L2 grammar-for-interaction: Functions of “and”-prefaced turns in L2 students’ collaborative talk

This article examines how second language (L2) inter-actional competence is manifested in students’ use of “and”-prefaced turns when doing meaning-focused oral tasks in pairs and small groups. Drawing on video recordings from English-as-a-foreign-language upper-secondary classes recorded in Czechia and Finland, 86 sequences involving “and”-prefaced turns were scrutinized using multimodal conversation analysis, focusing on language, gaze, and material resources. The ﬁndings suggest that by producing “and”-prefaced turns, students orient to task progression. These turns have two functions: task managerial and contribution to the emerging task answer. By using task-managerial “and”-prefaced turns, the current speaker invites another student to participate, while in “and”-prefaced contributions to the task answer, a participant adds to, generalizes, or modiﬁes the previous task answer. The analysis shows that students mobilized their L2 inter-actional competence in producing “and”-prefaced turns in close coordination with embodied resources and with respect to the spatio-material surroundings and the nature of the task. These ﬁndings contribute to the multimodal reconceptualization of the grammar–body interface and research on turn-initial particles within L2 interactional competence.

Designing one's turn in interaction is a complex task for participants in terms of achieving intersubjectivity and coherence in interaction.In view of this, different kinds of turn-initial particles provide an important resource for participants to organize turn-taking and maintain the progressivity of ongoing talk.Interactants use words such as "and," "well," or "but" to provide an early indication of how their turn relates to the previous turn(s) and the kind of stance the emergent turn will take.A substantial part of research on turn-initial particles has investigated everyday conversations (e.g., Bolden, 2010Bolden, , 2018) ) and institutional interactions in first-language (L1) settings (Heritage & Sorjonen, 1994;Hutchby, 2020;Mondada, 2018;Pekarek Doehler, 2016), but comparatively fewer studies have explored how they are used in second-language (L2) interactions (for exceptions, see House, 2013;Polat, 2011), as part of L2 speakers' interactional competence (IC; also see García García, 2021; Y. Kim, 2009;Pekarek Doehler & Berger, 2018;Pekarek Doehler & Pochon-Berger, 2011).Yet, the pragmatics of turn-initial particles can be challenging for L2 speakers to master because of possible crosslinguistic differences and because these "little words" (Bolden, 2006) are not necessarily approached from an interactional perspective in pedagogical grammar handbooks or conventional textbooks, despite their frequency in social interaction.The present study aims to address this research gap and contribute to conversation analytic (CA) scholarship on turn beginnings by exploring what the use of the turn-initial "and" by L1 Czech and Finnish students in peer interaction in English-as-a-foreign-language (EFL) classrooms tells us about their interactional competence.Altogether, previous CA research on turn-initial particles has increased our understanding of the relationship between grammatical structures and interaction, and what "positionally sensitive grammar[s]" (Schegloff, 1996, p. 63) look like.Much prior research in this line of work is characterized by a focus on verbal dimensions of social interaction.However, along with the embodied turn in social interaction research (Nevile, 2015), there is growing awareness that turn-initial particles (along with other grammatical constructs) are resources that participants deploy, together with embodied and material resources, as they build action in situated ways.Such a multimodal reconceptualization is visible in a redirection of focus to what Pekarek Doehler et al. (2022) have recently termed as the "grammarbody interface," and in the increasing analytical attention to how, for example, turn-initial particles are deployed as part of multimodal action packages (e.g., Pekarek Doehler et al., 2021;Rönnqvist & Lindström, 2021).In the present study, we align with such a multimodal CA approach and analyze turn-initial "and" with a view on how embodied and material interactional resources configure and fine-tune the relatively broad meaning horizon of the particle in locally meaningful ways.Other resources that are particularly central in our institutional setting (L2 classroom) include various taskrelated texts such as textbooks, notepads, task sheets, and whiteboards, which participants recurrently orient to visually, read, and write on while interacting.
Our study addresses the following research questions: RQ1.What interactional functions does students' use of the turn-initial particle "and" have during collaborative speaking tasks in EFL classrooms?RQ2.What do students' ways of using turn-initial "and" suggest about their L2 interactional competence and L2 grammar-for-interaction?

TURN-INITIAL PARTICLES AND "AND"-PREFACED TURNS AS INTERACTIONAL RESOURCES
Linguistic objects that occupy the beginning of a turn have been variably called discourse markers (e.g., Bolden, 2006Bolden, , 2010;;García García, 2021; H.R.S. Kim, 2013;Y. Kim, 2009;Pekarek Doehler, 2016), response tokens (e.g., Golato, 2018;Hayashi, 2009;Hayashi & Kushida, 2013), and connectives (Heritage & Sorjonen, 1994;Rönnqvist & Lindström, 2021) in CA research.In line with Heritage and Sorjonen (2018), we prefer the term "turn-initial particle," since it refers to the use of a single uninflected element of language.Turn-initial particles are a critical and pivotal component of turn-taking, as they reveal the current speaker's understanding of what has been said and done, while also projecting the action of the current turn (Couper-Kuhlen & Selting, 2018;Kim & Kuroshima, 2013).Two main criteria have been key in defining turn-initial particles.First, they are linguistic objects that cannot be considered as grammatical constituents of the turns or turn-constructional units (TCUs) that they preface (Heritage, 2013;Kim & Kuroshima, 2013)-that is, they are not syntactically integrated within the TCU that follows.This may also explain why these elements can be omitted from subsequent repetitions of the turn if the turn in which they occur is identified as a trouble source in need of repair (Schegloff, 2004).Second, turn-initial particles are often "through-produced" (Couper-Kuhlen & Selting, 2018, p. 514), which means that there are no prosodic breaks such as pauses, lengthenings, or pitch breaks after the turn-initial particle.This is not always the case, however, and in L2 interaction such as our data, for example, speakers may sometimes produce turn-initial particles as separate prosodic units from the ensuing TCU and with distinctive intonation contours.
This study focuses on "and"-prefaced turns.According to commonly used English grammars (e.g., Biber et al., 1999;Quirk et al., 1985), "and" represents a coordinating conjunction that may connect phrases or clauses.However, these grammars typically do not distinguish between a turn-initial or other kind of positioning within a turn-at-talk in terms of functions of "and."A notable exception is the grammar by Biber et al. (1999), which acknowledges that "turn-initial coordinators [including "and"] are considerably more common in conversation than sentence-initial coordination in the written registers" (p.83).However, interactional functions of turn-initial particles are not discussed.In her study on discourse markers, Schiffrin (1987) has noted that "and" coordinates idea units and marks the current speaker's turn continuation.Unlike CA studies, these analyses do not systematically differentiate turn-initial "and" from its mid-turn and turn-final uses.
In contrast, CA research has investigated the different turn-constructional positions and uses of "and" and shown that it can preface different kinds of actions (e.g., questioning and informing).A typical function is that "and"-prefaced turns signal the continuation and progression of ongoing activity (Bolden, 2010;Heritage & Sorjonen, 1994, 2018).For instance, Heritage and Sorjonen (1994) have observed that in doctor-patient interaction-which is routinely organized as a series of questionanswer adjacency pairs-the doctor's "and"-prefacing tends to accompany questions that mark the beginning of "some larger course of action" (p.24).In contrast, doctors do not typically use "and"prefacing in their follow-up (or "contingent") questions that elaborate on the patient's previous answer.This suggests that turn-initial "and" offers resources for managing the progression of the current activity.However, what exact function the "and"-prefaced turn serves depends on the nature of the activity and its broader goals.
Turn-initial "and" can also occur in other kinds of sequential environments in everyday conversations.Bolden (2010) has shown that recipients can respond to extended informings with "and"-prefaced formulations performed as declarative assertions.Such formulations seek clarification or confirmation of understanding regarding something that the prior informing did not explicitly mention, and thus advance the informing activity and show affiliation with the speaker.However, participants may also use "and"-prefaced formulations pre-emptively to reshape the ongoing interaction in different ways, for example, by cutting straight to the point of an informing that is in progress.In that sense, Bolden's study also highlights that the turn-initial particle "and" may play an important role in negotiating whether the previous speaker has finished their turn.This suggests that turn-initial "and" can also offer a resource for participants to manage the organization of turn-taking in social interaction.

EMBODIED L2 INTERACTIONAL COMPETENCE AND GRAMMAR-FOR-INTERACTION
In the field of conversation analysis for second language acquisition (CA-SLA), the term "interactional competence" (IC) is understood as part of interactants' collaborative work to achieve intersubjectivity and make progress in social interaction-that is, it involves participants' "ability for joint action that is contingent on the details of the social interaction [they] participate in" (Pekarek Doehler, 2019, p. 30;emphasis in original).It is also situation-based, context-sensitive, and recipient-designed in ways that it is "'publicly' observable within participants' practices" and "brought about by the social interaction at hand" (Pekarek Doehler, 2019, p. 30).When investigating IC, the focus is on participants' ways of constructing social actions collaboratively in the here and now, and increasingly also on how they do this by drawing on different multimodal resources, not only language (see, e.g., Pekarek Doehler & Eskildsen, 2022).When it comes to L2 IC, learners are viewed as interactionally competent participants with existing interactional competencies that they need to recalibrate and refine, although the relationship between IC and the development of one's linguistic skills is not straightforward (Pekarek Doehler, 2019, p. 46).
Overall, research on L2 IC is diverse and embraces many settings and languages as well as both cross-sectional and longitudinal perspectives (e.g., Hall et al., 2011;Hellermann et al., 2019;Pekarek Doehler & Berger, 2018;Salaberry & Kunitz, 2019).A handful of earlier CA-SLA studies have explored L2 learners' developing IC by investigating the functions and emergence of particular grammatical resources in L2 interaction.There are, however, relatively few studies focusing on L2 learners' use of turn-initial particles within CA-SLA (although see Y. Kim, 2009, for the Korean "Kuntey [but]" and the studies referenced next).To our knowledge, García García's (2021) study on the functions of the Spanish turn-initial "y," an equivalent of the English "and," is the one that comes closest to our study.The author investigated its use during L2 conversation tasks.The findings show that "y" is used in topic proffers, where it links the currently exhausted topic and a new topic.Relatedly, speakers use turn-initial "y" to preface questions in what the author terms "rotation series"-that is, when multiple speakers, one by one, respond to the same question-a phenomenon present in our data, as the analysis will show.
In addition, for our purposes, two other studies of turn-initial particles in L2 interaction are relevant.In their cross-sectional study, Pekarek Doehler and Pochon-Berger (2011) compared intermediatelevel learners' methods for doing disagreements with those of advanced learners during French lessons.Their findings show that while both learner groups used polarity markers and the turn-initial "mais [but]" to display disagreement, the advanced learners were able to use more varied methods that resemble those of L1 speakers, such as hedges, the "yes-but" turn-format, and format-tying.In a longitudinal study, Pekarek Doehler and Berger (2018) showed that an L2 French speaker's competence to open storytellings developed so that she was able to use the turn-initial "mais [but]" to accomplish prefatory work that helped her display her story's relevance to prior talk and to project its tell-worthiness.By analyzing the learners' use of "mais," both studies demonstrated that as L2 learners' IC develops, they can use their linguistic skills to turn-and recipient-design their talk in more context-sensitive and diverse ways to fit the local contingencies of the interaction.More pertinently, learners are able to use linguistic constructions, such as "mais," to accomplish specific social actions, such as doing disagreement or opening a story, thus manifesting an emergence of L2 grammar-forinteraction, that is, "a grammar that serves as an instrumental tool for conduction and coordinating L2 talk-in-interaction" (Pekarek Doehler, 2018, p. 4).
Along with the embodied turn, studies on grammar have also started to investigate how different multimodal resources are used to design TCUs and interactional turns, and social actions more broadly (e.g., Pekarek Doehler & Eskildsen, 2022;Pekarek Doehler et al., 2022).In terms of turn beginnings, Deppermann (2013) has argued that participants have four tasks to consider when designing turns that are fitted to the sequential and temporal contingencies of interaction: "to achieve joint attention to the upcoming turn, to display uptake of prior turn(s), to deal with projections emanating from them, and to project properties of the upcoming turn" (p.91).Moreover, all four tasks are situated in specific activities in their spatio-material ecologies-that is, interactants design their TCUs by drawing on the various multimodal resources at hand in the situation and coordinating their use in context-and activity-specific ways, be it in L1 or L2 interactions.For instance, Rönnqvist and Lindström (2021) demonstrated how both L1 and L2 speakers of Swedish use the Swedish turn-initial particle "å sen [and then]," recurrently together with a pointing gesture at a relevant object to mark topic continuation."Å sen" is then especially used to specify, restrict, or redirect the topic in different ways, yet 15404781, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/modl.12885 by University Of Jyväskylä Library, Wiley Online Library on [05/11/2023].See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions)on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License its meaning for the participants is closely tied to the pointing gesture.The authors explain that the turn-initial particle in these uses does not itself elaborate on the previous talk but, thanks to its additive meaning, it makes the speaker change and the shift more subtle; it also makes it possible for the current speaker to produce contributions that are potentially nonaligning with the previous speakers (Rönnqvist & Lindström, 2021, p. 9).More relevantly, the "å sen" precedes the pointing gesture, thus serving mainly the linking function, while the embodied resource is employed as an interactional cue, for example, to draw participants' attention to the next item to be talked about and to signal turn completion.Important for our focus is the understanding that in object-based interaction, such as speaking about various pictures as in Rönnqvist and Lindström's (2021) study, participants orient predominantly to the task-relevant objects during the task activity, while directing their gaze to other participants at interactionally meaningful moments (e.g., Tuncer et al., 2019;Vänttinen, 2022).However, and as we will show in this study, the grammar-body interface is not limited to the kinds of recurrent and grammaticalized multimodal action packages exemplified by Rönnqvist and Lindström's (2021) study but instead encompasses a broader range of ad hoc combinations of various verbal, embodied, and material interactional resources, which are not necessarily routinized by the participants nor grammaticalized in any language community (see also Stukenbrock, 2021).

DATA AND METHOD
Our study is based on video recordings of pair and group work speaking tasks from Czech and Finnish upper-secondary school EFL lessons.We combined these two datasets to investigate what interactional practices students with different L1 backgrounds use when interacting in English.The Czech data comprises approximately 7 hours of video-recorded group or pair task interactions from 18 lessons in five different schools in programs leading to the upper-secondary school leaving (Maturita) examination: three grammar schools and two other schools providing education in specific fields (economics and fine arts).According to the Czech framework curricula, the target levels in English in uppersecondary education range from intermediate (B1) to upper-intermediate (B2).The students in our dataset were in their final year (i.e., 18 to 19 years old), and their proficiency levels varied from intermediate to advanced, as did the teaching materials and coursebooks that the students and teachers were using.In the two schools providing education in specialized fields, the activities were to some extent shaped by the requirements and format of the standardized Maturita examination, while in the three grammar schools, the lessons were more flexible and included a number of speaking activities whereby the students were supposed to express their views.Before data collection, the students signed informed consent.
The Finnish data consists of 1 hour and 31 minutes of video-and audio-recorded speaking tasks that originate from five lessons in one school.The lessons comprise various activities, such as teaching grammar, doing and checking exercises, and engaging with book chapters, which means that in each lesson, there is only room for short speaking tasks, hence the small amount of data compared to the Czech data.The course in which the students are enrolled is an optional course meant for secondand third-year students, who are 17 or 18 years old.While the target proficiency level in the uppersecondary school for English in Finland is B2.1 (Finnish National Agency for Education, 2019), based on our data, the students' proficiency level varies from low-intermediate to advanced learners.The students' parents or legal guardians signed informed consent for their participation in the study before data collection.
In order to secure comparability between the datasets, we focused solely on student-student interactions in meaning-focused oral tasks, which required students to express and share their views and ideas, invent or retell a story, reach an agreement, discuss and evaluate figures, and do roleplays.Many of these tasks included several subtasks, often materialized in the form of lists of questions.The tasks were semistructured in that the teachers or textbooks had provided some instructions and guiding questions, yet the way the students accomplished the tasks was negotiated moment-to-moment as the interaction unfolded (see Park, 2021).For the analysis, we went through the videos and the transcripts, which were produced following CA conventions for transcription (Jefferson, 2004), and created a collection of sequences in which "and" featured in turn-initial position, thus narrowing down the locus of analysis to a specific interactional phenomenon, that is, students' "and"-prefaced turns during the speaking tasks (for crosslinguistic comparability in interactional grammar research, see, e.g., Pekarek Doehler et al., 2021).In the end, our collection comprises a total of 86 sequences, 69 from the Czech dataset and 17 from the Finnish dataset.The collection excludes cases of self-continuation realized as a new TCU following a gap after a completed TCU by the same speaker.Excluded from the collection are also "and"-prefaced turns that were cut off shortly after the particle, since their functions were not identifiable (see line 8 in Excerpt 4 for an example).
The sequences in the collection were then analyzed using principles of multimodal conversation analysis (Goodwin, 2018) to scrutinize how participants produced the interaction action by action.We paid special attention to the sequential position of the "and"-prefaced turns and the role of verbal, embodied, and material resources used to construct and make sense of the TCUs prefaced by "and"-in particular, the participants' gaze directions, gestures, and orientation to the teaching materials.We also observed the prosodic details of the turn-initial particles to the extent possible in the noisy, multiparty classroom setting.The excerpts presented in this article have been multimodally transcribed, applying Mondada's ( 2019) conventions (see Appendix).The students' names are pseudonyms, and to further protect the participants' identity, we use simplified line drawings to illustrate the spatio-material setting of key moments.

TWO FUNCTIONS OF "AND"-PREFACED TURNS IN L2 PEER INTERACTION
The analysis demonstrates the students' IC to use "and"-prefaced turns as a resource (a) for taskmanagerial purposes, that is, to organize their participation during pair or group work, and (b) for contributing to the emerging task answer, that is, to propose additions, elaborations, or summaries to emerging task answers.In both cases, students orient themselves to ensuring the progress of the ongoing speaking task.In what follows, we illustrate the two functions of "and"-prefaced turns by analyzing selected excerpts from both datasets, paying close attention to how the group constellation and the task goal shape the multimodal production of the focal turns.

"And"-prefaced turns as a resource for task-managerial purposes
The speaking tasks in our data have a somewhat predetermined agenda, and one typical use for turninitial "and" in our dataset includes turns that are explicitly designed to achieve task-managerial purposes-that is, they serve to organize students' participation in view of turn-allocation and speaker change.In such instances, the turn-initial "and" always prefaces a first action, typically a question targeted at another participant (see also García García, 2021; Heritage & Sorjonen, 1994).Excerpt 1 demonstrates the use of an "and"-prefaced turn in giving the floor to the last student to speak during an oral group task, while Excerpt 2 illustrates its use in a paired task, in which students make written notes of their answers.
Excerpt 1, from the Finnish dataset, comes from a small-group speaking task organized as a round robin (Mortensen & Hazel, 2011) in the sense that each student is expected to tell the group which vacation destination they would like to travel to and why.The destinations are listed in the students' textbooks, which they have in front of them on desks or in their hands and which provide the students with information and relevant vocabulary for the task.Our focal group has four members, who are placed so that Kerttu and Paula have turned in their seats to face Kirsi and Tuomas (see Figure 1), that is, they have visual access to one another.Prior to the excerpt, Kirsi and Kerttu have already shared their travel destinations, but Paula has encountered difficulties in producing one more reason why she would want to travel to the Maldives, her selected destination.As the excerpt begins, Kerttu offers in Finnish a potential reason to Paula (which Kerttu herself mentioned earlier).After Paula accepts this  Paula first accepts Kerttu's suggestion in Finnish ("joo.")and then verbalizes it in English (3), thereby producing her task-relevant turn.Kerttu acknowledges Paula's response (4), and Kerttu and Kirsi establish a brief mutual gaze and laugh in unison (4, 5), possibly orienting to a shared joke between the two of them.Paula only briefly gazes at Kerttu and Kirsi, after which she shifts her gaze to her book (5), thus marking the completion of her sharing turn.Orienting to this, Kerttu allocates the next sharing turn to Tuomas with an "and"-prefaced question (6).The "and" is preceded by a vocalization ("uhh") in prebeginning position and accompanied by Kerttu's gaze shift to Tuomas that signals the turn's recipient (Lerner, 2003).The "uhh" also serves to draw Paula's gaze to Tuomas (6).The turn-initial particle is followed by an address term that verbally disambiguates Tuomas as the recipient, and thus calls for his attention.It is possible that the address term orients to the fact that Tuomas has not visually attended to the group activity during the preceding turns (Lerner, 2003); instead he has been looking away from the others and thus has not seen Kerttu's gaze shift to him.Despite this, Tuomas responds to the question (8), shifting his gaze from his notebook to Kerttu toward the end of his turn.Kerttu acknowledges the response, after which she asks for Tuomas's reasons for traveling to Galapagos Islands (10−11).
Several noteworthy points can be made from the excerpt.First, Kerttu's "and"-prefaced turn (6) clearly orients to the social organization and progress of the task by requesting the only contribution that is missing from the round robin sharing activity.In this sense, the "and"-prefaced turn is both backward and forward looking (Heritage & Sorjonen, 2018), since it marks the return to the sharing round (also Heritage & Sorjonen, 1994) after an extended turn by the previous speaker and reinvokes the goal-oriented character of the task, including the use of the target language (English) after temporary L1 use (see also García García, 2021).Second, the other participants' gaze shifts at line 6 reconfigure the embodied participation framework so that it now includes Tuomas, who has so far been excluded not only by himself but also by the others' orientation to one another and their own books (see also Evnitskaya & Berger, 2017).Namely, while Kirsi shifts her gaze to Kerttu, the current speaker, both Kerttu and Paula shift their gaze to Tuomas at the beginning of Kerttu's "and"-prefaced turn, thus orienting to his imminent speakership (see Figure 2).The change in the participation framework prepares the space for Tuomas to contribute to the task, which eventually ensures task completion.Finally, Kerttu's "and"-prefaced turn to allocate the last turn to Tuomas can also be seen as a display of deontic authority (Stevanovic & Peräkylä, 2012) on her part to steer the task forward, as if chairing a meeting.Overall, in Excerpt 1, the group constellation (four participants) and the task goal (task answers are produced only orally) influence the task progress in that the participants need to manage the round robin explicitly by nominating the next speaker and adjusting the embodied participation framework in relevant ways.
While in Excerpt 1 the "and"-prefaced turn signals the last speaker of the round robin in an orally accomplished group task, the management of speaker change can look different in paired speaking tasks where students are also expected to write their answers, as Excerpt 2 demonstrates.It comes from a paired task in which Czech students were instructed to ask each other about their favorite area of culture and then ask and answer follow-up questions.They were also encouraged by the teacher to take notes to report on their peer's answers to the class later.Our focal pair comprises Simona and Alena, who are seated next to each other and take individual notes of the other's answers.Prior to the excerpt, Simona has expressed her interest in films and Alena in music, and here we see how the participants use "and"-prefacing on two occasions (3, 8) to mark their turns as follow-up questions that elaborate on the areas of culture that they have selected.Both "and"-prefaced turns serve to forward the task interaction, but in ways designed to fit the local sequential and temporal contingencies of asking questions and writing notes.In both cases, we see that the embodied participation framework of the activity is shaped by the participants' orientation to the 15404781, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/modl.12885 by University Of Jyväskylä Library, Wiley Online Library on [05/11/2023].See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions)on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License parallel activity of note-taking (see also Svinhufvud, 2016) during the other's answer (i.e., Alena writing in 1−5 and Simona in 6−8).While the other is writing, long silences emerge (2, 7), during which the one waiting to ask their question visually monitors the writing to be able to produce the follow-up question at a point in which the note-taker is available.For example, Simona monitors Alena's writing during the long silence (2), makes her own notes, and waits before launching the follow-up question.Her "and"-prefaced question (3) is accompanied first by a gaze shift to Alena and later toward the end of the turn by her bodily lean toward Alena (Figure 3).The lean helps position Simona closer to Alena's peripheral vision, thus attracting Alena's attention and marking the imminence of the speaker change.Indeed, after a 0.4 second silence, Alena stops writing for the duration of her response (5), which Simona acknowledges and takes a note of (6).
Alena's "and"-prefaced follow-up question in line 8 is similarly addressed to a recipient who is engaged in note-taking and with whom she, therefore, does not have mutual gaze.In fact, Alena herself is still writing her notes as she begins the turn with an elongated "and" that is followed by a long silence (8).As Alena is still gazing down and writing, it is possible that the elongation, together with the silence, serves as a harbinger that she is preparing her next question while talking and writing.The silence is then followed by a self-repair, with the turn-initial "and" repeated with a slight elongation, while still gazing at her notes (Figure 4).She shifts her gaze toward Simona at the end of the turn (8), thus orienting to the speaker change.However, because Alena has been preoccupied with writing her own notes, it may be that she has not noticed that Simona is also doing the same, and thus is not available for immediate recipiency.The problem of alignment becomes visible when it takes 0.6 seconds for Simona to react to Alena's question.At this point (10), Simona stops erasing her text, establishes mutual gaze with Alena, and initiates repair by asking Alena to repeat her question, which Alena does in line 11.The repaired question ( 11) is produced without a turn-initial "and," which suggests that, in contrast to lines 3 and 8, there is no need to flag the question as a new item in a series.This supports earlier observations that turn-initial particles tend not to be repeated when repaired, as they are not syntactically integrated in the TCU (Schegloff, 2004).
In this section, we have shown how the students' IC manifests in their ability to use turn-initial "and" to preface "first" actions that contribute to task progression.A key difference between the analyzed excerpts is that in Excerpt 1, students are free to organize their turns-at-talk as they see fit within the round robin, and the "and"-prefaced turn both manages to accomplish speaker change and ensures that the round robin organization is sustained until the end.In Excerpt 2, the students are specifically instructed to ask each other questions by taking turns.In both cases, however, the use of the turn-initial "and" highlights the serial nature of the ongoing task, be that by requesting a "missing" contribution (Excerpt 1), by flagging a turn as a follow-up question (Excerpt 2), or by reading out loud the next item of the task (see Excerpt 3, line 9, in the next section).Interestingly, in Excerpt 2, the followup questions related to the students' areas of interest are prefaced with "and"-unlike in Heritage and Sorjonen's (1994) study, where the follow-up questions seeking elaboration or clarification were produced without the particle.The use of the "and" thus seems to laminate the actions with an explicit sense of "moving on" (see also Rönnqvist & Lindström, 2021).Moreover, "and"-prefacing is one resource among others (e.g., gaze and the material artifacts) that contributes to making these turns intersubjectively understandable as particular actions.In this respect, they are multimodally designed to fit the local sequential and temporal contingencies of the task interaction and the demands of the broader task-related goals.

"And"-prefaced turns as a resource for contributing to the emerging task answer
Another use for turn-initial "and" in our dataset is as a resource for contributing to the emerging task answer.In such instances, the "and"-prefaced turns add to, elaborate, summarize, generalize, or modify the answers given so far, whereby they continue the current topic in different ways (see also Bolden, 2010; García García, 2021; Rönnqvist & Lindström, 2021).In essence, the "and"-prefacing functions as a tying device that signals the import of the turn to the emerging collaborative task answer 15404781, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/modl.12885 by University Of Jyväskylä Library, Wiley Online Library on [05/11/2023].See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions)on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License and as such is both backward and forward looking (Deppermann, 2013;Heritage & Sorjonen, 2018).An orientation to the agenda-based nature of the task and task progress is thus visible in these uses of "and" as well (e.g., García García, 2021; Heritage & Sorjonen, 1994).Excerpt 3 demonstrates an instance in which an "and"-prefaced turn adds to the previous answer (turn), while Excerpts 4 and 5 illustrate how "and"-prefaced turns can modify the suggestion made in the previous turn in ways that imply disagreement, highlighting that turn-initial "and" can be used for sensitive interactional work.
In Excerpt 3, two students, Eva and Roman, are responding to the question "Who is the target audience?"after having watched an advertisement issued by a council for accident prevention.In the video, a young person gets hit by a car as he crosses a street while listening to music on a headset and texting at the same time.Prior to the transcript, Eva has said that the advertisement is probably not intended for elderly people, who, according to her, need to concentrate on writing and are unable to walk and write at the same time.Then she summarizes her position in line 1, where "they" refers to elderly people.In line 6, Roman adds to Eva's answer, using an "and"-prefaced turn.In lines 1−2, and 4, Eva characterizes elderly people as those who stop when they want to write something on their phones, and then continue walking.Roman displays recipiency by repeating a part of Eva's turn with falling intonation (3) and nodding (5).Eva starts gazing back at Roman at the end of line 5, by which point they have established an embodied participation framework favoring collaboration (see Figure 5).It is in the subsequent line that Roman produces the "and"-prefaced turn ("and they wait for the car," 6), the object being accompanied by a tapping gesture on the desk that emphasizes its relevance to the point Roman is making.In his turn, Roman not only adds to Eva's answer, but also explicates the point Eva has been making: Elderly people would not only stop and write their text but also pay attention to the traffic and wait for a car to pass before crossing the street.Hence, in his contribution, Roman articulates what Eva did not say (see also Bolden, 2010) and underlines the fact that the advertisement is not meant for elderly people.Eva agrees in the subsequent line, which ends the exchange regarding the target audience of the advertisement.Roman then moves the task forward with another "and"-prefaced turn ( 9), which performs a task-managerial action similar to the instances described in the previous section.
Excerpt 4 comes from the Finnish dataset and occurred during a small group task, during which the students discussed the (dis)advantages of different energy sources.The discussion was based on a text on environmental issues and accompanying pie charts displaying the breakdown of energy sources in Finland and the United States.The focal group consists of four students (Elisa, Tuuli, Nina, and Mari), who are huddled together as shown in Figure 6.Prior to the excerpt, they have discussed the use of fossil fuels and nuclear power, and at the beginning of the excerpt, Elisa directs the group's attention to the (dis)advantages of renewable energy sources (1−2).The "and"-prefaced turn in focus is produced in lines 7 and 9 as a generalization that pinpoints the main advantage of renewable energy sources.In overlap with Elisa's initiation, Tuuli starts providing an answer by focusing on the advantages of solar power (3−5).Her turn is multimodally constructed as the verbal explanation is accompanied by a thumb-raising gesture (3; see also Figure 6) that indicates a list to come.However, she encounters problems in producing the turn and initiates an embodied word search by way of gesturing a rectangle in front of her (5).While she is engaged in the search, Mari produces an "and"-prefaced turn (7, 9) that shares similarities with what Bolden (2010) has described as pre-emptive formulation: Not only is it uttered before Tuuli's turn is complete but it also pre-emptively summarizes the main advantage of the three power sources (i.e., "they" and "all;" 7).Unlike in Bolden's data, here the "and"-prefaced turn is its own TCU that does not serve to complete Tuuli's turn; instead, due to its sequential placement and broader task-relevant scope ("all" vs. "solar power"), it makes the completion of Tuuli's interrupted explanation less crucial from the perspective of task completion.Tuuli visibly orients to this as she disengages from her embodied search and shifts gaze to Mari (7, see Figure 7).Mari's turn (7, 9) can thus be seen as nonaligning in terms of how it curtails the completion of Tuuli's turn.Yet, simultaneously, it ensures task progression as it leads the students to discuss the disadvantages of the energy sources (data not shown) after Nina's positive acknowledgment of Mari's answer (10).
The way the group accomplishes the task is characterized by self-selection, which is visible in the competition for turns and resulting overlaps.Due to this, there is an example of another "and"-prefaced turn by Elisa (8) that she cuts off.Mari's "and"-prefaced turn, in contrast, is produced at a moment when Tuuli is in the midst of the word search.This makes it possible for Mari to self-select and launch her generalization, which both Tuuli and Elisa visibly orient to as they turn their gazes toward Mari; both also cut off their turns (5 and 8, respectively).Interestingly, Nina, the only nongazing participant, acknowledges Mari's response (10).More relevantly, the embodied participation framework of the situation shows that students treat the EFL textbooks as resources that mediate their task talk, visible in the way they visually attend to and scrutinize the book and its semiotic resources.In other words, students' gaze orientation is predominantly to the books (see also Tůma, 2022;Vänttinen, 2022).In addition, the students treat the task question as an agenda that sets a particular order of topics for their talk, which is revealed in the way "and"-prefaced turns are used to make progress with the activity.
Excerpt 5 shows that "and"-prefaced turns can be used to modify the task answer.The excerpt comes from the Czech dataset and features three students whose task is to create identities for two men displayed in a picture on the projection screen and to invent their story.The group has come up with the idea that they are a homosexual couple and engage in brainstorming details about their lives.Relevant to our analysis is the way Jana negotiates the length of the two men's relationship through an "and"-prefaced turn (5) that modifies Klára's earlier proposition (2).
EXCERPT 5 Klára proposes that the men "have been together for three and half years" (2), which she produces while looking at both her peers, who subsequently acknowledge and verbally accept the proposal as part of the emerging narrative about the two men (3−4).Klára then turns around to look at the projection screen to see the picture of the two men (see Figure 8), by which she temporarily withdraws from a participation framework that favors collaboration.Marcela, who is responsible for taking notes, also disengages and starts writing (2), which further works as a public indication of the acceptance of Klára's proposal.
Then, Jana produces a turn with a turn-initial "and" (5), followed by the possessive pronoun "their," thus tying her emerging turn to Klára's turn (2).These two components of Jana's turn seem to recruit Klára's attention, as she turns back at her peers and gazes at Jana, whose turn can be heard as an other-continuation of and addition to Klára's turn, which is further evident from Jana's gaze shift toward Klára (Figure 9).However, Jana's turn contradicts Klára's turn content-wise, as Jana suggests that the anniversary of the relationship is coming soon-typically, an anniversary does not happen in the middle of a year.Therefore, Jana's suggestion in fact modifies Klára's original proposal.A similar principle-employing a turn-initial particle with an additive meaning as other-continuation, while in fact introducing a new line of reasoning-is reported by Rönnqvist and Lindström (2021).
15404781, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/modl.12885 by University Of Jyväskylä Library, Wiley Online Library on [05/11/2023].See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions)on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Interestingly, Jana's "and"-prefaced turn ( 5) is immediately accepted by Klára (6), whose smiley voice may index, on the one hand, acceptance and amusement, but on the other hand, it may be indicative of the contradiction, which is, however, not problematized overtly.
The "and"-prefaced turn in Excerpt 5 can thus be seen as a borderline case.On the surface, it is produced as an addition to the previous proposal, similar to line 6 in Excerpt 3, while content-wise it in fact modifies it.It can be argued that the way it is produced (i.e., a turn-initial "and" followed by a pronoun with the same referent as in the utterance which is being modified), and the task nature (i.e., brainstorming ideas), have led to an acceptance of the proposal.Thus, turn-initial "and" can also serve as a resource for what Nielsen (2012) has described as a "decoupling" of the activities of generating and evaluating ideas in brainstorming activities, enabling the students to develop their ideas about the storyline "without any immediate concern for their adequacy" (p.105), as can be seen from the students' immediate and amused reactions (3, 4; also 6).The issue at hand is how participants in these types of activities can modify and revise each other's proposals when producing a shared story.Our analysis suggests that "and"-prefacing and tying to the previous ideas by using anaphoric pronouns (in Rönnqvist & Lindström, 2021, the tying was achieved by pointing gestures instead) is a resource that allows Jana to modify Klára's proposal in a very implicit manner.
In this section, we have analyzed how students use "and"-prefaced turns to contribute to the emerging task answer by adding to it (Excerpt 3), generalizing it (Excerpt 4), and modifying it (Excerpt 5).Since these actions enable the students to make progress in the task, such "and"-prefaced turns represent a central resource in the students' IC.Here, gaze helps participants disambiguate whose turn is being continued and what stance the participants adopt.Of special interest are turns that generalize or modify the preceding contribution, as they may be seen as sensitive actions because they can potentially be heard to imply criticism of the preceding contribution and constitute a face threat.In Excerpt 4, the student who produces the "and"-prefaced turn (i.e., Mari; 7, 9) keeps gazing at her textbook, while two other students shift their gaze toward that person abstaining from acknowledging the contribution, thus showing their stance toward the generalization.Similarly, in Excerpt 5, the modifying "and"-prefaced turn ( 5) is responded to by Klára immediately.However, she produces it in a smiley voice and somewhat ironically, thus potentially mitigating the new proposition or displaying her stance.

CONCLUSIONS AND IMPLICATIONS FOR L2 EDUCATION
This study has examined L2 students' use of "and"-prefaced turns during speaking tasks in EFL lessons and what their use can tell us about the students' L2 interactional competence.In line with recent calls for reconceptualizing grammar as an interactional and multimodal phenomenon (e.g., Pekarek Doehler et al., 2022), we have analyzed "and"-prefaced turns as a particular grammar-body interface by focusing on how embodied and material resources configure and fine-tune the relatively broad meaning horizon of the particle in locally meaningful ways.
Altogether, we have shown how the students employ "and"-prefaced turns for two interactional functions: for task-managerial purposes and for contributing to emerging task answers.This highlights the students' IC, as they can utilize one linguistic resource to perform two interactional functions that display their orientation to task progression (see also, e.g., Bolden, 2010;García García, 2021;Heritage & Sorjonen, 1994, 2018) but in different ways.Task-managerial "and"-prefaced turns allocate the turn-at-talk to another participant.Such turns are accompanied by gaze shifts toward the next speaker, whereby the shift serves as an embodied means of next speaker selection (also Lerner, 2003).The gaze shift helps disambiguate the addressee, together with a potential address term (Excerpt 1) or referential term (Excerpt 2), and show imminent recipiency on the part of the current speaker.The onset of the gaze shift in relation to the emerging turn can vary depending on whether the speaking task involves the coordination of talk with learning materials such as personal notes (Excerpt 2) or the board (Excerpt 3).In such cases, the gaze shifts typically occur toward the end of the turn.The 15404781, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/modl.12885 by University Of Jyväskylä Library, Wiley Online Library on [05/11/2023].See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions)on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License particle "and," together with a gaze shift, thus seems to laminate the action with an explicit sense of "moving on" (also Rönnqvist & Lindström, 2021), thereby ensuring task progression as a result of turn-allocation and speaker change.
Similarly, students' "and"-prefaced contributions to the emerging task answers show an orientation to making progress with the ongoing task-that is, they build on the turn(s) produced by the previous speaker(s) by adding to them (Excerpt 3), making a generalization based on them (Excerpt 4), or modifying the previous answer (Excerpt 5).Using some of these "and"-prefaced turns, speakers may express disagreement with the prior speaker's contributions.In such turns, for the most part, the students' gaze orientation is toward the materials during tasks that involve books or instruction sheets, and when gaze shifts toward co-participants occur, they are produced for specific purposes (see also Tůma, 2022;Vänttinen, 2022), such as allocating the turn to the next speaker.
Overall, our study offers new insights into research on the use of turn-initial particles in the fields of interactional linguistics, L2 interactional competence, and L2 grammar-for-interaction. First, previous studies on the functions of "and" in L1 settings have not addressed its use as a resource for task-managerial purposes.Second, as was mentioned earlier, the use of turn-initial particles from the perspective of L2 learning has been relatively underresearched, especially research focusing on "and" (cf.García García, 2021).Finally, our findings are based on naturally occurring classroom data from two countries (Czechia and Finland) as opposed to those of García García (2021) and Rönnqvist and Lindström (2021), who used researcher-elicited data.
Previous CA studies have investigated the grammar-body interface by analyzing multimodal action packages that are recurring combinations of vocal and bodily conduct (e.g., Pekarek Doehler et al., 2021;Rönnqvist & Lindström, 2021).Our analysis adds to these studies by showing that in materially rich settings such as language classrooms, participants together assemble task-relevant materials, embodied resources, and "and"-prefaced turns to produce social action.Such "and"-prefaced turns are fitted to the temporal, material, and sequential contingencies of the unfolding task interaction in ways akin to what Stukenbrock (2021) has called ephemeral multimodal gestalts.In contrast to socially sedimented and grammaticalized gestalts, ephemeral gestalts consist of an interplay among verbal, embodied, and material resources that is only locally routinized and employed in the here and now for these purposes.For example, the findings show that students use gaze in various ways to achieve turn allocation or display orientation to task-related materials.This underscores the role of the spatio-material ecology of the settings, including spatial arrangements (such as the position of the task-relevant materials), embodied participation frameworks, and the varying local contingencies, such as the nature of the task, the requirement to take notes, and the distribution of the task of note-taking among the participants (see also Deppermann, 2013).
In sum, both functions of the "and"-prefaced turns manifest the students' L2 interactional competence in that they were able to tie their turns together, display sensitivity to the sequential placement of their "and"-prefaced turns, and monitor and adapt their actions collaboratively with respect to their peers (cf. Pekarek Doehler & Berger, 2018;Pekarek Doehler & Pochon-Berger, 2011) while simultaneously reconfiguring the embodied participation framework accordingly.It thus follows that turn-initial "and" plays an important role in how the students construct the L2 speaking tasks as collaborative, since they are able to design their turns in such a way that their co-participants orient to, understand, and accept them for the action they are designed for (e.g., Pekarek Doehler, 2019;Pekarek Doehler & Berger, 2018).That the students-in two different countries with diverging pedagogical environments-are able to utilize the turn-initial particle in ways similar to those described in studies of "and"-prefacing in L1 interaction (Bolden, 2010;Heritage & Sorjonen, 1994, 2018) to coordinate and organize their L2 interaction demonstrates that it has become a routinized linguistic resource for them (see also Pekarek Doehler, 2018).As such, it can therefore be argued that the ease with which the students deploy turn-initial "and" in varied contexts and with nuance showcases their developed L2 grammar-for-interaction.
One limitation of our study is its cross-sectional design and dataset.This does not enable us to pinpoint when exactly the interactional functions described in the study typically emerge in the 15404781, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/modl.12885 by University Of Jyväskylä Library, Wiley Online Library on [05/11/2023].See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions)on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License developmental trajectory of L2 users of English, and the possible variation in the emergence within the two groups of participants we have studied.However, we have observed that, especially in the Finnish dataset, the lower intermediate level students' interactions did not include any "and"-prefaced turns, unlike in their more advanced peers (see Excerpts 1 and 4).Given that our students' English proficiency ranges from lower intermediate to advanced levels, it could be that the use of turn-initial "and" emerges later than the use of "and" as a coordinating conjunction, around (upper-)intermediate proficiency levels.An interactionally competent use of turn-initial "and" requires a participant to monitor what has happened just before and fit their turn in the ongoing interaction, which is a complex task to accomplish in an L2.Further longitudinal studies could shed light on this issue in the future, as well as comparative studies of turn-initial particles across languages (e.g., Heritage & Sorjonen, 2018).We are not aware of microanalytical interactional studies that would have explored the equivalent of turninitial "and" in Czech and Finnish languages ("a" and "ja," respectively); even if Czech and Finnish learners of English could transfer resources from their L1s to L2 English, it does not mean transfer in all L1-L2 combinations would work in equally transparent ways.
Our findings also have implications for how commonplace interactional phenomena, such as turninitial particles, should be taken into consideration in L2 education.Recently, calls have been made to bridge the gap between research and practice in terms of L2 IC (e.g., Salaberry & Kunitz, 2019).Such calls emphasize the fact that teachers and teacher educators need to become more aware of the interactional and social character of language (learning) in order to reconsider grammar teaching and include interactional practices in the focus as well (e.g., Pekarek Doehler, 2018, 2019).This is relevant considering that the two uses of the "and"-prefaced turns reported here differ from normative and pedagogical grammars (e.g., Biber et al., 1999;Quirk et al., 1985), where "and" is discussed as a coordinating conjunction mainly in written communication and rarely as a particle that can be used to initiate an action.In this respect, our findings can help respecify pedagogical grammars by extending the environments in which and the functions of how "and" is used.On a more general level, an actionoriented view of language can also challenge a conception of language competence as an individual property (see also Huth, 2021).Furthermore, our findings show that speaking tasks that are already used in L2 lessons provide a natural context for practicing different aspects of grammar, including such little words (Bolden, 2006) as turn-initial particles.A key step is to ensure that (preservice) teachers become aware of peer interaction as a site for complex language use that requires students to be able to, among other things, manage participation on their own and add to their peers' contributions.These actions are highly relevant for smooth task accomplishment and should be considered by language teachers and textbook writers, who may want to include examples of such language use, possibly based on naturally occurring data, in the task instructions and objectives of collaborative speaking tasks in order to raise learner awareness of aspects of spoken language use that might otherwise go unnoticed.In this study, we have focused on the interactional functions of "and"-prefaced turns and shown students' interactional competence in how they employ it.However, more research in the use of turn-initial particles is needed to help teachers implement a broader range of functions and particles into collaborative L2 speaking tasks.

A C K N O W L E D G M E N T S
The collection of the Czech data was supported by the Czech Science Foundation (grant number 18-02363S).This article project was supported by the University of Jyväskylä Visiting Fellow Programme (2021) and the Academy of Finland (grant number 343480).We would like to thank Arja Piirainen-Marsh for her feedback on a previous version of the manuscript, the anonymous reviewers for their helpful comments, and Neea-Stiina Urpilainen for collecting the Finnish EFL data.