee and Schallert2008; Liu & Tannacito, Reference Liu and Tannacito2013). Similarly, membership in these categories may also cause students to be differentially positioned as relatively ‘expert’ or ‘novice’ compared with each other, even in settings that supposedly value multilingualism and multicultural perspectives, thus impacting their socialization (e.g., Morita, Reference Morita2009; Ou & Gu, Reference Ou and Gu2021). These findings raise intriguing questions regarding the situated, contingent, and ideological nature of expertise that might be productively explored in EFL settings, especially those that bring together individuals with diverse linguistic and educational backgrounds, thus creating an environment in which individuals may be positioned along a continuum as more or less ‘knowing’ in a particular knowledge domain. For example, Duff and Uchida's (Reference Duff and Uchida1997) study at a private language school in Japan examined how the professional identities of four EFL instructors (two American and two Japanese) were co-constructed around different kinds and levels of linguistic, cultural, and professional expertise and how these, in turn, shaped students’ responses to each instructor's teaching. Other contexts in which to further this line of inquiry would be settings where local instructors co-teach with foreign ‘native speakers’ (who may be untrained in language pedagogy) or those that include instructors and/or students who have studied abroad.

I propose two research tasks related to this topic. When researching either, it is important to remember that the process of identity construction does not take place in isolation, but is situated within specific historical and sociopolitical contexts (Duff, Reference Duff, Gass and Mackey2012, Reference Duff and Barkhuizen2017; Norton & De Costa, Reference Norton and De Costa2018). Interviews with stakeholders (e.g., instructors, students, policy makers, school administrators, parents, etc.) and collection of data regarding educational policy, school hiring standards (e.g., employment advertisements), or admissions criteria would allow for a critical analysis of what counts as ratified knowledge and who is assumed to possess it within a specific setting.

This task requires recordings of classroom interactions (instructor–student and/or student–student), ideally collected over the course of a semester or academic year in order to document any changes over time. Analysis of these interactions would trace how claims of expertise are co-constructed across turns as they unfold moment-by-moment in talk. Possible foci would be the discursive features through which claims of expertise are made (e.g., correcting or evaluating others’ utterances, using metalanguage, providing explanations, etc.), the degree of certainty with which these claims are made (e.g., falling vs. rising intonation), and the extent to which they are validated by others.

While classroom observations can provide insights into how expertise is co-constructed at particular moments in time, it can also be informative to examine how instructors’ personal histories have shaped their self-identification as ‘experts.’ Research on professional socialization in North American Teaching English to Speakers of Other Languages (TESOL) programs has provided a mixed picture regarding EAL speakers’ success in negotiating expert identities as English language instructors (e.g., Cho, Reference Cho2013; Ho, Reference Ho2011). Examining professional socialization in programs outside of English-dominant countries could provide insights regarding the extent to which struggles and negotiations over the right to claim expertise and the relevance of social categories such as ‘nonnative speaker’ are also operational in this setting. One approach to this task would be a case study that follows one or more students through a language teacher education program. Extending the study into participants’ own classrooms following graduation would allow for investigation into the extent to which pre-service socialization experiences shape actual teaching practice, an issue of particular importance given reports of mismatches between how English academic discourse expertise is construed in TESOL programs (especially those based on Western methodologies) and the realities of teaching in EFL classrooms (e.g., Casanave, Reference Casanave and Manchón2009; Lee, Reference Lee2013). Another possibility that would allow for a longer-term view of socialization experiences involves narrative analysis (e.g., Benson, Reference Benson2014) of instructors’ accounts of professional identity development before, during, and/or after their training. For example, using data from interviews and reflective diaries, Tsui (Reference Tsui2007) traced a Chinese EFL teacher's struggle to formulate a competent teacher identity at a university in Hong Kong, where his village origins and resistance to the preferred ‘communicative’ teaching methodology initially marginalized him.

6. ADS beyond the classroom

Although classrooms have long been the primary sites for ADS research, instructors are not the only, or necessarily the most competent, agents of socialization (e.g., Duff, Reference Duff, Marriott, Moore and Spence-Brown2007, Reference Duff2010; Duff & Anderson, Reference Duff, Anderson and Markee2015; Morita & Kobayashi, Reference Morita, Kobayashi, Duff and Hornberger2008). ADS studies of international students in North American universities (Kobayashi, Reference Kobayashi2003; Séror, Reference Séror2011; Zappa-Hollman & Duff, Reference Zappa-Hollman and Duff2015) and an English language program in Japan (Bankier, Reference Bankier2019) have identified peers as potential sources of support and socialization into both oral and written academic discourse. In some cases, peers were found to be more capable and influential than instructors, thus calling into question the ‘expert instructor/novice student’ dichotomy that has often been the focus in ADS research (Duff, Reference Duff, Marriott, Moore and Spence-Brown2007; Duff & Anderson, Reference Duff, Anderson and Markee2015). A related line of research has demonstrated the role of ‘literacy brokers’ (Lillis & Curry, Reference Lillis and Curry2010), such as journal editors and reviewers (Lillis & Curry, Reference Lillis and Curry2010), graduate advisors (e.g., Ferenz, Reference Ferenz2005), and language and disciplinary experts within and beyond the local academic community (e.g., Li, Reference Li2007; Lillis & Curry, Reference Lillis and Curry2010; Luo & Hyland, Reference Luo and Hyland2016), in shaping the writing of both established and novice multilingual scholars who are working in English outside of English-dominant countries.

Digital environments represent another area that has been receiving increasing research attention as settings for ADS (e.g., Reinhardt & Thorne, Reference Reinhardt, Thorne, Duff and May2017; Thorne et al., Reference Thorne, Black and Sykes2009). This work has highlighted the affordances of technology in support of academic discourse development as well as the hybridity of discourses in online spaces, which may involve use of multiple languages, ‘non-academic’ registers, or remixing material (e.g., pictures, video) from other sources (e.g., Schreiber, Reference Schreiber2015; Wu, Reference Wu2020). Interaction in digital spaces therefore has the potential for socializing users into new practices, genres, conceptualizations of text, and identities as multilingual users of English that may conflict with those promoted in the classroom (Reinhardt & Thorne, Reference Reinhardt, Thorne, Duff and May2017; Thorne et al., Reference Thorne, Black and Sykes2009).

Findings from this diverse body of research point to the need to shift from the sole focus on classrooms, which has been characteristic of many ADS studies, in favor of further consideration of the impact of socializing interactions outside of formal class settings and attention to students’ roles as active agents in seeking out alternative resources for their own socialization (Duff, Reference Duff, Gass and Mackey2012; Duff & Anderson, Reference Duff, Anderson and Markee2015). Taking up this inquiry in EFL settings would allow for investigation into a range of issues, such as how different types of expertise and knowledge (e.g., linguistic, disciplinary, cultural) are displayed and valued within these alternative spaces and how factors such as power dynamics, relative language proficiency, and personal relationships shape participation. Other critical questions include how the limited access to outside resources (e.g., private tutors, writing centers, disciplinary or language experts, Internet connections, etc.) that exists in some EFL settings affects the trajectory and outcomes of socialization, the extent to which socializing messages in these contexts support or challenge those of the classroom, and how students respond to these mixed messages.

This task would best be pursued through a case study design that follows one or more focal participants as they engage with academic discourse across spaces, interlocutors, and rhetorical situations (e.g. different assignments or genres). There are two main challenges in conducting such research. First are the practical and ethical concerns regarding observation and recording of interactions outside of classrooms, which may be spontaneous, occur in spaces where it is difficult to record, or unfold in settings (homes, professors’ offices) where the presence of a researcher might be unwelcome or invasive. For this reason, most researchers have relied on indirect data sources, such as interviews and/or logs in which writers document their experiences (e.g., Bankier, Reference Bankier2019; Kim & Duff, Reference Kim and Duff2012; Li, Reference Li2007; Zappa-Hollman & Duff, Reference Zappa-Hollman and Duff2015). However, given the limitations of self-reports, researchers should also seek out more direct sources where possible. For example, collecting written correspondence between participants and potential socializing agents (e.g., emails, reviewer comments, feedback, transcripts of online chat, etc.), in addition to self-report data, would allow for a more finely grained and triangulated analysis of the socialization process (e.g., Lillis & Curry, Reference Lillis and Curry2010; Séror, Reference Séror2011). Another option is to supply participants with recorders and ask them to record relevant interactions as they arise, as in Surtees's (Reference Surtees2018) dissertation on the peer socialization of Japanese undergraduate study abroad students at a Canadian university.

The second challenge is how to capture the complex interplay among socializing interactions that may take place across multiple spaces, individuals, and academic communities. One option is to choose a single text (e.g., a draft of a research article by a Chinese doctoral student in Li (Reference Li2007)) or project (e.g., a group oral presentation by Japanese undergraduate students in Canada in Kobayashi (Reference Kobayashi2003)) and trace its development over time (see Lillis & Curry (Reference Lillis and Curry2010) for an example of this on a larger scale). Alternatively, some ADS studies have focused on individual participants, examining the type and intensity of their interactions with members of their social networks (e.g., Bankier, Reference Bankier2019; Ferenz, Reference Ferenz2005; Lillis & Curry, Reference Lillis and Curry2010; Zappa-Hollman & Duff, Reference Zappa-Hollman and Duff2015).

7. ADS and multicompetence

In recent years, the ‘multilingual turn’ in applied linguistics has advocated for a multicompetence (Cook, Reference Cook1992) perspective to the study of language learning that repositions multilinguals as competent users of a hybrid semiotic system rather than imperfect users of an additional language (e.g., Douglas Fir, 2016; Ortega, Reference Ortega2013). One result has been increasing research attention to how multilingual students and/or instructors employ a range of languages and other semiotic resources in oral or written academic discourse (e.g., Canagarajah, Reference Canagarajah2013, Reference Canagarajah2018). There have also been calls to reframe ‘second language’ literacy as biliteracy (Gentil, Reference Gentil2011; Ortega & Carson, Reference Ortega, Carson, Silva and Matsuda2010) and to investigate how multilinguals draw strategically from their full repertoire of linguistic, rhetorical, and genre knowledge to suit the demands of different discourse communities and rhetorical contexts.

Since students in EFL contexts may be continuing to develop academic discourse competence in the local language of instruction as they simultaneously develop such competence in English, researching multilingual socialization in EFL settings could address a problem noted by Gentil (Reference Gentil2011) regarding the difficulty of finding individuals who are engaging in academic discourse in multiple languages in the monolingual environments of English-dominant universities. However, as Hall et al. (Reference Hall, Cheng and Carlson2006) noted in their critique of the multicompetence model, the attainment of multicompetence cannot be assumed even in these settings; rather, it is a variable process that emerges within specific communities of practice and is contingent on a range of factors such as the frequency and type of encounters with each language. For example, the Korean scholars interviewed by Cho (Reference Cho2010) claimed that owing to lack of attention to writing in the Korean educational system and the dominance of English in academic publishing, they had developed advanced academic literacy in English, but not in Korean. In addition, language ideologies can shape understandings regarding which languages are suitable for academic discourse (e.g., Beiler, Reference Beiler2020) or the acceptability of multilingual practices such as translanguaging in academic settings (e.g., Séror & Gentil, Reference Séror and Gentil2020).

These findings argue for research that not only examines how individuals use multilingual and material resources to create meaning, but also the micro-, meso-, and macro-level contexts in which they are socialized into the beliefs and knowledge that facilitate or constrain such practices. With its long tradition of research on multilingual socialization in postcolonial or immigrant communities (de León & García-Sánchez, Reference de León and García-Sánchez2021; Garrett & Baquedano-López, Reference Garrett and Baquedano-López2002; Moore, Reference Moore, Duff and May2017), language socialization can provide a rich theoretical framework for exploring how (or whether) students in EFL settings are socialized into multicompetence in academic discourse. Possible foci include how practices and policies at the classroom and institutional levels both shape and are shaped by ideologies about language learning and use, the value placed on the various languages in students’ repertoires as a medium for academic discourse, or the extent to which English is seen as a threat to the vitality of the local language.

As with the research tasks listed above, this task would best be undertaken through a longitudinal ethnographic or case study approach, ideally including recording of socializing interactions in or outside of the classroom, in order to capture the process of socialization as it unfolds. Including multiple sites where students engage with academic discourse in various languages would allow for a more comprehensive picture of the linguistic resources and academic practices made available to students across the curriculum and the extent to which they are being socialized into seeing these as assets. When multiple languages are used in the classroom, attention to how they are functionally distributed (e.g., English for academic work, the local language for small talk) can also be revealing of the implicit socializing messages regarding what each language is best suited for (e.g., Beiler, Reference Beiler2020). Interviews with instructors and students would provide both a window into participants’ perspectives and attitudes towards multilingual practices, as well as accounts of past socialization experiences that may have contributed to shaping these perspectives and attitudes (e.g., Séror & Gentil, Reference Séror and Gentil2020). Collection and analysis of classroom interaction and/or student-produced academic texts in a variety of genres (oral and/or written) could provide evidence regarding the extent to which students had taken up ideologies or practices related to multicompetence (e.g., Beiler, Reference Beiler2020; Cavazos, Reference Cavazos2017). Students’ texts can also be the basis for text-based interviews in which students comment on the choices they made during the process of writing and extent to which attitudes towards multilingualism shaped these choices (e.g., Beiler, Reference Beiler2020; Kaufhold, Reference Kaufhold2018).