"Skogmyr Marian's research offers us a master class in social observation. Through careful sequential analysis, she demonstrates how L2 learners' interactional repertoires are developed and diversified, both in the moment and over time."

## **Tim Greer**, *Kobe University, Japan*

"Framed within the construct of interactional competence, this book elegantly demonstrates how to do longitudinal, second language learningrelated research from the perspective of multimodal, ethnomethodological conversation analysis. Specifcally, it is particularly notable for its brilliant use of emic learning behavior tracking techniques. A must-read contribution to the literature."

## **Numa Markee**, *University of Illinois at Urbana-Champaign, USA*

"This book is a double intellectual treat: a well-executed longitudinal study with enlightening fndings and a thoughtful discussion of theoretical and methodological issues with far-reaching implications for future research. An in-depth exploration into the development of interactional competence in a second language, it is a stellar addition to the feld."

**Hanh thi Nguyen**, *Hawaii Pacifc University, USA* 

# **The Development of L2 Interactional Competence**

This book presents unique insights into the development of L2 interactional competence through the lens of complaining, demonstrating how a closer study of complaining as a social activity can enhance our understanding of certain aspects of language learning with implications for future L2 research.

 The volume employs a multimodal, longitudinal conversation analytic (CA) approach in its analysis of data from video-recorded interactions of several elementary and more advanced L2 speakers of French as they build their interactional competence, understood as the ability to accomplish social actions and activities in the L2 in context-dependent and recipient-designed ways. Skogmyr Marian calls attention to three key dimensions of complaining in these conversations – its structural organization, the interactional resources people use when they complain, and how speakers' shared interactional histories and changing social relationships afect complaint practices. The volume underscores the fundamentally multimodal, socially situated, and co-constructed nature of L2 interactional competence and the socialization processes involved in its development, indicating paths for new work on interactional competence and L2 research more broadly.

This book will be of appeal to students and scholars interested in second language acquisition, social interaction, and applied linguistics.

**Klara Skogmyr Marian** completed her PhD and postdoc at the University of Neuchâtel, Switzerland, and is now assistant professor at Stockholm University, Sweden. Her research focuses primarily on L2 learning and social interaction from a conversation analytic and multimodal perspective. Her works have been published in *Research on Language and Social Interaction*, *Frontiers in Psychology*, and *The Modern Language Journal*.

# **Routledge Advances in Second Language Studies**  Edited by John Hellermann and Søren Wind Eskildsen

**Racialized Identity in Second Language Learning**  Speaking Blackness in Brazil *Uju Anya* 

**Teaching and Testing L2 Interactional Competence**  Bridging Theory and Practice *Edited by M. Rafaiel Salaberry and Silvia Kunitz* 

**A Conversation Analysis Approach to French L2 Learning**  Introducing and Closing Topics in Everyday Interactions *Clelia König* 

**Interaction, Language Use, and Second Language Teaching**  *Thorsten Huth* 

**The Development of L2 Interactional Competence**  A Multimodal Study of Complaining in French Interactions *Klara Skogmyr Marian* 

For more information about this series, please visit: www.routledge.com/ Routledge-Advances-in-Second-Language-Studies/book-series/RASLS

# **The Development of L2 Interactional Competence**

A Multimodal Study of Complaining in French Interactions

**Klara Skogmyr Marian** 

First published 2023 by Routledge 605 Third Avenue, New York, NY 10158

and by Routledge 4 Park Square, Milton Park, Abingdon, Oxon, OX14 4RN

*Routledge is an imprint of the Taylor & Francis Group, an informa business* 

Published with the support of the Swiss National Science Foundation

© 2023 Klara Skogmyr Marian

The right of Klara Skogmyr Marian to be identifed as author of this work has been asserted in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988.

The Open Access version of this book, available at www.taylorfrancis. com, has been made available under a Creative Commons Attribution 4.0 license.

*Trademark notice*: Product or corporate names may be trademarks or registered trademarks, and are used only for identifcation and explanation without intent to infringe.

*Library of Congress Cataloging-in-Publication Data*  A catalog record for this book has been requested

ISBN: 978-1-032-22116-8 (hbk) ISBN: 978-1-032-22119-9 (pbk) ISBN: 978-1-003-27121-5 (ebk)

DOI: 10.4324/9781003271215

Typeset in Sabon by Apex CoVantage, LLC **To Cornel** 

# **Contents**


	- *3.4 Complaining in L2 interaction 37*
	- *3.5 Cumulative evidence about complaining and research gaps 39*

# **4 Investigating change longitudinally: methodological concerns and data** 42

	- 4.3.1 Setting and participants 46
	- 4.3.2 Recordings and supplementary material 47
	- 4.3.3 Transcription and anonymization 47
	- 4.3.4 Determining the analytical focus and establishing collections 48
	- 4.3.5 Focal participants 49
	- 4.3.6 Overview of collections 51
	- 4.3.7 Comparability 52

### **5 The structural organization of L2 complaints** 54

	- 5.2.1 Elementary speakers 63
	- 5.2.2 Upper-intermediate/advanced speakers 70
	- 5.3.1 Elementary speakers 80
	- 5.3.2 Upper-intermediate/advanced speakers 91

### **6 Interactional resources for complaining in the L2** 105

	- 6.1.1 Elementary speakers 106
	- 6.1.2 Upper-intermediate/advanced speakers 118
	- 6.1.3 Negative assessments: quantitative comparison 127
	- 6.2.1 Elementary speakers 130
	- 6.2.2 Upper-intermediate/advanced speakers 141
	- 6.2.3 Direct-reported speech and reenactments: quantitative comparison 149

# **7 The interactional history of a complainable** 155 *7.1 Case study 1: Suresh 155*  7.1.1 Profering a complaint about the heat 156 7.1.2 Profering a complaint about the cold 160 7.1.3 Suresh: summary and intermediate discussion 172 *7.2 Case study 2: Malia 174*  7.2.1 Fall semester of 2016 (months 1–3) 174 7.2.2 Spring semester of 2017 (months 4–9) – same coparticipants 187 7.2.3 Fall semester of 2017 (months 11–15) – new coparticipants 196 7.2.4 Malia: summary and intermediate discussion 209 *7.3 Discussion 211*  **8 Discussion of results and perspectives** 215 *8.1 Understanding L2 interactional competence and its development 215*  8.1.1 Stability in the basic composition of conversational activities 217 8.1.2 Turn-taking management: increased synchronization and co-construction 218 8.1.3 Sequence and preference organization: diversifcation of methods 219 8.1.4 Linguistic resources for action: diversifcation and routinization 220 8.1.5 Language and the body for action: change in multimodal practices 222 8.1.6 Socialization processes in L2 learning: shared experiences as an interactional resource 223



# **Foreword**

The study of second language acquisition has come a long way. What started in the 1970s with a focus on the development of learners' grammars was revolutionary since it claimed and demonstrated that second language development followed linguistic regularities and was not just a chaotic ragbag of errors. Over the recent decades, interest has broadened from interlanguage to interactional competence, where *interaction*  refers to the full ecology of social practices, that is, to the whole range of linguistic and embodied resources that speakers make use of in interactions. To understand the order of social practices and their change over time, studies must focus on the development of recognizable, well-defned social practices, their sequential structure, and the changes observable over time.

Very few studies of this kind are currently available. The research into the development of social practices is still at its beginning, and one of the reasons that it moves so slowly is that this research needs carefully selected video documentation of environments where participants regularly meet and talk and form social bonds between them. Collecting these data is difcult and time-consuming.

Klara Skogmyr Marian's book is an impressive example of what studies of this kind can contribute and what we can learn from them. The practice studied is *complaining in interaction*. Analytically, complaints are interesting since they are delicate and complex social activities that make them a proper analytic object for studying the development of second language speakers' interactional competence. The data are drawn from *conversation circles* of students at a university in the French-speaking part of Switzerland. Complaining is quite common in these data and seems to be a practice through which the participants build rapport as brother- or sisterhood of *les misérables*. Complaints work as a resource to forge bonds between the students, as learners of French, as (PhD) students, and as non-Swiss citizens in a majority society. This is the sociological side shown by this book. Complaining is an interactional achievement, it has a social role, and these students do it increasingly competently.

With respect to the function and structure of complaints, this book contributes to our general understanding of complaints as a ubiquitous social activity. Complaints are complex in that they are sensitive to the evolving social relations between the participants and draw on a wide variety of possible topics and linguistic and embodied resources. The book does not only break new grounds in its detailed descriptions of complaining as an embodied multimodal activity, but it also informs about complaining's role in the social relations of the people who take part in it. Complaints are one resource to build interactional histories with other people and to create friendships and solidarity.

Skogmyr Marian's detailed longitudinal study of participants' interactions shows not only the development of their interactional competence but also the ways in which the students build social relations where their interactional competence is the central tool for socializing into new relations. While the overall structural composition of complaining is essentially similar among elementary and more advanced speakers, the ways in which complainants and their hearers move into complaints difer. More advanced speakers do more joint complaints and more afliative work and contribute to the sequential unfolding of the complaints. Over time, complaints are produced more fuently, are better synchronized, and emerge as co-constructed talk. In other words, the participants progressively diversify their interactional procedures – or 'ethnomethods'.

The book demonstrates convincingly that learning a second language is about becoming a member of a new social world. Only a few studies come to mind that have been able to show this, but none of them demonstrates a comparable richness of the material.

> Johannes Wagner *Professor, Department of Design and Communication, University of Southern Denmark*

# **Acknowledgments**

 This book constitutes a revised version of my doctoral dissertation, which I defended at the University of Neuchâtel, Switzerland, in October 2020. The pre-press stage of this publication and its Gold Open Access were supported by the Swiss National Science Foundation (grant no. 10BP12\_213629). I also thank the Swiss National Science Foundation for generously fnancing my research stay in Denmark (grant no. P1NEP1\_184343).

My heartfelt thanks go to the participants of the study, who agreed to be recorded and have their interactions studied in micro-detail. Without you, this book would never have seen the light of day.

I sincerely thank Simona Pekarek Doehler for supervising my PhD research, for providing generous scientifc guidance and feedback, and for creating an intellectually stimulating and constructive research environment in Neuchâtel. I thank Esther González-Martínez, Catrin Norrby, and Johannes Wagner for accepting to be part of my dissertation committee and for ofering useful insights on my research from diferent perspectives. I especially thank Johannes Wagner for inviting me to carry out a research stay at the University of Southern Denmark during my PhD studies and for ofering valuable comments during data sessions and on early drafts of my analyses. I am also grateful to Numa Markee for many helpful remarks and encouraging words during the preparation of both my dissertation and the book manuscript.

Many thanks go to the series editors, Søren Wind Eskildsen and John Hellermann, as well as Elysse Preposi and her colleagues at Routledge, for helpful comments and suggestions and for highly professional support throughout the publishing process.

I am thankful to my colleagues in Neuchâtel, Stockholm, and elsewhere (Adam Jones, Loanne Janin, Sophia Fiedler, Mélissa Juillet, Ioana-Maria Stoenica, Anne-Sylvie Horlacher, Silvia Kunitz, Olcay Sert, Ufuk Balaman, and the IVIP research group, to only mention a few) for inspiring scientifc discussions and well-needed cofee breaks and *apéros*.

Last, I thank my family and friends for their unwithering support throughout my PhD and during the authoring of this book. Special thanks go to Ronja Sannasdotter for the many serious and non-serious conversations about this wonderful thing called life and for putting up with my constant absences.

I dedicate this book to my husband Cornel, who day after day teaches me how to be a better version of myself – both as a researcher and as a human being.

# **1 Introduction**

Most of us have gone through the process of learning a second/foreign/ additional language (henceforth *second language* or *L2*) at some point in our lives, during school years and perhaps later in life. The opportunities for L2 learning have never been as diverse and easily accessible. Traditional classroom instruction is only one of many means through which people may develop L2 skills. With the help of inexpensive online courses and mobile applications, people across the world and their lifespan can practice vocabulary and grammar and basic dialogues in countless languages, even in times of a worldwide pandemic when international travel (and for many, also regular classroom instruction) may be restricted. The technological advances that allow for such opportunities are truly incredible, and their contribution to the democratization of language learning merits nothing less than awe. But regardless of the manifold possibilities for remote language learning – whether through foreign-language instruction or technological means – most people who eventually set foot in a place where the L2 is the main language of communication will be struck by the challenge of putting their L2 knowledge to use in real-life situations. All of a sudden, people seem to speak too fast and use regional accents and expressions, and the language learned through courses in high school or mobile applications just does not sufce for participating in spontaneous conversations.

Why is it so difcult to learn how to interact in another language, that is, to gain L2 interactional competence (IC)? And given this difculty, why do we not focus more on interactional skills in language instruction? As underscored for a long time within usage-based and socio-interactional strands of Second Language Acquisition (SLA) research, language learning is inextricably related to language *use* (see Cadierno & Eskildsen, 2015; Douglas Fir Group, 2016; Firth & Wagner, 1997, 2007; and Pekarek Doehler, 2010, among others). One cannot expect to become a profcient L2 interactant without actually practicing social interaction, without partaking in diferent types of conversations and social encounters in the L2. This is why learners need occasions to practice authentic language use and ecologically sound guidance on what is needed to participate efectively in L2 interactions. Unfortunately, due to a long-standing tradition within SLA to disregard the micro-details of social interaction (see Firth & Wagner, 1997) and to focus on distinct speaking and listening skills, the attention paid to IC is still limited in central policy documents like the Common European Framework of Reference (CEFR; Council of Europe, 2020) and, consequently, in language education. In addition, research on the *longitudinal trajectories* involved in the development of L2 interactional skills is even more scarce (Pekarek Doehler & Pochon-Berger, 2015), which means that we know very little about the steps speakers go through as they gain interactional profciency. Pedagogical resources for L2 learning, therefore, often fail to provide learners with empirically grounded advice on how to improve their interactional skills over time (Huth, 2020; Salaberry & Kunitz, 2019a). To change the view of L2 interaction and how it is treated within language instruction today, more research is needed on what it means to develop L2 IC.

 This book contributes to such a research endeavor by investigating longitudinally how L2 speakers of French develop their interactional practices for complaining in interaction. The focus on complaining may seem strange. Why would it be relevant to investigate complaining from an L2 perspective (and to dedicate a whole book to it)? There are two main motivations for this. The frst reason relates to the ubiquity of complaining in our daily lives, and the second to the interactional complexity of the phenomenon.

Complaining has been a human concern for quite some time. Take Seneca's observation from AD 49 that "[t]he majority of mortals . . . complain bitterly of the spitefulness of Nature" (Seneca, 49/1932). In layman's terms, the notion of complaining typically refers to the action of expressing grief, discontent, or some other type of negative stance toward a person, an object, or a situation. This can be done either informally, to a close person, a colleague, or a fellow commuter train passenger, or formally, to the electricity provider, workers union, housing agency, and so on. We complain about the weather, our bosses, our in-laws, our inability to remember to return overdue books to the library, and broken heaters, unacceptable working conditions, and loud neighbors. Based on such everyday experiences, we have likely acquired a basic appreciation, or commonsense 'members' knowledge' in Garfnkel's (1967) terms, of what it means to complain in diferent spheres of life. People's basic awareness of and interest in complaining can even be seen in the vast domain of self-help books, with popular-science eforts to help people complain 'the right way' to improve various situations (e.g., Winch, 2011). But contrary to both Seneca's (49/1932) rather negative account of complaining and to popular belief, some sociological and interactional research has highlighted more positive dimensions of this phenomenon, which is something that will also be shown throughout this book. Complaining allows people to share their troubles, 'let of steam', and typically express

their emotional support and sympathy with each other (Drew, 1998; Heinemann & Traverso, 2009; Selting, 2012), and may therefore have the positive potential of strengthening social relationships (Boxer, 1993; Günthner, 1997; Hanna, 1981). Such social dimensions likely contribute to the prevalence of complaining in our daily interactions.

The second and perhaps even more important reason is that the complex nature of complaining makes it an ideal analytical object for investigating the development of L2 speakers'1 interactional skills over time. It is not until the last 40 years or so that complaining as a social activity has undergone systematic scientifc scrutiny, and recent micro-analytic research has precisely highlighted the interactional intricacy of the activity (see particularly Chapter 3). As I illustrate in Section 1.1, complaining entails the organization of long sequences of actions, fne-grained social coordination between participants, the management of delicate talk, stance-taking, afliation, and other related facets of social interaction that are essential to an ordinary conversation but may be challenging for L2 speakers. Complaining thus provides a site for many central components of what it means to be interactionally competent, yet it has never been investigated from a micro-analytic, action-oriented, and longitudinal perspective in SLA. This is precisely what I do in this book.

### **1.1 A praxeological research perspective**

The research presented in this book draws on conversation analysis (CA) to investigate L2 use and learning, an approach also called CA in/for SLA, or CA-SLA. The use of CA entails the adoption of a set of epistemological and methodological principles that have implications for the conceptualization and study of L2 talk and of complaining (see particularly important notions in bold).

CA is a **micro-level**, **praxeological** (action-oriented) approach that analyzes social interaction **sequentially**, turn-by-turn, to observe how speakers show each other that they orient to each other (Sacks et al., 1974; Scheglof et al., 1977). The analyst is thus interested in the participants' own interpretations of the unfolding interaction, adopting what we call an insider's or **emic perspective** (Kasper, 2006). In the context of L2 research, this means, for example, that the analyst does not treat linguistic errors in the participants' talk as problematic unless the participants themselves orient to such errors as problematic for establishing mutual understanding. The praxeological approach furthermore means that the analytical focus is not on linguistic structures *per se*. Language is only one of the various semiotic resources that humans deploy to do things (such as to assess, disagree, or explain a word) in interaction. Video recordings of interactions allow the researcher to consider the range of linguistic, prosodic, and sometimes embodied2 resources that speakers use to reach intersubjectivity (e.g., Goodwin, 2000, 2018; Mondada, 2014; Streeck,

2009). My research is part of the growing eforts within CA-SLA to investigate L2 development from such a **multimodal perspective**.

The praxeological approach has profound implications for the conceptualization of learning. L2 learning is not understood as the individual learner's internalization of increasingly target-like linguistic structures, as is typically the case within cognitivist SLA approaches (see, e.g., Doughty & Long, 2003; VanPatten & Williams, 2015). Instead, both the goal and driving force of L2 learning are speakers' efective participation in meaningful interactions. L2 speakers are seen as active agents who, over time, in and through social interaction, work to become increasingly more competent members of the L2 community (Pekarek Doehler, 2010; Wagner & Gardner, 2004). While some research within CA-SLA focuses on L2 learning as an interactive process (e.g., Markee, 2008; Sahlström, 2011), this book primarily investigates the longitudinal outcomes, or 'products' of L2 learning – even though the two are naturally intimately interrelated. It does so by documenting changes in L2 speakers' interactional skills over time or, in other words, the *development of L2 interactional competence* (Hellermann, 2008; Pekarek Doehler & Pochon-Berger, 2015).

This perspective also has consequences for our understanding of complaining. Conversation analysts are interested in complaining as a situated, social activity and as an interactive process. Sequential analysis of authentic complaint sequences allows us to document the precise actions and interactional resources people use when complaining, in what way recipients contribute to the complaint, and how the complainant (the person complaining) works to obtain afliative or sympathetic responses from coparticipants.

Excerpt 1.1 illustrates the relevancy of investigating complaining from a conversation analytic and developmental L2 perspective. The excerpt shows the L2 French speaker Aurelia (AUR) complaining to her coparticipant Mia (MIA). The object of the complaint, that is, the *complainable*, is some undesirable conduct of people in Switzerland, a recurrent complaint topic in my data. The analysis highlights interactional resources that participants regularly deploy to construct complaints (in bold) and the interactive process through which the participants negotiate the development of the sequence.

The complaint is already underway as we join the excerpt. Aurelia has criticized Swiss people for being aggressive and 'hostile with rules'. In lines 1–2, Mia elicits an account (a justifcation) by asking Aurelia whether someone has been angry at her, and in line 12, Aurelia initiates a **story**  through which she develops and backs up the complaint with a precise example (Günthner, 1995).

*Excerpt 1.1* 'Better inside' (Lun\_2018–02–26)

## *Introduction*

 In the omitted lines (lines 13–39), Aurelia introduces the story setting, explaining that she and her friend were walking in the city and stopped in a space between two streets to greet some other friends. Having specifed that it was not *in* the street (line 40), Aurelia introduces the antagonist of the story ('a man', line 42), who stopped his car and started scolding Aurelia and her friends (lines 42–43) for not standing in the 'good place' (line 44). Aurelia then uses a range of verbal and embodied means to detail and criticize the man's conduct and show how the situation afects her negatively. Through **direct-reported speech** (lines 43–44, 46–47, 53–55), **marked prosody** (vowel elongations, stress), and **gestures** (FG.1.1), Aurelia **reenacts** (Sidnell, 2006) the event and contrasts her own, morally defensible conduct with the unreasonable and morally reprehensible behavior of the man (Drew, 1998; Drew & Holt, 1988). Twice, after each report of the man's talk, Aurelia pauses (lines 45, 56) and displays her stance by **frowning** and fipping her hands palm up in an expression of disapproval and incomprehension (FG.1.2; see also line 56). With a **rhetorical question** (line 57), Aurelia concludes the telling and again shows her disapproval of the man's conduct.

 Mia participates in the complaint by eliciting elaborations and clarifcations (lines 1, 5, 10–11, 48–49), granting Aurelia access to the foor for extended turns, and producing listenership tokens at appropriate moments (lines 41, 51). Following the end of Aurelia's telling, she also provides an **assessment of the story** in the form of a **non-lexical vocalization**, *wo*↓*:w* (line 59), by which she displays some emotional support, or afliation, with Aurelia. As seen in Aurelia's embodiedly completed **summary assessment** (line 60, FG.1.3) and **generalization** of her criticism (lines 60–61), Aurelia thereafter expands the complaint until she receives stronger displays of afliation from her coparticipant (not shown).

The brief analysis of this excerpt demonstrates what a CA analysis of authentic interactions can contribute with to our understanding of complaining. The excerpt sheds light on the immense interactional complexity of complaints. Besides exemplifying interactional resources for complaining, the excerpt showed the interactive process involved in complaints, the status of complaints as social activities rather than actions produced by a single speaker, and the participants' emic orientations to the interactional purpose of complaining, that is, to obtain afliation or sympathy. But the excerpt also gives rise to questions regarding complaining from an L2-learning perspective. The sequence involved speakers with quite advanced profciency in French. At some points in the interaction, the participants engaged in repair practices that displayed language-related difculties (e.g., line 2), but these did not seem to threaten mutual understanding. In many respects, the complaint resembles complaints of L1 speakers (similar structure and basic components, the use of negative assessments, direct-reported speech, etc.). What about the complaints of L2 speakers with lower profciency, who may

# 8 *Introduction*

not have as diverse interactional resources in the L2 at their disposal as more advanced speakers? How do they engage in complaining, and how do they respond to other speakers' complaints? And how does the developmental trajectory of complaint practices look like as speakers gain L2 profciency?

### **1.2 Aim and research questions**

This book aims to enrich our understanding of the longitudinal trajectories involved in the development of L2 IC. To do so, it investigates how L2 French speakers over time change their practices for complaining about non-present third parties, objects, or situations. By means of three empirical sub-studies, the book addresses the following research questions:


The data of the study consist of video recordings of adult L2 speakers of French participating in a 'conversation circle' in the French-speaking part of Switzerland. The recordings took place in a university cafeteria, where the participants met regularly to interact in French over a cup of cofee. The analysis adopts a longitudinal and pseudo-longitudinal comparative perspective, analyzing speakers' multimodal practices for complaining over time and across profciency levels.

In sum, this book tackles L2 learning from a praxeological perspective that focuses on social action from a holistic, multimodal point of view and that considers L2 speakers as active and cooperative social agents. The methodological principles of CA allow me to analyze the micro-level details of social interaction from a participant-relevant perspective. By investigating primarily the longitudinal 'products' of L2 learning over time, this research contributes to the cumulative evidence about the overall trajectories involved in the development of L2 IC, which ultimately has the potential to inform pedagogical practices for the teaching and learning of L2 interactional skills.

### **1.3 Outline of the book**

The book comprises eight chapters. Following this introduction, Chapter 2 situates the study in its larger theoretical research framework and reviews empirical studies on the development of L2 IC. Chapter 3 addresses the main analytical object of the book, complaining, by presenting prior work on complaining in interaction. Chapter 4 outlines the methodological approach and the empirical material of the study and discusses challenges associated with the research design and the interpretation of the fndings. Chapters 5, 6, and 7 present the results of the three empirical sub-studies. In Chapter 5, I investigate the interactional organization of complaining. I show that the overall structural composition of L2 complaints remains the same over time but that there is a longitudinal change in terms of practices for initiating complaints and in how participants together co-construct complaint sequences. Chapter 6 examines the multimodal interactional resources participants deploy to construct complaints. In that chapter, I demonstrate a longitudinal change in speakers' practices for negatively assessing and for reporting on other people's talk or conduct. Chapter 7 presents two case studies that longitudinally track specifc recurrent complainables in two participants' interactions. I document how speakers draw on their shared interactional histories and deepening social relationships with their coparticipants to engage in complaints in context-sensitive and recipient-designed ways. In Chapter 8, I discuss the implications of the empirical fndings for our understanding of the development of L2 IC and for complaining in interaction and suggest directions for future research.

# **Notes**


# **2 L2 interactional competence and its development**

The past few decades have seen an unprecedented interest in the social dimensions of L2 learning. Social-interactional research strands that were until recently called 'alternative approaches' to SLA (Atkinson, 2011a) can hardly be qualifed as such today, considering the substantial expansion of these strands and the increased interest in more holistic views of L2 learning and teaching across disciplines (Douglas Fir Group, 2016; but see Atkinson, 2019, on the continued prominence of cognitivist SLA). Specifcally, CA research on L2 interaction has expanded remarkably in the last ten years, establishing CA-SLA as a prominent research paradigm within applied linguistics. The same research has ofered a wealth of empirical investigations into the nature of L2 IC and has increasingly documented the longitudinal trajectory of its development. In this chapter, I outline the historical developments within SLA that have led to current understandings of L2 IC (Section 2.1) and review empirical CA work on the development of L2 IC (Section 2.2). I then synthesize existing fndings and identify gaps in the literature (Section 2.3).

### **2.1 Epistemological roots and current understanding of L2 IC**

The present work is part of a research tradition within SLA that emerged in the 1990s as an attempt to reconceptualize language learning in social terms. In a seminal paper published in the midst of the SLA 'paradigm wars' (Douglas Fir Group, 2016), Firth and Wagner (1997) criticized mainstream SLA for being too individualistic, theory-driven, and concerned with learners' errors. Instead, they advocated for a social, datadriven, and emic approach to SLA. The authors highlighted the crucial role of social interaction in language learning and the need for empirical research on what L2 speakers *can do* in social interaction rather than on what they *cannot do*. Firth and Wagner, together with other scholars who embraced the same ideas, embarked on an overarching endeavor to refne the understanding of social interaction in L2 learning (Hellermann, 2008; Markee, 2000; Pekarek Doehler, 2010). The same movement gave birth to

the feld of CA-SLA (Kasper & Wagner, 2011; Markee & Kasper, 2004), an approach to SLA that draws on the epistemological underpinnings and methodological tools of CA.1 Most of the recent research on the development of L2 IC emanates from this approach.

The interest in L2 IC can be traced back to early criticism of Chomsky's (1965) dichotomy between linguistic competence and performance, which presupposed that language competence was an innate capacity worthy of scientifc inquiry, whereas language performance was merely a messy by-product to be ignored by linguists. The US American sociolinguist Hymes (1972) opposed the idea of a separation between competence and performance and instead argued for a more holistic and contextual view of speakers' communicative abilities. He introduced the term *communicative competence* as an alternative, a notion that considered speakers' sociolinguistic and pragmatic competence in addition to purely linguistic knowledge. Others (e.g., Canale & Swain, 1980) have later adopted and redefned the same term, and the study of communicative competence is still today a prosperous research feld – for example, within the strand of interlanguage pragmatics (see the chapters in Bardovi-Harlig & Hartford, 2005; Kasper & Blum-Kulka, 1993).

Along with the conceptual expansions and transformations of SLA as a feld around the 1990s came criticism also of the notion of communicative competence. Communicative competence was seen as too static of a notion, too focused on individual learners and their (in)capacities, and not sensitive enough to the micro-level workings of social interaction. Within the felds of sociocultural learning and situated learning theory, researchers started documenting L2 speakers' 'interactive' or 'discursive' practices (Hall, 1993, 1995; Young, 2000, 2003; Young & Miller, 2004). The term *interactional competence*, notably deployed by Kramsch (1986) in a critique of the accuracy-concerned language profciency movements in the US, has later been adopted by researchers within CA-SLA. While the debate about the most suitable terminology is still ongoing (see Hall, 2018; Markee, 2019), most researchers presently conducting empirical developmental research using ethnomethodologically inspired CA (EMCA) deploy the term *interactional competence*.

From the perspective of EMCA, L2 IC may be described as speakers' interactional 'methods' (Garfnkel, 1967) for accomplishing social actions and establishing mutual understanding in L2 interaction. These methods involve systematic interactional procedures for managing – for example, turn-taking, repair, topic transitions, and disagreements – in a way that is sensitive to the recipient and to the local circumstances of the interaction (Hellermann, 2008; Pekarek Doehler & Pochon-Berger, 2011, 2015; Pekarek Doehler & Berger, 2018). An important aspect of this competence is its fundamentally socially shared and co-constructed

## 12 *L2 interactional competence and its development*

nature (Greer, 2019; Hauser, 2019; He & Young, 1998; Kasper & Wagner, 2014). The development of L2 IC, in turn, refers to change over time in L2 interactional methods.

The relationship between longitudinal change in interactional practices, the development of L2 IC, and language learning is not unproblematic from a CA perspective, specifcally as it concerns the possibility of maintaining an emic, participant-relevant perspective when interpreting change over time in terms of development. I return to this important discussion in Chapters 4 and 8. At this point, sufce it to say that empirical studies on the development of L2 IC often document increased recognizability and local efcaciousness of interactional practices over time, which may be considered emic evidence for development (Wagner et al., 2018).

#### **2.2 Empirical fndings about the development of L2 IC**

Empirical research on the development of L2 IC has started uncovering the precise workings of this development (see also Skogmyr Marian & Balaman, 2018, for an overview). There is growing evidence for how speakers develop both their practices for accomplishing social actions and activities in the L2 and how they use (and change their use of) specifc linguistic resources for interactional purposes. The empirical basis now includes investigations of a range of analytical objects, L2s, and types of interactional contexts. A basic observation derived from these studies is that the foundational mechanisms governing the generic features of interaction appear to remain the same in L2 interactions over time and across profciency levels. That is, L2 interactions are not based on entirely diferent mechanisms for sequence organization, turn-taking, and repair than L1 interactions. They rather refect the orderly, basic infrastructure for social interaction that all humans share (Levinson, 2006). L2 speakers likely draw on certain aspects of their IC from the L1 and apply these in the L2. However, as argued elsewhere (Pekarek Doehler & Pochon-Berger, 2015), IC is not simply *transferred* from the L1 to the L2. Speakers progressively 'recalibrate' and adapt their interactional practices, typically to increasingly approximate the practices of L1 speakers. In what follows, I present a selection of the work showing this recalibration, providing both an overview of general fndings and specifc observations that are of importance for the study of L2 complaining.

#### *2.2.1 Developing practices for action*

Developmental studies on social actions and activities typically investigate how one or a few participants (sometimes a group of participants) accomplish one specifc action or activity in similar environments over time. Studies include investigations of disagreements (Pekarek Doehler & Pochon-Berger, 2011), displays of active listenership and other types of recipient responses (Dings, 2014; Ishida, 2011; Kunitz & Yeh, 2019; Sert, 2019; Taleghani-Nikazm, 2019), the opening of phone calls, tasks, and storytellings (Brouwer & Wagner, 2004; Hellermann, 2008; Pekarek Doehler & Berger, 2018), repair (Balaman, 2016; Hellermann, 2009, 2011; Pekarek Doehler & Berger, 2019), requests (Al-Gahtani & Roever, 2012; Youn, 2015), topic management (Hellermann & Lee, 2021; Kim, 2017; König, 2019; Kunitz & Yeh, 2019; Nguyen, 2011), and turnconstruction and turn-taking (Cekaite, 2007; Nguyen, 2019; Watanabe, 2016; Young & Miller, 2004). Most of these studies converge in showing a longitudinal diversifcation of interactional methods for accomplishing the studied actions or activities, a complexifcation of methods (for example, longer turn-constructional units, TCUs), and an increased ability over time to adjust the methods to the interactional context and the recipients.

One particularly relevant set of studies on conversational actions concerns the opening of storytellings. Both Hellermann (2008) and Pekarek Doehler and Berger's (2018) longitudinal investigations document the emergence and use of increasingly elaborate prefatory work in storytelling openings in L2 English and French, respectively. As exemplifed in Chapter 1 and detailed in Chapter 3, storytelling typically constitutes an important component of complaints. Successful complaining thus often (but not always) requires the ability to initiate and manage storytelling, which can be a complex undertaking. Future storytellers need to suspend the regular turn-taking machinery to secure the right to a longer turn, and they usually establish the relevancy of the upcoming story to the prior talk and provide some indications about the nature of the story to help coparticipants anticipate appropriate responses (Jeferson, 1978; Mandelbaum, 2013; Sacks, 1974).

Hellermann (2008) and Pekarek Doehler and Berger (2018) show that L2 speakers without much L2 conversational experience typically initiate stories *in medias res*, without any interactional preparation or with only limited prefatory work. This often has negative consequences for the subsequent development and reception of the story since coparticipants are unprepared for what is coming and what is expected of them as story recipients. With time and increased L2 profciency, story prefaces become more frequent and interactionally more complex. Excerpt 2.1, from Pekarek Doehler and Berger (2018), illustrates the more elaborate prefatory work that the upper-intermediate L2 French-speaking *au pair* Julie (JUL) produces toward the end of her nine-month stay with the host family as compared to the beginning of the stay.

### 14 *L2 interactional competence and its development*

*Excerpt 2.1* (Pekarek Doehler & Berger, 2018: 571)

In response to the host mother Marie's (MAR) frst story, Julie ofers the disjunct marker *mais* ('but', line 4) to signal that the nature of the upcoming talk contrasts somehow with Marie's displayed stance (Pekarek Doehler & Berger, 2018). She then introduces the key referents of the story – Robert Giroud's children – in a stepwise manner through a left-dislocation to check the recognizability of the referent to the recipient (lines 5, 7; see rising intonation in line 5 and Marie's confrmation in line 8) before launching the story (line 10). The excerpt illustrates Julie's general tendency later during her stay to use more extensive prefatory work that displays relatedness to prior talk, secures recipiency, and foreshadows the nature of the upcoming telling. These practices help Julie recruit aligning and afliative responses, which was sometimes difcult for her at the beginning of the stay. Whereas Hellermann's (2008) fndings show a change from no or extremely minimal to *some* prefatory work in storytelling openings by beginner and lower-intermediate speakers of English, Pekarek Doehler and Berger's (2018) observations illustrate what happens further on in the developmental trajectory. The observation that more advanced L2 speakers increasingly foreshadow the stance that will be conveyed in the story already in the story opening is of high relevance for the present study, as stance displays are central to complaining.

Another particularly pertinent line of investigation examines longitudinal change in practices for taking conversational initiatives and providing relevant recipient responses to co-construct interactions and maintain the progressivity of talk (Dings, 2014; Hellermann & Lee, 2021; Ishida, 2011; Kim, 2017; König, 2019; Kunitz & Yeh, 2019;

 Sert, 2019; Taleghani-Nikazm, 2019). Because complaints are co-constructed activities (see Section 3.1), they require close collaboration by coparticipants. Several studies have observed that L2 learners at lower profciency levels often have difculties providing appropriate responses to other participants' contributions (Kunitz & Yeh, 2019; Sert, 2019; Taleghani-Nikazm, 2019). Focusing on informal multi-party interactions, Sert (2019) examines the development of practices for displaying active listenership and specifcally collaborative turn completions in out-of-class interactions recorded over two semesters of universitylevel English-as-a-foreign-language (EFL) instruction. At the beginning of the frst semester, the participants tend to organize their discussions in a rather 'monologic' manner, with longer turns by each speaker and few displays of active listenership from the coparticipants. Excerpt 2.2 comes from one of the groups' frst interactions. The participants are discussing data privacy.

```
Excerpt 2.2 (Sert, 2019: 149)
```
As highlighted by Sert (2019), the lack of recipient responses is notable specifcally in line 183, as SED has reached a transition-relevant place (line 182) after a longer turn, but no coparticipant self-selects to respond. SED, therefore, produces an open-ended question to recruit the coparticipants' expressions of opinions (line 184), and after a long silence (line 185), BUS fnally takes the turn (line 186). Once BUS's turn has come to an end (line 189), another silence ensues before AYS

self-selects to express her own stance (line 191). The interaction hence adopts a round-robin format through which the participants take longer turns one at a time and rarely ofer verbal displays of recipiency during coparticipants' turns.

Sert (2019) shows that, with time, the students' interactions become much more conversational in nature, adopting a more conventional turntaking system that involves more verbal displays of active listenership, including collaborative turn-completions. Excerpt 2.3 illustrates this change.

```
Excerpt 2.3 (Sert, 2019: 154)
```
Here coparticipants ofer candidate turn completions both in response to word searches (line 15) and in an anticipatory manner (lines 19, 21, 22), showing their active listenership and contributing to the progressivity of talk. Overall, Sert (2019) observed an increased number of collaborative completions over time and a more diverse use of listenership tokens with a growing interactional experience. The students' ability to complete other speakers' turns was closely related to their growing linguistic repertoires, observable in the emergence of subordinate clause completions, turn-initial conjunctions, and candidate lexical items (Sert, 2019). This fnding indicates that L2 speakers' increasing linguistic skills may allow them to participate as more active interactants in peer conversations. The study contributes to our yet-limited understanding of how L2 speakers develop their practices for managing turn-taking and jointly co-constructing social activities, something that I build upon in Chapters 5 and 6.

#### *2.2.2 Developing linguistic resources for interaction*

Longitudinal investigations that take linguistic resources as starting point have documented the emergence and use of particular discourse markers or grammatical constructions in interaction (Eskildsen, 2012; Ishida, 2009; Kim, 2009; Pekarek Doehler, 2018; Pekarek Doehler & Balaman, 2021; Pekarek Doehler & Berger, 2019), interactional uses of lexical items (Markee, 2008), and change over time in the use of language-body assemblies for social action (Eskildsen & Wagner, 2015, 2018; Pekarek Doehler & Skogmyr Marian, 2022). Studies that document longitudinal change in the use of linguistic resources for accomplishing precise social actions, such as assessments (Hellermann, 2008; Nguyen, 2019) or directives (Skogmyr Marian, 2018), similarly contribute to our understanding of the development of an L2 'grammar-for-interaction' (Pekarek Doehler, 2018), that is, linguistic resources used for action formation and interaction-organization (for recent contributions on this issue, see the papers in Eskildsen & Pekarek Doehler, 2022).

Given the central role of assessments in complaints (see Section 3.3.1), Hellermann's (2008) and Nguyen's (2019) observations about linguistic resources used in assessment turns in L2 English are relevant for the present study. Focusing on closing-implicative assessments, Hellermann documents that an English-as-a-second-language (ESL) student over ten months goes from producing an "awkward sounding pronoun-adjective combination" (*everybody nice*) to the use of repetition of an assessment adjective (*nice grandfather nice*) to more canonical adverb-adjective combinations such as *very good* (2008: 123). These observations suggest a possible trajectory of action formats specifcally for high-grade frst assessments, with repetition of assessment adjectives being used early on and the use of intensifying adverbials emerging later. Nguyen (2019), in turn, reports a progression from the use of a passe-partout format for positive assessments (*beautiful*) by a Vietnamese L2 speaker of English to a larger repertoire of assessment adjectives deployed in similar activity contexts. Both studies, although limited in scope, show how the diversifcation of linguistic resources contributes to assessment practices at very early stages of L2 development. In Chapter 6 (Section 6.1), I extend this line of inquiry by examining L2 speakers' linguistic resources used in negative assessments in complaints.

Research on L2 speakers' use of linguistic resources for interaction-organizational purposes (i.e., for managing aspects related to the architecture of interactions, see Pekarek Doehler, 2018), on the one hand, and on the role of embodied conduct in the development of L2 IC, on the other hand, is still scarce. Pekarek Doehler and Skogmyr Marian's (2022) study on the progressive routinization of a language-body assembly used for foor-holding in word searches ofers some empirical evidence about both of these issues, however. Both Pekarek Doehler and Berger (2019) and Pekarek Doehler and Skogmyr Marian (2022) observe L2 French speakers' initial tendency to use

### 18 *L2 interactional competence and its development*

the expression *comment on dit* ('how do you say') primarily to request assistance from a coparticipant to complete word searches, while subsequently routinizing the same expression as a marker of cognitive search serving foor-holding purposes. Pekarek Doehler and Skogmyr Marian (2022) show that this linguistic change co-occurs with a change in embodied conduct. Whereas *comment on dit* (often produced as *comment dit*, 'how to/do you say') deployed to request help consistently occurs with gaze at recipient and often depictive gestures (see Streeck, 2009), the same expression used as a marker of cognitive search co-occurs with gaze aversion (Excerpt 2.4):

*Excerpt 2.4* (Pekarek Doehler & Skogmyr Marian, 2022: 36)

Here Malia (MAL) employs *comment dit* ('how do you say') to display cognitive search and hold the foor while seeking to solve the word search herself – which she rapidly does (line 3). The prosodic delivery of the expression and the simultaneous embodied conduct are typical of this type of use: The expression is produced at a fast pace and with gaze averted from coparticipants until the end of the turn (FG.2.1–2.2). Javier's (JAV) understanding of Malia's verbal and embodied conduct as indexing cognitive search is visible in that he refrains from taking the turn and rapidly responds once Malia has returned her gaze to him (line 5). The excerpt exemplifes the emergence and progressive routinization of a discourse marker-like metalinguistic expression used to hold the foor. This fnding converges with earlier observations about L2 speakers' tendency to frst deploy linguistic constructions in a literal sense and, with time, develop more discourse marker-like uses of the same expressions (Pekarek Doehler,

 2018). In addition, the fnding about decreased use of response-mobilizing gaze (see Stivers & Rossano, 2010) and gestures in word searches over time concurs with Eskildsen and Wagner's (2015; 2018) observations about the linguistic and embodied change involved in an ESL speaker's use of the prepositions *under* and *across* (Eskildsen & Wagner, 2015) and constructions with the verbs *ask*, *tell*, and *say* (Eskildsen & Wagner, 2018). In both cases, the authors noted a reduced scope of gestures over time as linguistic resources took more prominent interactional roles.

Although still at an early stage, the research on the development of grammatical resources for interaction, on the one hand, and on change in embodied conduct over time, on the other hand, provides important avenues for more extensive investigations of these issues, to which the present work to some degree contributes (see Chapter 6).

### *2.2.3 Developing interactional routines and shared interactional histories*

The development of L2 IC is indisputably tied to socialization processes. Research within language socialization (e.g., Schiefelin & Ochs, 1986) and situated learning theory (Lave & Wenger, 1991) has for long stressed the locally specifc nature of (language) learning and investigated participants' socialization into precise communities of practice. Particularly some of the early studies on the development of L2 IC (or *interactive/discursive practices*) were inspired by these research traditions (Brouwer & Wagner, 2004; Cekaite, 2007; Young & Miller, 2004). For example, Brouwer and Wagner (2004) analyzed a series of telephone opening sequences in L2 Danish and German and observed mutual adaptation processes by the participants over time, which made the openings smoother and less problematic.

 Socialization processes have received very little attention in the more ethnomethodologically grounded research on the development of L2 IC. A few studies on L1-L2 conversations (Greer, 2019; Kim, 2017; Pekarek Doehler & Berger, 2019; Skogmyr Marian, 2018) indicate that people who are in the process of getting increasingly acquainted adjust their interactional practices to each other as they engage in similar types of interactional encounters over time. Some recent longitudinal CA studies have also shown the efects of speakers' shared interactional histories on recipient-design in L1 talk (e.g., Deppermann, 2018; Norrthon, 2019). The term *interactional history* refers to participants' shared interactional experiences and has been used by Deppermann (2018) as a way to conceptualize the establishment of common ground (Clark, 1996) – that is, shared meanings ascribed to semiotic resources and shared expectations about coparticipants' actions and language use.

Excerpt 2.5, from Kim (2017), illustrates these points. The conversation takes place between the ESL speaker Chungho (C) and the L1 speaker Tom (T), and it comes from the participants' tenth meeting, after six months of conversations-for-learning together.

# 20 *L2 interactional competence and its development*

*Excerpt 2.5* (Kim, 2017: 98; boldface added)

Chungho's topic announcement in line 268 breaks a lapse in the conversation (see lines 265 and 267). It builds upon the participants' common interest in cars, which they have discovered through repeated meetings. As discussed by Kim (2017), the announcement refects Chungho's growing participation in the conversations over time, whereby he, more frequently and with more diverse means, initiates new topics. Chungho thus draws upon the participants' shared interactional history to take interactional initiatives and be a more active interactant than at the beginning of the recording period. This change involves a growing capacity to adjust one's interactional practices to the particular recipient and interactional context, which pertains to both the L2 and L1 speakers involved. Reporting similar fndings, Greer (2019) speaks of a *joint* development of IC of all concerned parties as they together build shared interactional histories and interactional routines (see also Eskildsen, 2021a; Waring, 2013; Watanabe, 2016, on L2 speakers' increased participation through interactional routines, and Hellermann & Lee, 2021, on change in topic management and shifting participation frameworks in a group of L2 speakers engaging in similar types of conversations over time). I further explore the role of shared interactional histories and routines and evolving relationships in the development of L2 IC in Chapter 7.

### **2.3 Cumulative evidence about the development of L2 IC and research gaps**

To summarize existing research on the development of L2 IC, there is evidence that this development involves the following:

• A progressive *diversifcation* of interactional procedures, or methods, for accomplishing recognizable social actions and organizing interaction. This diversifcation involves the systematic use of an increasingly varied set of practices for accomplishing locally more efective actions, such as opening or closing a phone call, giving a directive, or making a request.


In addition, we have *some* evidence of the following:


The main consequence of these changes is an increased ability for recipientdesign and context-sensitivity (Sacks et al., 1974). When speakers gain IC, they become interactionally more agile; they manifest a higher level of interactional fexibility to adjust to local contingencies. They also accomplish locally more efcacious actions that are recognizable for what they are designed to do. These fndings converge with socio-cognitive approaches to SLA (Atkinson, 2011b) that view L2 learning as crucially involving 'learning how to ft in' (Atkinson, 2019), but the research on the development of L2 IC does so by relying entirely on socio-praxeological evidence.

Although the previously mentioned literature has provided many valuable insights into the nature of the development of L2 IC, there is yet much to discover. More research is needed to answer the following questions:

• How, more precisely, do L2 speakers develop their practices for jointly coordinating larger conversational *activities*? How does an increased ability to synchronize and co-construct conversational activities manifest itself?

# 22 *L2 interactional competence and its development*


Studies involving L2 speakers interacting spontaneously with other L2 speakers would specifcally allow for a better understanding of what these speakers can do without the assistance of teachers or L1 'experts'. Such research will provide useful insights, for example, for the feld of *lingua franca* studies (e.g., Mauranen & Ranta, 2009), but also for the many assessment contexts in which language students are assessed based on their interactional conduct with other L2 speakers (Sandlund et al., 2016). I address these issues through the empirical studies presented in Chapters 5–7.

# **Note**

1. For more on CA and its methodological premises, see Section 4.1.

# **3 Complaining in L1 interaction**

Chapter 1 briefy illustrated what a micro-level sequential analysis of a complaint sequence can contribute to our understanding of complaining as a social phenomenon and why it is relevant to study this phenomenon from a developmental L2 perspective. This chapter goes deeper into the notion of complaining and reviews prior fndings about this topic from primarily the CA literature. The chapter covers the core features of complaints (Section 3.1), the structure of complaint activities (Section 3.2), recurrent actions and interactional resources deployed in complaints (Section 3.3), and how complaints have been scarcely addressed in the L2 literature (Section 3.4). The chapter closes with a summary and discussion of research gaps (Section 3.5).

#### **3.1 Core features of complaints**

The terms *complaint* and *complaining* are colloquially used to designate several diferent types of interactional phenomena. Sometimes people express their dissatisfaction with an issue directly to the person or entity (institution, group, etc.) that has caused the problematic situation: These complaints may be called *direct complaints*. Concrete examples include complaints about a defective product addressed to the manufacturer and complaints to one's partner about his or her late working hours. Such complaints are often used in an attempt to change the situation to the better in some way (see Dersley & Wootton, 2000; Kevoe-Feldman, 2018; Laforest, 2009; and Scheglof, 2005, among others).

In contrast, the present study concerns what may be called *indirect complaints*. In indirect complaints, people express dissatisfaction with a (typically non-present) third party, object, or state of afairs to someone who is not in any way responsible for having caused the negative situation (e.g., Drew, 1998; Drew & Holt, 1988; Drew & Walker, 2009; Holt, 2012, Ruusuvuori et al., 2019; Traverso, 2009). The interactional 'project' (Levinson, 2013) embodied through these complaints is typically not to change the problematic circumstances but to obtain afliation and/ or sympathy from the recipients. Indirect complaining has therefore been argued to serve important interpersonal purposes (Boxer, 1993; Günthner, 1997; Hanna, 1981). Examples include complaints about tiresome work tasks, difcult course work, or bad weather to fellow coworkers, classmates, or neighbors who are likely to understand the negative experience themselves. The social purposes of indirect complaints are thus radically diferent from direct complaints, and this is refected in their interactional accomplishment and in the way recipients respond to the complaints. The following literature review focuses mainly on indirect complaints. Before going deeper into the interactional workings of these complaints, I summarize the core features of complaining as highlighted in prior research.

Complaining involves the basic tenet of expressing a negative stance toward a person, a thing, or an issue, that is, the *complainable* (the object of the complaint, Scheglof, 2005), that has afected the *complainant* (the person complaining, Drew, 1998) personally. Although negative stance expressions sometimes are subtle (Ruusuvuori et al., 2019), in many cases the complainant uses overt interactional means that clearly index the unreasonable, egregious nature of the situation and show the speaker's afective1 involvement in the activity (Drew, 1998; Günthner, 1997; Selting, 2010a). To justify negative stance expressions, speakers engage in accounting practices, for example, by detailing the complaint-worthy situation or conduct through reports or storytellings (Drew, 1998; Drew & Holt, 1988; Günthner, 1995; Selting, 2010a). Moreover, complaints bring morality to the interactional surface (Drew, 1998; Holt, 2012; Ruusuvuori et al., 2019). When complaining, speakers show their orientations toward what is right and wrong, reasonable and unreasonable. In the case of person-related complaints, speakers hold other people accountable for their behavior and make explicit how the reprehensible conduct breaches normative expectations of morality (Drew, 1988). Even in complaints about inanimate matters for which no one can be attributed responsibility – such as complaints about the weather – speakers engage in 'micro-interactional moral calibrations' (Stivers et al., 2011: 3) as they position themselves *vis-à-vis* a complainable and their interlocutors and thereby make relevant displays of alignment and afliation. Finally, complaints are typically considered conversational *activities* composed of more than one adjacency pair (see Robinson, 2013, on the notion of activity) rather than distinct actions (Heinemann & Traverso, 2009). Participants together negotiate the emergence and incremental development of complaints by picking up potential complainables, ratifying or denying their existence, and co-constructing their sequential development.

Complaining bears similarities with several other conversational phenomena, including criticism, accusations, talk about troubles, and gossip, which in turn may all be part of complaining (Edwards, 2005). Criticism and accusations are typical components of complaints about third parties, used by complainants to convey what is complaint-worthy about the third party's conduct. This does not mean that all criticism and accusations are parts of complaints, however. Troubles talk or troubles tellings (Jefferson, 1980, 1984a, 1984b, 1988; Jeferson & Lee, 1981) may also be part of complaints or work as self-standing activities on their own. Both complaints and troubles talk involve longer sequences of actions in which the speaker expresses a negative stance so as to recruit afliative responses from coparticipants. Drew (1998) argues that complaints differ from troubles talk precisely in the moral dimension of complaints, in the complainant's expression of grievance and orientation to unfairness. In troubles talk, issues of morality stay more implicit. Gossip, too, bears similarities, especially with complaints about non-present third parties. Like complaints, gossip normally involves afect-laden storytelling, and there is a tendency to publicly deny the engagement in gossip (Bergmann, 1987). Similar to complaining, gossip invokes issues of morality, and the activity serves important interpersonal purposes. Gossip does not necessarily involve the speaker's expression of grievance, however.

### **3.2 Structural organization of complaints**

Limited research has addressed the structural and sequential organization of indirect complaining. As pointed out by Laforest (2009), the kind of adjacency pair structure described by Scheglof (2005), consisting of a complaint proper (the main complaint formulation) and its response, sometimes applies for direct complaints but rarely for indirect ones. When such sequences do occur, they are typically preceded and succeeded by fne-grained interactional work that ought to be considered to be part of the same interactional activity (Traverso, 2009). Indirect complaints, therefore, tend to make up long sequences (Heinemann & Traverso, 2009), or what Sacks calls 'big packages' (1992, Vol. II: 354). The development of complaint initiations into full-fedged complaints and the subsequent organization of the activity are contingent on the coparticipants' contributions (Drew & Walker, 2009; Heinemann & Traverso, 2009; Ruusuvuori et al., 2019; Traverso, 2009). Moreover, observations about the structure of complaint stories (Günthner, 2000; Selting, 2010a, 2012) indicate that these, to a high degree, structurally resemble other types of stories. In the following sections, I focus primarily on the organization of complaints generally while providing some brief observations about the structure of complaint stories specifcally.

### *3.2.1 Complaint initiations*

 Existing research shows that speakers often move into complaining in a stepwise manner to test the grounds of the complaint before launching the activity fully (Heinemann & Traverso, 2009; Traverso, 2009; Ruusuvuori et al., 2019). Complaint initiations may be accepted, rejected, or merely disattended (Mandelbaum, 1991; Scheglof, 2005), and the

 initiator orients to these contingencies already from the start of (or even before) the sequence (Edwards, 2005; Traverso, 2009). Some diferences have been observed in the extent to which speakers orient to complaining as a delicate, or even dispreferred, activity in the initiation (cf. diferent observations in Traverso, 2009; Ruusuvuori et al., 2019; Günthner, 1997, 2000), which appears to relate to the interactional context and the participant framework – and possibly the type of complainable. In most cases, overt criticism or other strong expressions of negative stance are ofered only after some more subtle hints at the complaint-worthy situation (Pomerantz, 1986; Scheglof, 2005; Ruusuvuori et al., 2019).

Traverso's (2009) analysis of indirect complaints in French interactions between friends provides some evidence for how participants with well-established relationships initiate complaints in informal settings. The author observes that future complainants typically frst do some work to gauge whether the coparticipants will recognize and accept the complaint initiation and give their go-ahead signal for the initiator to continue. Even before the initiation, however, the initiator may somehow indicate a shift toward upcoming troubles talk, such as through a heavy sigh. While at the initiation stage the complaint is only a 'potential' one, according to Traverso (2009), it nevertheless resembles actual complaints at the surface level, in that it is uttered with linguistic, prosodic, and paraverbal features that are similar to those of full-fedged complaints. Consider Excerpt 3.1. The original is in French, but Traverso (2009) also ofers an English transcript version (forward slash indicates rising intonation).

*Excerpt 3.1* (Traverso, 2009: 2389, boldface added)

According to Traverso (2009), the initiation occurs in line 5, where C sighs, thereby foreshadowing the stance of the upcoming turn, and then shifts the topic slightly from the problems with the bathroom to the house as a whole. The sigh and the explicit expression of frustration and dissatisfaction make the turn hearable as a complaint-initiation, but whether it develops into an actual complaint depends on the recipient's response to the initiation.

As shown by Edwards (2005; see also Drew, 1998), in complaints between close friends or relatives, complainants regularly announce the way they are afected by the complaint-worthy situation before describing it, and this is also what seems to be happening in Excerpt 3.1. It is equally similar to what has been observed for complaint story openings, where speakers produce story prefaces to frame the upcoming story as a negative telling and hint at the expected afliative or sympathetic responses (Couper-Kuhlen, 2012; Selting, 2010a, 2012).

In contrast to the previously mentioned research on complaints among friends and relatives, Ruusuvuori et al.'s (2019) study of employees' complaints about coworkers in performance appraisal interviews in Denmark and Finland shows how people may go about when moving into complaining in more formal, professional settings. The authors observed highly subtle and careful ways in which the participants initiated complaints to display an orientation to these as delicate matters. The employees used various means to delay and mitigate complaint initiations, such as hesitation markers, restarts, and hedges. The managers also created opportunities for the employees to complain and facilitated complaint initiations by collaborating in the expression of shared afective stances and joint epistemic access to the complainable (Ruusuvuori et al., 2019).

Future complainants thus regularly hint at the negative valence of the upcoming talk and delay delicate actions such as negative criticism before fully launching complaints. Besides using non-linguistic resources such as non-lexical vocalizations that index negative a stance (see Section 3.3.2), speakers routinely use what Sacks (1992, Vol. I: 359) refers to as 'praisebut'-complaint initiations. By prefacing complaints with some kind of praise (which carries positive valence) and a 'but', what follows does not necessarily have to be clearly negatively formatted to be understood as a complaint. As I show in Section 5.2, this practice regularly occurs in my data too.

The use of self-praise in complaint initiations may be a way for speakers to manage what Edwards (2005) calls the 'subjective side' of complaining – namely, the speakers' portrayal of themselves. Self-praise can work to convey speakers' own reasonableness and legitimacy as complainants in the face of the complainable situation. Orientations to the delicacy of complaining are hence often dual in complaint initiations, seen in both the delay and mitigation of criticism and in speakers' positive portrayals of themselves, but the extent of such orientations seems to vary based on the interactional setting and participant framework.

#### *3.2.2 Complaint development*

 What follows the initiation of the complaint depends on coparticipants' responses. In Traverso's (2009) data on ordinary French interactions, coparticipants rarely ofer immediate afliation. More often, there is a negotiation about the complaint initiation after 'blatant' or 'subtle' disattending responses (cf. Mandelbaum, 1991), requiring the initiator to try several times to move into the complaint proper. If coparticipants ratify the complaint initiation, the activity proceeds to what Traverso (2009) calls the 'complaint development'. At this stage, the complainant works to attain further afliation from coparticipants by underlining why the situation is worth complaining about. In Traverso's collection, this is mostly done through repetitions, 'amplifcation', or tellings of stories or anecdotes.

Drew and Holt (1988) identify two distinct components of the formulation of complaints, which seem to correspond to the post-initiation phase described by Traverso (2009). On the one hand, participants report on the circumstantial details of the grievance, and on the other hand, they explicitly formulate, or name, the grievance through an idiomatic expression – the latter constituting the actual complaint (Drew & Holt, 1988):

*Excerpt 3.2* (Drew & Holt, 1988: 404)


In Emma's frst turn (lines 1–4), she details the circumstances of the problem; in her last turn (line 6), she formulates it with the help of the idiomatic expression 'two cents worth'. Together, Traverso's (2009) and Drew and Holt's (1988) observations indicate that the accounting practices involved in complaining, which are typically achieved through diferent types of reports or tellings, can be done either before (as reported by Drew & Holt, 1988) or after (as reported by Traverso, 2009) the main complaint formulation.

### *3.2.3 Recipient responses to complaints*

As previously mentioned, indirect complaints are designed to recruit afliative (and/or sympathetic) responses from the coparticipants. Afliation may be understood as a display of support and endorsement of another speaker's conveyed afective stance (Couper-Kuhlen, 2012; Stivers, 2008). In contrast to alignment, which has to do with *structural*  cooperation, afliation thus works on the *afective* side of cooperation (Stivers, 2008). When complainants express their afective negative stance toward an issue, they expect their coparticipants to show afective support. If they do not (immediately) get such support, they typically expand the sequence to provide further opportunities for the coparticipants to afliate (Couper-Kuhlen, 2012; Drew & Holt, 1988; Holt, 2012; Selting, 2010a, 2012; Traverso, 2009). Such sequence expansions support the idea of complaints as instantiating an interactional project

(cf. Levinson, 2013): The main purpose is not, for example, to tell a story about an unfortunate event but to obtain a particular type of recipient response.

 Documented practices and resources for displaying afliation in response to complaints include, among others: claims of understanding, displays of agreement, negative assessments, non-lexical vocalizations, and embodied stance expressions that match or upgrade the complainant's expressed stance, as well as second stories/complaints that are congruent with the frst complaint (Couper-Kuhlen, 2012; Selting, 2010a, 2012). Couper-Kuhlen (2012) shows that verbal afliative responses typically are delivered in a timely fashion and with prosodic matching or upgrading of the complainant's talk. Merely vocal displays of afliation, such as non-lexical vocalizations, are usually "reinforced verbally in following turns, suggesting that they may be perceived as momentary and feeting" (Couper-Kuhlen, 2012: 142) and not treated by participants as sufcient displays of afliation on their own. Particularly strong displays of afliation may lead to joint complaining (Drew & Walker, 2009; Heinemann, 2009; Laforest, 2009; Rääbis et al., 2019), that is, complaints in which several participants adopt the role of complainant expressing their dissatisfaction about a common complainable. Although research on joint complaining is still scarce, it appears that joint complaints may serve specifc bonding purposes between participants, allowing them to establish interactional 'coalitions' (Laforest, 2009) against the complained-about third party or issue.

Complaints are not always responded to with afliation, however. Uncooperative complaint responses have been discussed in terms of 'diffusing', 'disattending', 'preempting', or 'rejecting' the complaint underway (Holt, 2012; Mandelbaum, 1991; Ruusuvuori & Lindfors, 2009; Scheglof, 2005). Given the preference for cooperation in conversation (Stivers, 2008), participants develop methods for accomplishing noncooperation that nevertheless permit them to maintain social solidarity with the complainant. Mandelbaum (1991) documents how complaint recipients accomplish 'subtle disattending' by attending to and developing factual points in the telling that lead away from the actual complainable (see also Ruusuvuori & Lindfors, 2009). Holt (2012) moreover shows that laughter can avert the development of a complaint without jeopardizing social solidarity.

#### *3.2.4 Complaint closings*

Even less is known about complaint closings than about initiations. According to Traverso (2009), one of the diferences between complaining and storytelling is that complaints lack the kind of early projection of the overall structure that storytellings typically have. This converges with Jeferson's (1988) observations about troubles talk, in which she

 proposes a 'sequence candidate' of sequentially ordered elements that reoccur to a varying extent rather than the interactional work-up to a projectable climax. In Jeferson's data, the closing of troubles talk often co-occurred with conversational closing, making troubles talk the last topic after which no other topics should be introduced. Traverso (2009) did not observe complaining as the last topic of conversations, but closings sometimes coincided with marked changes in the situation, such as the arrival or departure of a participant. In both Jeferson's (1988) and Traverso's (2009) collections, speakers show strong orientations toward the relevance of receiving afliative or sympathetic responses before moving to a close.

In terms of interactional practices for moving toward sequence closure, Jeferson (1988) documents the use of 'optimistic projections', invocation of the status quo, and making light of the trouble. Another resource complainants deploy to close down complaints is prosody. Ogden (2010) notes that speakers produce closing-implicative turns (typically consisting of summary assessments, idiomatic expressions, or lexical recycling of prior talk) with lower pitch onset than in prior turns, with relatively quiet and lax voice, and with a narrow pitch span. These prosodic features difer from those used with turns designed to invite afliation and thus contribute to the various cues informing recipients about expected responses. As mentioned earlier, laughter, typically used together with other 'topically disengaged' responses that convey some level of afliation but do not invite further elaboration, seems to work as a particularly useful resource when coparticipants set out to close a complaint and initiate topic shift (Holt, 2012).

Regarding complaint stories, the literature suggests that these rely on similar principles for sequence closure as complaints generally. If, after the delivery of the story climax, tellers do not receive sufciently afliative responses, they normally expand the sequence by further insisting on the complaint-worthiness of the reported situation (Couper-Kuhlen, 2012; Selting, 2010a, 2012). Complaint stories lend themselves particularly to the initiation of second complaints (Selting, 2012). Similar to what was observed by Jeferson and Lee (1981) for advice given prematurely in response to troubles tellings, second stories initiated 'too early' may be oriented to as non-afliative by the frst complainant (Selting, 2012) as the frst complaint is interrupted before the complainant had a chance to fully develop it.

The research presented in this section paints a complex picture of complaint sequence structure. Most available evidence converges around a characterization of complaints as highly contingent activities, the structure and sequential development of which depend strongly on coparticipants' contributions and contextual factors. I further address the structure of complaints in Section 5.1, although from an L2 perspective.

#### **3.3 Interactional resources for complaining**

One of the main jobs of complainants is to underline the unreasonable and egregious character of the reported situation to show that the complainable is worthy of complaining. At the same time, complainants often do interactional work to handle the 'subjective side' of complaints, to portray themselves as credible and legitimate complainants (Edwards, 2005). The literature highlights several actions and activities complainants typically use to manage both the 'objective' and the 'subjective' side of complaints, such as negative assessments, precise descriptions of complaint-worthy conduct, and storytellings. Within and besides these actions and activities, complainants deploy a range of linguistic, rhetorical, prosodic, and embodied resources to show a negative stance and construct 'complaintworthiness'. I refer to these as interactional *resources* for complaining in a general sense, despite their diferent statuses as actions, practices, or resources according to more traditional CA terminology (see, e.g., Clift, 2016; Heritage, 2010).

#### *3.3.1 Verbal and linguistic resources*

Negative assessments2 are a key resource for formulating complaints and expressing the complainant's stance toward the complainable. Drew (1998: 310) refers to these as "overt expressions of moral indignation", by which speakers morally condemn the behavior of the other, although sometimes through frst-person assessments (e.g., 'I was so upset'). Negative assessments are also used by coparticipants to respond afliatively to other speakers' complaints and to co-construct joint complaints. Although no study has systematically investigated the role of negative assessments in complaint sequences, existing evidence suggests that these typically are of low-grade nature or take rather implicit forms at the beginning of complaints (Edwards, 2005; Ruusuvuori et al., 2019). High-grade assessments are used later as part of an escalation of the accusations against the third party or complainable (Drew, 1998; Pomerantz, 1986; Rääbis et al., 2019), such as in summary assessments. This converges with participants' orientation to complaints as delicate matters that need to be introduced in a stepwise fashion. As pointed out earlier, even neutrally formatted descriptions may serve the purpose of a negative assessment by virtue of their placement after a positively valenced element and a contrastive marker. Excerpt 3.3 illustrates both the stepwise escalation of negative assessments and the use of contrastive formulations. The excerpt comes from a performance appraisal interview between a manager (M) and an employee (E) at a Danish company (Ruusuvuori et al., 2019). M has just asked E how it is going with 'the other departments'.

*Excerpt 3.3* (Ruusuvuori et al., 2019: 50; gaze indications omitted)

Upon M's request, E assesses the situation as 'generally good' (line 4). This assessment, while positive, also implies that something is *not* good. According to Ruusuvuori et al. (2019), the manager's acknowledgment of this assessment, his repeat (line 5), displays an expectation that E should expand on the topic. E does so in the ensuing talk, as he claims that there is something that has been going on with the department Blommeballe for months and years (lines 8–9). In lines 11–12, he ofers a mildly positive assessment of the situation, that it would be wrong to call it 'bad'. This assessment, too, indexes a negative undertone, and the more clearly negative assessment follows the contrastive conjunction 'but': The relationship is artifcial. The complaint proper is formulated only after further elaboration and takes the form of an idiomatic expression with clear negative connotations (Ruusuvuori et al., 2019).

Second assessments often fgure in complaints when recipients express their afliation or sympathy with the complainant (Couper-Kuhlen, 2012; Drew, 1998; Selting, 2010a, 2012) or help construct a joint complaint (Rääbis et al., 2019; Drew & Walker, 2009; Heinemann, 2009). Afliative second assessments upgrade the frst assessment (Pomerantz, 1984) and typically contribute to the escalation of the afective stance (Rääbis et al., 2019). Excerpt 3.4 illustrates such a case in which three Estonian teenage girls complain about their physics teacher.

*Excerpt 3.4* (Rääbis et al., 2019: 29)

In response to H's assertion that she needs to study for a test (line 4), R deploys a fgurative expression that negatively assesses her teacher as 'sick' (line 5). J responds to this with the strongly upgraded second assessment of the teacher as an 'old moron damned runt' (line 6) and subsequently the ironic 'good teacher' (line 8). As a further escalation of the complaint, H asserts that she would 'kill' the teacher (line 9; see detailed analysis by Rääbis et al., 2019). The participants' high-grade assessments rely on another common resource deployed for constructing complaint-worthiness – namely, extreme-case formulations (Pomerantz, 1986). Extreme-case formulations are assessments or descriptions that contain extreme terms such as 'brand new', 'everyone', and 'all day Sunday' (Pomerantz, 1986). In the context of complaints, they may serve as a way "to defend against or to counter challenges to the legitimacy of complaints" (Pomerantz, 1986: 219).

Two related resources are generalizations and idiomatic expressions. Mandelbaum (1991: 120) notes that a statement such as 'he always gets mixed up' may be used to support the recurrent nature of the complainable and thus the legitimacy of the complaint. Idiomatic expressions sometimes serve to create a similar, generalizing efect, as these typically appeal to general, public wisdom. Drew and Holt (1988) note that complainants recurrently deploy idiomatic expressions in a similar manner as summary assessments to formulate the gist of the complaint. The fgurative nature of idiomatic expressions makes them a particularly powerful resource for complaining: In contrast to descriptions with concrete facts that may be empirically tested and falsifed, idioms "have a certain resistance to being tested or challenged on the empirical facts of the matter" (Drew & Holt, 1988: 406). This is why they often occur after insufciently afliative responses to a more factual description as a means to increase chances of obtaining afliative displays.

### 34 *Complaining in L1 interaction*

As mentioned earlier, other central tasks of complainants are to describe and account for the way in which the complainable situation constitutes a transgression. Through descriptive detailing, complainants construct the conduct of the other as morally reprehensible and their own conduct as morally defensible (Drew, 1998), thereby justifying the complaint. Storytellings and other types of tellings are key resources for accomplishing such descriptive work. In some cases, tellings are used as 'exemplary stories' (Günthner, 1995), providing precise examples of the complainable. In other cases, complaints emerge progressively *through* storytellings (Selting, 2010a, 2012). Either way, the telling details the circumstances of the complainable situation to substantiate the complaint.

Recurrent features of storytellings are reported speech and (re)enactments, which serve particular interactional purposes in complaints. There are many alternative terms for referring to reports of talk and conduct, such as *reported talk/speech/discourse/thought*, *constructed dialogue*, *conversational quoting*, *represented talk*, *enactments*, and *reenactments*  (Clark & Gerrig, 1990; Clift & Holt, 2007; Prior, 2015); I henceforth use the terms *reported speech* and *reenactments*. The term *reenactment*  typically refers to reports of real or fctive embodied actions (Sidnell, 2006; see also Holt, 2007, for *enactment*), although studies of 'reported speech' regularly encompass reported non-verbal conduct. A basic distinction made in the literature is between *direct-reported speech* (DRS) and *indirect-reported speech* (IRS). When speakers deploy DRS, they claim to adopt the voice of the quoted person and reproduce his/her exact words; in IRS, they integrate the reported talk into their own discourse by adapting it to the circumstances of the here-and-now. The distinction between the two is not always clear-cut, however (Clift & Holt, 2007; see also Haakana, 2007, on reported thought). As discussed by Clift and Holt (2007), marked prosody is typically part of all reported speech and may both serve to distinguish between reported utterances and the speaker's own voice and to convey an evaluation of the reported talk.

Specifcally, DRS and reenactments are useful resources for speakers to scene-set tellings in an animated and witnessable way and for implicitly assessing reported events. When quoting or reenacting what another person has said or done, a speaker claims to provide an accurate, objective account of such conduct (Drew, 1998). At the same time, the speaker portrays the other's conduct through the eyes of the speaker himself/herself, thereby ofering a highly subjective picture of the situation. Complainants regularly deploy DRS and reenactments to show, rather than merely describe, in what way the conduct of the third party constitutes a complaint-worthy transgression and how this contrasts with their own reasonable conduct (Berger & Pekarek Doehler, 2015; Drew, 1998; Günthner, 1995, 1997, 2000; Holt, 1996, 2000; Kasper & Prior, 2015; Selting, 2010a, 2012). Through marked prosody, lexical choices, and other interactional means, the complainant illustrates the ofensiveness and unpleasantness of the third party (Drew, 1998). By instead deploying

more neutral prosodic and paralinguistic resources when reporting on their own conduct, speakers portray themselves in a more positive light (Berger & Pekarek Doehler, 2015). Reported speech and reenactments thus increase the 'witnessability' of the complaint, which makes these resources particularly suitable for marking the climax of a complaint story (Drew, 1998; Holt, 2000). After the climax, there is not necessarily a need for the complainant to explicitly evaluate the reprehensible behavior, instead he/she leaves the reported conduct to 'speak for itself'.

DRS and reenactments can also be efective means to engage coparticipants in the telling and recruit afliative responses (Holt, 2000; Sidnell, 2006). Excerpt 3.5 provides an example in which speaker D's direct-reported dialogue (lines 14–17) immediately receives an afliative response from K:

*Excerpt 3.5* (Holt, 2000: 446; boldface added)


K's response in line 18 consists of a negative assessment of the third party, which shows afliation with D's complaint. As also invoked by Holt (2000), such recipient display of afliation often results in a rapidly produced second assessment by the speaker himself/herself (line 19), which confrms the recipient's interpretation of the reported event and allows the participants to exchange mutual displays of alignment and afliation.

#### *3.3.2 Prosodic and other non-linguistic resources*

As already highlighted in the discussion of complaint closings and in relation to reported speech and reenactments, participants use marked prosody and certain non-linguistic resources such as non-lexical vocalizations and laughter to both construct and respond to complaints (Couper-Kuhlen, 2012; Drew, 1998; Drew & Holt, 1988; Günthner, 1995, 1997, 2000; Holt, 1996, 2000; Kasper & Prior, 2015; Ogden, 2010; Selting, 2010a, 2012). Because a large proportion of the studies on complaining has relied on telephone data, less is known about speakers' use of embodied resources like gestures and gaze shifts.

### 36 *Complaining in L1 interaction*

Complainants recurrently modulate their prosodic3 delivery through shifts in pitch, volume, and tempo to display their afective stances toward the complainable situation, to portray the characters reported on in their tellings (see the previous discussion on DRS), and to animate their complaints (Couper-Kuhlen, 2012; Günthner, 1997, 2000; Selting, 2010a, 2012). While not all complaints involve strong displays of afect (Ruusuvuori et al., 2019), many do (Rääbis et al., 2019; Selting, 2012). Selting (2010a, 2012) has identifed a range of prosodic features that complainants assemble in context-sensitive ways as 'bundles of co-occurring cues' to display their emotive involvement and particular afective stances such as annoyance and indignation. These include accentuation on key terms, dense accentuation, marked changes in pitch and volume, distinct contrasts in contours, marked rhythm, syllable lengthening, laughter, and laugh particles. In the enactment of diferent voices in DRS, speakers have also been observed to use partly conventionalized stereotypic features such as falsetto voice quality and hyper- and hypoarticulation (Selting, 2012). In addition, as mentioned in Section 3.2.4, prosody contributes to the diferent cues complainants use to index whether their complaint turns invite afliation or are designed to close down the sequence (Ogden, 2010). Last, Edwards (2005) demonstrates that Jeferson's (1984b) observations about laughter as a means to show troubles-resistance also holds for complaints: Complainants may use laughter to convey that they remain in good spirits despite the unfortunate circumstances and thereby portray themselves in good light, as someone who can manage the difcult situation rather than being a dispositional moaner.

As for responses to complaints, Couper-Kuhlen (2012) has found that recipients often prosodically match or upgrade the complainant's prior turns to show afliation, whereas prosodic downgrading typically indexes non-afliation. This was particularly true in the case of responses in the form of stand-alone non-lexical vocalizations, although – as also documented by Couper-Kuhlen (2012) – non-lexical vocalizations were typically accompanied by subsequent verbal resources (such as verbal assessments) and hence not treated as sufcient displays of afliation on their own (see the previous discussion). Outside the literature on complaining, research on (non-lexical) vocalizations (Keevallik & Ogden, 2020) has shown that these may be used both as assessments on their own and as displays of stance and afect more generally (Baldauf-Quilliatre, 2016; Couper-Kuhlen, 2009; Goodwin & Goodwin, 2000; Hoey, 2014; Reber, 2012; Wiggins, 2013; see also Gofman's notion of *response cries*, Gofman, 1981). Several studies document the use of nonlexical vocalizations in turn-initial position within negative assessment turns, in which the sound object forecasts the negative valence of the upcoming turn (Baldauf-Quilliatre, 2016; Hoey, 2014; Traverso, 2009).

Only a few studies on complaining address how speakers' gestures and gazes contribute to the accomplishment of the activity. As with prosodic features and non-lexical vocalizations, embodied displays become meaningful communicative cues in their interactional context and in concert with other resources. Analyzing frst and second complaint stories, Selting (2010a, 2012) documents the use of the following visual displays by complainants and complaint recipients: facial expressions (raised eyebrows, frowning, smirking, smiling versus unsmiling), gaze shifts, head movements (head nods/shakes, head postures), hand movements (knocking/pointing with fnger, slashing with arm), and conventionalized postures of 'helplessness' and 'no understanding', such as raised shoulders and eyebrows and spread out arms. Particularly in the context of DRS and reenactments, speakers have been observed to combine multiple semiotic resources to display their stances and heightened afective involvement (Günthner, 1997, 2000; Selting, 2010a, 2012; Sidnell, 2006).

Also relevant for complaining, research on assessments has documented the highly multimodal nature of assessment turns and activities. Speakers deploy prosody and bodily conduct to foreshadow incipient assessments, modulate the strength of verbal assessment segments, mobilize recipient assessments, and more generally display their afective involvement in the assessment activity (Goodwin, 1980; Goodwin & Goodwin, 1992; Haddington, 2006; Lindström & Mondada, 2009). Stance displays in the form of facial expressions can help stretch the temporal boundaries of assessments of stories and topics (Ruusuvuori & Peräkylä, 2009), and turn-initial frowns, in particular, have been observed to foreshadow negative assessments (Kaukomaa et al., 2014). Chapter 6, which examines how L2 speakers make use of both linguistic and non-linguistic resources in constructing and responding to complaints, ofers a small contribution to the still limited research on embodied conduct in negative assessments (see also Skogmyr Marian, 2021a, on embodied completions of verbally incomplete assessments).

#### **3.4 Complaining in L2 interaction**

This book presents the frst longitudinal investigation on complaining within research on the development of L2 IC. The SLA literature more generally is similarly scarce regarding indirect complaints (for research on direct complaints within the feld of interlanguage pragmatics, see Olshtain & Weinbach, 1993; Wijayanto et al., 2017; Trosborg, 1995). A few non-CA studies on indirect L2 complaining and CA studies on related issues that have some bearing on the analysis of L2 complaints are nevertheless relevant to my work.

 The topically most relevant research is Boxer's (1993) ethnographic study of complaints as speech acts among American English L1 students and faculty and a group of Japanese ESL students at an American university campus. Based on (primarily) audio-recorded complaints, Boxer observed that complaints were much more frequent in L1 interactions than in L1-L2 speaker interactions. In L1-L2 interactions, almost 70% of the complaint sequences were initiated by L1 speakers. Complaints formulated by L2 speakers were most often elicited through a question from the L1 speakers, and many of the complaint initiations by the L2 speakers were problematic in that they were not recognizable as such to the L1 speakers. In addition, over half of the responses by L2 speakers to L1 complaint initiations were 'zero-answers' or topic shifts (in L1 conversations, these responses were very rare), meaning that the L2 speakers failed to respond in an afliative way that allowed for the development of the complaint. Boxer's (1993) fndings indicate that complaining is an activity that L2 speakers need to learn how to accomplish in the L2, both in terms of initiating complaints and responding to others' initiations.

 As regards CA research, Berger and Fasel Lauzon's (2016) study of the L2 French *au pair* Julie and her host family provides some insights into the kind of resources L2 French speakers deploy to respond to other speakers' displays of afect and what consequences this has for the management of interpersonal relationships. The authors report that Julie recurrently showed understanding of and afliation with the host mother's afective stance through second assessments, afect-laden sound objects, recycling of the mother's prior turns, prosodic matching, and laughter. Such afliative responses led to mutual displays of emotional solidarity that contributed to the establishment and maintenance of the social relationship between the participants. According to Berger and Fasel Lauzon (2016), the participants did not orient ostensibly to Julie's status as an L2 speaker in these contexts, and Julie's responses to the host mother's afect-laden talk testify to a high level of L2 IC. Interestingly, however, although not investigated specifcally by the authors, several of Julie's second assessments and displays of afliation were delayed, requiring some kind of repair (see Berger & Fasel Lauzon, 2016, Excerpt 3.1: 91; Excerpt 3.3a: 93–94; Excerpt 3.5a: 99–100). Consider Excerpt 3.6:

*Excerpt 3.6* (Berger & Fasel Lauzon, 2016: 91, boldface added)


In response to Mom's assessment of the missing element in the grammar book as *tellement bizarre* ('so weird', lines 1–3), Julie frst responds with the acknowledgment token *<ouais>* ('yeah', line 4). The silence in line 5 indicates potential trouble with Julie's response, which Julie repairs by ofering a second assessment (*oui c'est bizarre*, 'yes it's weird', line 6) that shows understanding of and afliates with Mom's afective stance. The several instances of delayed and repaired responses in the presented data indicate that it may be difcult even for quite advanced L2 speakers (which Julie was) to produce timely second assessments and stance displays in response to other speakers' afect-laden talk, similar to what Boxer (1993) noted about ESL speakers' responses to complaints. My study develops these observations by investigating more in detail L2 speakers' practices for assessing and responding to assessments and how such practices develop over time (see Section 6.1).

### **3.5 Cumulative evidence about complaining and research gaps**

The combined empirical evidence about complaining in interaction shows the complexity of this activity. An important, basic distinction to make when referring to complaining is the one between so-called direct and indirect complaints. Speakers use direct and indirect complaints for distinctly diferent social purposes, and these activities take diferent interactional forms. Only a few studies have investigated joint complaints (Drew & Walker, 2009; Heinemann, 2009; Laforest, 2009; Rääbis et al., 2019). While no research has directly compared complaint sequences across languages or cultures, the reviewed literature represents a range of diferent languages, primarily Indo-European but also other (European) languages, such as Danish (Heinemann, 2009; Ruusuvuori et al., 2019), English (e.g., Drew, 1998; Drew & Holt, 1988; Edwards, 2005; Mandelbaum, 1991), Estonian (Rääbis et al., 2019), Finnish (Ruusuvuori & Lindfors, 2009; Ruusuvuori et al., 2019), French (Laforest, 2009; Traverso, 2009), and German (Couper-Kuhlen, 2012; Günthner, 1995, 1997, 2000; Selting, 2010a, 2012). The precise linguistic resources people mobilize to complain in these various languages naturally difer, but the basic building blocks and practices that go into complaining do not seem to difer much across European languages.

Cumulatively, the literature on indirect complaining has revealed the following:


# 40 *Complaining in L1 interaction*


In addition, existing research provides some evidence about the following aspects of complaining:


The literature on complaining in interaction leaves room for further investigations of several dimensions of this phenomenon in both L1 and L2 interactions:


There is clearly much left to uncover about indirect complaining in interaction, specifcally as it concerns L2 speakers' practices. The empirical studies presented in Chapters 5–7 contribute to these lines of research.

# **Notes**


# **4 Investigating change longitudinally**  Methodological concerns and data

Longitudinal CA, and especially research interested in L2 development, comes with considerable epistemological and practical challenges. This chapter addresses these challenges by outlining and problematizing the methodological procedures and data used in the study. I frst review the foundational principles of CA and discuss the implications of these when used in the framework of SLA studies (Section 4.1). I then describe the presumptions underlying the longitudinal research design of the study and discuss the conceptual and practical challenges involved in conducting developmental studies of L2 IC (Section 4.2). In the third section of the chapter, I present the empirical material used in the investigation. I outline the steps involved in collecting data and establishing collections and describe the participants and the inventory of complaint sequences (Section 4.3).

### **4.1 EMCA and CA-SLA**

 The methodological approach of the study is grounded in ethnomethodological CA (EMCA). As a branch of sociology, ethnomethodology investigates the interactional 'methods' (Garfnkel, 1967) by which ordinary people make sense of their social world. CA has developed in part as an ofspring of ethnomethodology, focusing on the sensemaking practices people deploy as they engage in non-elicited, naturally occurring talk, or *talk-in-interaction*. CA aims to discover and describe people's orderly, recurrent, and systematic practices for managing different aspects of social interaction, such as turn-taking (Sacks et al., 1974), sequence organization (Scheglof, 2007), and repair (Scheglof et al., 1977). CA, therefore, documents the kind of everyday competence people rely on when engaging in ordinary conversation (Garfnkel & Sacks, 1970; Heritage, 1984b; Psathas, 1990) – in other words, their IC. While some researchers deploy CA merely as an analytic tool within other epistemological frameworks (e.g., ethnography of communication, language socialization), EMCA adheres strongly to the principles of ethnomethodology as an inductive, data-driven, and emic discipline that

stays indiferent to *a priori* theory (see Clift, 2016; and Sidnell, 2010, for comprehensive introductions to CA).

Conversation analysts draw on a specifc set of analytical and methodological procedures for analyzing social interaction (Psathas, 1990). Observation lies at the heart of this method: "We will be using observation as a basis for theorizing. Thus we can start with things that are not currently imaginable, by showing that they happened" (Sacks, 1984: 25).

 With the help of detailed transcripts of recorded talk (see Jeferson, 2004), the researcher analyzes the unfolding interaction sequentially, that is, turn by turn, to document how participants accomplish social actions and reach intersubjectivity. Initially, CA was concerned mainly with people's verbal practices (but see, e.g., Goodwin, 1979, 1980, for early multimodal CA). The feld's 'embodied turn' (Nevile, 2015) around the year 2000 occurred through a marked rise of empirical studies on embodiment and has contributed to a more holistic understanding of how people interact. Many conversation analysts now transcribe (see, e.g., Mondada, 2019) and analyze a range of diferent semiotic resources that participants deploy to accomplish social actions, including embodied conduct (gaze, gestures, postures, facial expressions, etc.) and cultural artifacts. Whereas some CA research focuses on participants' practices in unique interactional encounters, most studies rely on collections of cases for purposes of grasping the recurrent, systematic nature of members' methods (Scheglof, 1993). The cumulative fndings of these studies shed light on the generic organizational principles of social interaction (Scheglof, 2007; but see, e.g., Lynch & Macbeth, 2016; Lynch & Wong, 2016, for more critical stances toward collection-based CA). Empirical observations about real-life encounters thus form the basis for theorizing about the 'machinery' underlying human social conduct (Sacks, 1984).

As outlined in Section 2.1, the feld of CA-SLA draws on the methodological foundations of CA to investigate L2 interaction, learning, and development. An important and contentious issue within this feld concerns the ability of CA to address issues of learning (Kasper, 2006). Whereas most SLA research traditionally has focused on theory-building and the testing of L2 acquisitional models (see, e.g., VanPatten & Williams, 2015), CA's data-driven and emic approach does not ofer a theoretical framework that explains learning. This is why some scholars combine CA with sociocultural theory, language socialization, or other pre-established theories to interpret fndings about L2 practices in terms of learning. Atkinson's collection of 'alternative approaches to SLA' (2011a) neatly exemplifes some of these diferent approaches, including more ethnomethodologically oriented research like the present study that does not rely on any *a priori* theory of L2 learning. So how can such EMCAgrounded research investigate L2 learning and contribute to the feld of SLA while staying true to the emic, participant-relevant perspective? What kind of evidence of L2 learning can CA ofer?

### 44 *Investigating change longitudinally*

Some CA-SLA studies examine L2 speakers' practices for 'doing learning' – that is, observably orienting to learning processes (Sahlström, 2011). Sometimes this work adopts a (micro-)longitudinal perspective that documents change over time in participants' *learning behaviors*  (Markee, 2008; Kunitz & Skogmyr Marian, 2017). These studies thus provide emic evidence for how learning takes place – the process of L2 learning – sometimes across a few diferent social encounters. They do not address long-term change in interactional practices, however, which is the main focus of the present investigation.

If one is interested in the longitudinal development of L2 IC, and hence the outcome or 'product' of learning, the emic perspective is more problematic. How can we interpret change over time in interactional practices as a development from the participants' own perspective? Speakers rarely orient to diferences in their interactional conduct at one point in time *vis-à-vis* earlier occasions. And if participants interact with diferent coparticipants, to whom do we attribute the change? What is the link between change in social practices and learning? As discussed by Deppermann and Pekarek Doehler (2021) and Wagner et al. (2018), participants' orientations to their coparticipants' practices as more or less competent can provide an emic perspective on development in the sense of 'positive change'. Competent conduct is recognizable to coparticipants (and thus, to researchers) for what it is designed to do and hence "provides no grounds for comment or repair" (Wagner et al., 2018: 27; see also Garfnkel, 1967; Garfnkel & Sacks, 1970). Empirical studies on the development of L2 IC often document increased recognizability and local efcaciousness of interactional practices over time (see Chapter 2). Such change can thus be considered emic evidence for development, which in turn may be seen as refecting the result of learning. A problem with this perspective is, however, the general permissiveness and 'let it pass' tendency (Garfnkel, 1967: 3) observed in interactions with L2 speakers, whereby recipients typically show a high threshold of acceptance toward interactional conduct that would be treated as problematic in L1 talk (cf. Firth, 1996, on *lingua franca* interactions). Although there is no clear-cut solution to this difculty, in Chapter 8 I ofer some refections on these issues in light of my own data and empirical fndings. At this point, it is worth underlining that the present study adheres to the foundational methodological procedures of EMCA by deploying sequential analysis of talk-in-interaction without recourse to exogenous learning theory. I also refrain from discussing learning in the context of my empirical analyses and instead address the implications of the fndings for L2 learning in Chapter 8 and, to a limited extent, in the discussions at the end of each analytical chapter. This way, I ensure *emic validation* of my fndings (see Clayman & Heritage, 2021) and, only as a second step, discuss how these can inform SLA.

#### **4.2 Longitudinal CA: research design and challenges**

Research on the development of L2 IC is fundamentally comparative: In order to say something about change over time, the analysis has to document diferences in practices or interactional resources across chronologically ordered (collections of) cases (Wagner et al., 2018). Most studies on the development of L2 IC adopt a longitudinal research design. These studies focus on one or a few target participants and compare their practices for accomplishing particular social actions or activities at diferent points in time. For example, König (2019) tracked three French L2 speakers' practices for opening, shifting, and closing topics at the beginning, middle, and end of a period of six to ten months. The granularity of the analysis varies across studies: Some studies compare practices at two different points in time, and others analyze practices on multiple occasions at regular intervals. Instead of using a longitudinal research design, certain studies on the development of L2 IC draw on a cross-sectional design to compare the interactional conduct of groups of participants at diferent profciency levels (e.g., Al-Gahtani & Roever, 2012). Although such an approach has some practical advantages, observed diferences across groups are merely indicative of a potential longitudinal trajectory since they do not represent actual change.

The present study relies primarily on a longitudinal research design by using longitudinal data from several participants at diferent profciency levels and by comparing interactional practices both within participants over time and between participants. To enhance the robustness of the fndings (see the following section), I have also included data from additional participants who stayed at the same profciency level over time or for whom a longitudinal analysis was not possible. This research design maintains the benefts of longitudinal analysis while also allowing for observations past the individual level.

The comparative approach involves several methodological challenges. A basic requirement when tracing changes in the accomplishment of an action or activity is to show that the action or activity remains the same while the participants' practices for accomplishing it change (Koschmann, 2013). Wagner et al. (2018) highlight three methodological difculties associated with this premise: (1) warranting comparability, (2) building collections, and (3) providing robust evidence for longitudinal change. The frst of these concerns the interactional context of the cases. To argue that a change in practices for accomplishing an action or activity has taken place, one must ensure that observed diferences are due to a change in practices and not a change in the activity itself or in the interactional context. Therefore, the analyst needs to use cases where the speech exchange system, the (type of) participants, the activity type, and the sequential environment are comparable. The second challenge concerns the procedures for establishing collections that allow for longitudinal comparison. Instead of drawing on one single homogenous collection, one must build several collections of the studied phenomenon from diferent points in time. This requires designing the data collection in a way that provides sufcient comparable interactional data and cases over an extended time period. Third, Wagner et al. (2018) invoke the challenge of providing robust evidence for the documented change. To prove that a longitudinal development has taken place, one needs to show *systematic* change – that a change in practice is not just a one-time happenstance. Some quantifcation may therefore be useful considering that we know that "the proportion of occurrence of an interactional phenomenon changes as part of people's (increased) adaptation to the local circumstantial details of the ongoing interactions" (Wagner et al., 2018: 25). Although not necessary, desirable, or possible in all studies, quantifcation thus helps showing systematicity and routinization of interactional practices over time. As discussed by Clayman and Heritage (2021), quantitative analysis can also mitigate the risk of confrmation bias in the selective presentation of data excerpts in the qualitative analysis. Importantly, quantifcation does not replace single case analysis but instead "is built on its back" (Scheglof, 1993: 102). As shown in the next section, I have collected the data and established collections in ways that take into account the methodological challenges discussed so far.

### **4.3 This study: empirical material**

#### *4.3.1 Setting and participants*

The primary empirical material of the study consists of video recordings of L2 French speakers participating in French-language conversation groups that took place between October 2016 and June 2018. The 'conversation circles', as these were called, ofered an informal conversation-practice activity for students attending a French-language institute at a university in the French-speaking part of Switzerland. The participants were, with the exception of one person, university students enrolled in either an L2 French support course or in a more comprehensive L2 French language program. The support courses targeted four diferent profciency levels of the CEFR (Council of Europe, 2020) scale: A1, A2–B1, B2, and B2–C1. The more comprehensive language courses were intensive programs with B1 and B2 prerequisite levels, respectively. The conversation circle provided an optional, complementary activity allowing students to practice their spoken French outside the classroom.

The participants were between 21 and 42 years old and came from 17 diferent countries. They were placed in groups of three to four people with similar estimated profciency level and based on their study schedules. Five participants, who attended four diferent conversation groups, were chosen as focal participants of the study (see Section 4.3.5). The meetings took place in a small university cafeteria every two weeks during the semester, with longer breaks during academic holidays. The participants were free to speak about whatever they wished during these meetings. Meetings lasted 30–60 minutes.

In line with standard research ethics guidelines, the participants were informed about the general aims of the research, its methods, data handling, and the fact that participation was voluntary and may be terminated at any time. All participants gave their written consent to participate in the research.

#### *4.3.2 Recordings and supplementary material*

The meetings were recorded with two video cameras and an external audio recording device. The recording equipment was positioned as discreetly as possible in the cafeteria to minimize intrusion on the interaction while nevertheless capturing the interaction from diferent angles to permit detailed multimodal analysis. Figure 4.1 illustrates the interactional setting.

*Figure 4.1* Interactional setting seen from two angles.

A total of 63.5 hours of recordings were included in the study. These came from four conversation groups, referred to as Lundi ('Monday'), Mercredi-1 ('Wednesday-1'), Mercredi-2 ('Wednesday-2'), and Jeudi ('Thursday') based on the day of the recording. Besides these recordings, I used an online background questionnaire to collect complementary information about the participants at the start of their participation. Unless the participants already had a recent certifcate indicating their level of French, they were asked to complete the online profciency test Dialang, which is based on the CEFR scale.1

### *4.3.3 Transcription and anonymization*

The interactions were transcribed according to standard Jefersonian transcription conventions (Jeferson, 2004). I also used Mondada's (2019) conventions for transcription of multimodal conduct, with some modifcations (see Appendix). Descriptions of embodied conduct are only included to the level of detail necessary for the points made in the analysis. Framegrabs (shortened FG in transcripts) show particularly important embodied conduct. All framegrabs, including participants' faces, have been anonymized (blurred). Participants' names, references to places, and other sensitive information have also been replaced by pseudonyms and fctive names in the transcripts. The town in which the recordings took place is consistently referred to as 'Launève'.

English translations of French talk appear in italics below the original line. I have attempted to fnd a balance between providing as idiomatic translations as possible and maintaining features of the original turn design. I have typically not translated non-lexical vocalizations (*eu:h*, *pf, o:h*, etc.) or response tokens such as *mm-hm*, *uh-huh*, or *okay* unless these have conventional meanings in English that are diferent from French. In the case of non-standard or unintelligible talk, I have ofered my best guesses of target items.

For identifcation purposes, excerpt headers include information about the conversation circle group (abbreviated as 'Lun', 'Mer1', 'Mer2', 'Jeu'), the date of the recording, and a keyword/phrase from the sequence.

### *4.3.4 Determining the analytical focus and establishing collections*

The decision to focus the analysis on indirect complaints was not motivated by any particular interest in complaining. In accordance with EMCA's principle of unmotivated looking (Psathas, 1990; Sacks, 1984), I did not determine the analytical focus of the study before collecting the data. The decision to investigate complaining was thus data-driven and emerged as I started working with the recordings. It was based on a wish to focus on an analytical object that was salient in the conversational data at hand, which had not yet been examined in research on the development of L2 IC, and which aforded the opportunity to concurrently examine several aspects of the development of IC (action formation, sequence organization, etc.). The initial screening of the recordings showed that complaining occurred relatively frequently in the interactions and therefore was representative of the type of activities in which the participants regularly engaged in the conversation circle. It is thus an analytical object that was sufciently frequent to allow for longitudinal analysis and which no prior study had investigated from a CA-SLA perspective.

The collections were established based on the core features of complaining highlighted in prior literature (see Section 3.1). After broadly collecting potential cases, I narrowed my focus to interactional episodes exhibiting the following characteristics:

(1) They include *expressions of negative stance* about an issue that, according to the speaker, has afected him/her personally in an unfair or unreasonable manner. This may be about either non-present third parties, inanimate objects, or situations but not present parties (as in direct complaints).


Sequences involving an expression of negative stance, criticism, dislike, and so on that lack afective involvement, or single turns or expressions of a negative stance that include afect and personal involvement but that are not responded to as complaints by coparticipants and not pursued as such by the speaker were not included in the collections. Similarly, troubles tellings that do not exhibit any orientation to the troublesome issue as being unfair or unreasonable, and thus complaint-worthy, have been excluded.

### *4.3.5 Focal participants*

 The longitudinal analysis focuses specifcally on fve of the conversation circle participants. These participants were chosen because they represent distinct profciency levels at the start of their participation in the recordings – elementary (A1–A2) and upper-intermediate/advanced levels (B2–C1) – and because of their extended participation (except for Mariana). Table 4.1 shows an overview of the participants included in the analysis.


*Table 4.1* Focal participants

The personal details that follow come from information provided in the background questionnaires, through profciency measures, and in personal communication with the participants.

# *Mariana*

Mariana was a 24-year-old PhD student from Spain. Besides Spanish, she spoke English and some Italian. She participated in the conversation circle for approximately two and a half months in the group Mercredi-1, which is the same as Malia. The combined information about Mariana's French skills suggests that she was an elementary (A1–A2) speaker. She took part in six recordings, during which she was the main or one of the main complainants in seven complaint sequences. The short time span of Mariana's participation in the recordings made it impossible to analyze her complaint practices longitudinally. Instead, the analysis of her complaints serves to complement the overall observations about the practices of elementary speakers.

# *Suresh*

 Suresh was a 27-year-old master's student from India. Besides speaking Hindi, his L1, he was highly profcient in English. Suresh participated in two diferent groups (frst Lundi, then Jeudi) for approximately 19 months but with certain breaks. The combined information about Suresh's French skills suggests that he was a lower-elementary (A1) speaker at the beginning of his participation. Because Suresh was enrolled in an English-language master's program and mainly interacted in English in his everyday life, Suresh's experience with French was limited. During his frst three semesters in the conversation circle, his French-language contributions to the interactions were limited compared to other participants. During the fourth semester, he participated more actively and interacted freely with his coparticipants, showing interactional skills of a lowerintermediate (B1) speaker. He participated actively in seven complaints during the 28 recordings in which he took part. Four of these occurred during his time as an elementary speaker and three when he had reached lower-intermediate profciency.

# *Aurelia*

Aurelia was a 25-year-old PhD student from England. Besides English, Aurelia also spoke fuent Spanish. The available information about Aurelia's French skills suggests that she was approximately at upper-elementary (A2) level at the beginning of her participation. She participated for 15 months, but most regularly in the frst and second semesters. In her third semester, she took the B2–C1 French course, and she assessed her profciency level as at least B2 at this point. Aurelia participated in the group Lundi, which is the same as Suresh. Aurelia was a highly active interactant. During the 18 recordings with her, she was the main or one of the main complainants in 35 complaint sequences. The longitudinal distribution and many complaints in Aurelia's recordings allow for a longitudinal analysis of her practices. To include a comparable number of complaints by Aurelia *vis-à-vis* the other participants in the analysis, only the complaints from the frst two months (11 sequences) of her time as an elementary speaker were analyzed. Ten complaints from Aurelia's time as an intermediate speaker (months 6–9), and nine complaints from her

time as an upper-intermediate/advanced speaker (months 11–15), were also included in the analysis.

# *Malia*

Malia was a 30-year-old PhD student from Iran. Besides Farsi, Malia spoke fuent English. She participated in the conversation circle for 16 months. The cumulative information about Malia's French skills placed her as an upper-elementary (A2) speaker at the beginning of her participation. In her third semester, Malia took the B2–C1 support course, as she had already completed a B2-level summer course. She participated in the group Mercredi-1 (same as Mariana). Like Aurelia, Malia was a highly active interactant. In the 23 recordings with Malia, she was the main or one of the main complainants in 35 complaint sequences. The longitudinal analysis of Malia's complaints is based on the frst ten complaints (months 1–3, elementary level), eight complaints from the second semester (months 7–9, intermediate level), and eleven complaints from the third semester (months 11–15, upper-intermediate/advanced level).

# *Cassandra*

Cassandra was a 23-year-old bachelor's degree student from Italy. At the start of the recordings, she had lived in Launève for three years. She studied L2 French as one of her main subjects, with courses at the B2 level. Besides Italian and French, she spoke intermediary English. She participated in the conversation circle for nine months in the group Mercredi-2. In the 13 recordings in which she took part, she was the main or one of the main complainants in 16 sequences. The limited number of complaints, particularly at the beginning of the recordings, makes a longitudinal analysis of Cassandra's complaint practices difcult. A preliminary analysis of her complaints did not reveal any distinct changes in complaint practices over time. This may be due to her already high French level at the beginning of her participation (for similar observations about topic management, see König, 2019). The analysis, therefore, includes all of Cassandra's complaints in the upper-intermediate/ advanced collection.

#### *4.3.6 Overview of collections*

The focus on the fve participants just presented allows for an analysis of complaint practices longitudinally within participants (Suresh, Aurelia, Malia) and across profciency levels (elementary, intermediate, and upperintermediate/advanced). In total, I have analyzed 86 complaint sequences distributed across three collections corresponding to the three profciency levels (Table 4.2)2 .

# 52 *Investigating change longitudinally*



While I initially included all three levels in the analyses, in Chapters 5 (on the structural organization of complaining) and 6 (on interactional resources for complaining), I only compare elementary (31 sequences) and upper-intermediate/advanced level (36 sequences) complaints. The preliminary analysis suggested that the intermediate level sequences indeed represent an 'in between' stage between elementary and upper-intermediate/advanced levels. Considering the lengthy nature of many complaints, the choice to exclude the intermediate-level complaints from Chapters 5 and 6 is also due to practical reasons. In contrast, the two case studies presented in Chapter 7 (on the interactional history of a complainable) track all complaints in the data made by Suresh and Malia, respectively, about two particular complainables, including complaints from the intermediate level collection.

### *4.3.7 Comparability*

To enhance the comparability of my analyses, I have kept the speech exchange system, the (type of) participants, and the activity type constant over time and across participants. More specifcally, the conversation circle provides for a speech exchange system consisting of small group interactions that closely resemble ordinary conversation. This setting remains constant over time. The participants are L2-speaking peers who are (all except one) university students. Upon their start in the conversation circle, they were matched with other speakers at approximately the same French level. Over time, a few participants left the activity, and a few joined. In both Aurelia and Malia's cases, comparability is nevertheless high since the coparticipants of both groups (Lundi and Mercredi-1) changed in the third semester. Both the elementary collection and the upper-intermediate/ advanced collection thus include complaints made by Aurelia and Malia to relatively new acquaintances (the same applies to Mariana). As for Cassandra, her later sequences are from interactions with coparticipants that she has known for quite a while. These later sequences thus illustrate the practices of the most profcient speakers among well-known acquaintances. Naturally, this schematized characterization of the participants' relationships does not represent the dynamic nature of human relations. The relationships also change as the participants meet regularly over time and increasingly get to know each other, and this may afect the observed complaint practices. Chapter 7 addresses these questions specifcally by investigating how the observed changes in participants' complaining intersect with the development of shared interactional histories and evolving social relationships.

The activity type is kept constant over time, as the analysis only investigates sequences with indirect complaining. While many studies on indirect complaining focus on complaints about third parties, I also include complaints about inanimate objects and states of afairs. This was a data-driven decision since many of the complaints, in fact, concern these types of complainables, such as the French language, Swiss society and culture, and university courses. These are topics that the participants have in common as university students and newcomers to Switzerland. A small diference is observable between the complaints of elementary and upper-intermediate/advanced speakers, with a slightly higher proportion of complaints about third parties among more advanced speakers than among elementary speakers.

As for the sequential environment, which is also mentioned by Wagner et al. (2018) as a relevant factor for enhancing comparability, I decided not to control for this in the establishment of the collections. In contrast to investigations of distinct social actions, such as questions, responses to questions, or requests, which typically are done in particular sequential positions, the nature of complaints as social activities consisting of larger sequences of actions (with several adjacency pairs) would make such control difcult. I discuss the sequential position of complaint *initiations* in Section 5.2.

# **Notes**


# **5 The structural organization of L2 complaints**

This frst analytical chapter concerns the structural organization of indirect complaints in L2 French. It focuses on the core actions involved in complaints, the ways in which speakers move into complaining, and the co-construction of complaint sequences. I present both similarities and differences in how elementary (A1–A2) and upper-intermediate/advanced (B2– C1) speakers structurally organize their complaints. I frst identify a series of 'building blocks' that recurs in the complaint sequences of both elementary and more advanced speakers (Section 5.1). I then document diferences over time in the way complaints are initiated (Section 5.2) and co-constructed by the participants (Section 5.3). I fnally discuss the implications of the fndings for our understanding of the development of L2 IC (Section 5.4).

#### **5.1 Interactional building blocks of indirect complaints**

As discussed in Section 3.2, research focusing on the structural and sequential organization of complaining is scarce. We know that complaint sequences tend to be long and complex and that their structure is highly contingent on coparticipants' contributions. The contingent and variable nature of complaints is most obviously evidenced in my data by the fact that sequences are between 30 seconds and more than 20 minutes long. This variability supports the conceptualization of complaints as interactional projects (Levinson, 2013; see also Chapter 3), whereby the interactional project (the complaint) must be diferentiated from the sequence in which it is produced. Despite the apparent variability in length of the sequences, the basic building blocks, or core actions, remain stable, even when comparing the complaints of elementary and upper-intermediate/advanced speakers. In a similar vein as Jeferson (1988) and Traverso (2009), I now present the 'sequence candidate' (Jeferson, 1988: 418) of indirect complaints observed in my complaint collection. This sequence candidate does not present an exhaustive structure that accounts for all possible sequential developments but rather outlines a recurrent pattern in the data.

I present the sequence candidate with the help of a complaint at the elementary (A1–A2) level. In this excerpt (Excerpt 5.1), Malia (MAL) complains about the fact that she did 'nothing' (presumably nothing fun or interesting) the preceding weekend because she had to study as preparation for her students' upcoming exam (as a PhD assistant, Malia occasionally administers exams to students). The omitted lines (lines 1–4) include a pre-sequence in which Malia produces a circumstantial preface, and the coparticipants confrm their listenership (see Excerpt 5.2). In lines 5–7, Malia presents what will become the object of her complaint:

```
Excerpt 5.1 'Last weekend' (Mer1_2016–12–07)
```
Although Malia's assertion that she did *rien* ('nothing') last weekend (lines 5, 7) could be interpreted as a neutral report of her past days, the extreme-case formulation (Pomerantz, 1986) *rien*, the prosodic stress on the same word, and Malia's embodied conduct of closing her eyes and shaking her head (FG.5.1) render the assertion understandable as a negative one. By virtue of its negative valence, the assertion projects a complaint as *possibly* upcoming. Mariana (MAR) and Theo (THE) respond to this assertion with verbal and embodied conduct (lines 8–9, 11) that Malia treats as expressions of doubt, as she, in lines 12–13, insists on the veracity of her assertions (*non,* 'no'; *vraiment*, 'really'). She then develops

an account in which she details and justifes the expressions of negative stance (Drew, 1998).

Malia explains that the reason that she did not do anything last weekend is that she had to study for her students' exam (lines 15–34). The format and delivery of the account express the unreasonable and complaintworthy nature of the situation, frst through the verb *devoir* (*devai:s*, 'had to', line 15), which reports an obligation, and the prosodically marked *étudie:r* ('study', line 15) and then, once Malia has responded to a clarifcation request from Mariana regarding the exam (lines 19–24), through the emphatic assertion that it was only one paper sheet that took such a long time to study; see the stress on the frst *même* ('even') and *une feuille* ('one sheet', line 26). After repeating *rien* ('nothing', line 29), Malia restates the reason that she did nothing: She had to study for her students' exam (lines 30–31, 33–34).

As seen in lines 36–37, Malia's account is responded to with only some alignment tokens from Mariana and Theo. What follows is a typical sequence-closing sequence (Scheglof, 2007), through which Malia initiates a move toward closure while simultaneously seeking more afliative responses from her coparticipants. She initiates a summary assessment with *c'est très-* ('it's very', line 38). By gazing from one coparticipant to another during the delivery of this turn, Malia seeks close embodied engagement with her coparticipants – likely to recruit their engagement and afliative responses – but she encounters difculties with producing the assessment segment (see micro-pause, lowered gaze, and the nonlexical vocalization that delays the delivery of the adjective). She settles on the adjective *drôle* ('funny') and then gazes down (line 39). As no response is forthcoming, Malia ofers the closing-implicative statement *mais c'est ça* ('but that's that', line 40; cf. Jeferson, 1988). Mariana and Theo eventually respond by laughing (lines 41, 43), after which Malia takes back her assessment of the situation as 'funny' (lines 45–46). Instead, she assesses the situation as *horrible* ('horrible', line 50), which more accurately matches Malia's expressed stance. By lowering the volume and producing the extreme-case formulation in a laughter-infused smiley voice while still gazing down, Malia both embodies her claimed negative stance and shows some 'troubles-resistance', displaying her ability to take the troubles lightheartedly (Edwards, 2005; Jeferson, 1984b).

Whereas Zarah merely laughs with Malia (lines 51–52), Theo shows his understanding of Malia's situation and ofers a token of sympathy through the assessment adjective *fatigué* ('tired', line 55) in low volume. Malia immediately (line 56) confrms Theo's interpretation and upgrades it (see Holt, 2000). She then gazes into empty space (line 57), embodiedly disengaging from the sequence. After the fnal closing of the sequence (lines 58–61), Mariana initiates a new sequence by asking Theo a question (line 62).

The overall composition of Excerpt 5.1 is representative of the complaints at both elementary and upper-intermediate/advanced levels. The sequence candidate may be summarized as follows:

# (*1*) *An expression of a potential complainable*

This typically takes the form of a negative assessment turn or a neutrally formatted assertion with underlying negative valence (lines 5–7 in Excerpt 5.1). It presents a problem with or a criticism of a non-present third party or situation. Although many complaint sequences start with this component, at this point, the sequence has not yet developed into a complaint (cf. the 'initiation phase' in Traverso, 2009), as the speaker's action of (potentially) initiating a complaint has not yet been ratifed by the coparticipants. As seen in Excerpt 5.1, extreme-case formulations, marked prosody, and embodied conduct are some of the resources used to characterize the issue as a complainable. In some cases, the frst expression of the complainable is more subtle, merely implied (cf. Ruusuvuori et al., 2019; see also Section 5.2). It is typically responded to with displays of listenership or alignment tokens that ratify the speaker's course of action and allow him/her to continue, or by signs of resistance. Alternatively, coparticipants may actively add to the sequence with their own assessments or assertions that align with or upgrade the frst assessment/assertion; sometimes, this leads to joint complaining (see Section 5.3).

(*2*) *A detailing of the complainable situation or behavior* 

This is where the sequence more clearly develops into a complaint. This component often consists of a storytelling or a report that outlines, exemplifes, and accounts for the complaint (lines 15–34 in Excerpt 5.1). It is similar to what Drew and Holt (1988) call the circumstantial detailing of the complaint, in which descriptions of concrete facts are ofered as 'evidence' for the complaint, and to the 'development phase' outlined by Traverso (2009). The detailing varies in length, from a few turns (typical at the elementary level) to a longer series of examples and tellings (more common at the upper-intermediate/advanced level). DRS and reenactments are regularly part of stories (see Section 6.2). Tellings often adopt a humorous tone despite the underlying negative valence (Edwards, 2005; Jefferson, 1984b; see also Glenn & Holt, 2013, on the recurrence of laughter in moments of trouble).

(*3*) *A summary assessment, restatement, or formulation of the complaint* 

This component summarizes, restates, or formulates the complaintso-far, typically through an explicitly formulated high-grade negative

# 60 *The structural organization of L2 complaints*

assessment (lines 38, 50 in Excerpt 5.1) or (more rarely in my data) a negatively valenced idiomatic expression (see Drew & Holt, 1988; Rääbis et al., 2019; and Ruusuvuori et al., 2019, and discussion in Section 3.3.1). This component is similar to what Drew and Holt (1988) refer to as the explicit formulation or naming of the complaint. If recipient responses have been lacking or minimal so far, the formulation may work to elicit further displays of afliation. It also serves to mark the end of the telling or report and thereby prepare for potential sequence closure.

(*4*) *Recipient recognition of the complaint as a complaint* 

While afliative recipient responses are not limited to particular sequential positions, more elaborate (non-minimal) responses are expected at the end of a complaint telling and/or after the formulation or restatement of the complaint. Preferred responses include displays of afliation or sympathy, whereby coparticipants at the very least recognize the legitimacy of the complaint (line 55, Excerpt 5.1). At the elementary level, afliative displays are often verbally minimal (as in Excerpt 5.1); at the upper-intermediate/advanced level, they tend to be more elaborate (see Section 5.3). In cases where coparticipants have epistemic access to the complainable, afliative responses may include displays of agreement and upgrades of negative assessments, sometimes leading to joint complaints (see Section 5.3). Not all responses are afliative, however. Coparticipants sometimes resist the complaint by rejecting its grounds or by working to close or defuse it (Holt, 2012; Mandelbaum, 1991).

(*5*) *Expansion/closing* 

After the coparticipants' displays of recognition of the complaint as a complaint, complainants often expand the sequence with further stance displays or by ofering more evidence in support of the complaint. Coparticipants may also contribute to the expansion by producing 'my-side' tellings (second complaint stories; see Selting, 2012) or reports that support the overall complaint. Expansions may be quite elaborate, especially at upper-intermediate/advanced levels. Alternatively, participants immediately move to close the sequence once the recipients have shown afliation or sympathy, or they transition into another activity after unsupportive responses. In my data, most complaints eventually receive afliative responses of some kind (see Section 5.3).

It is typically the complainant who ratifes the closure of the sequence by ofering sequence-closing thirds (Scheglof, 2007) following displays of afliation and by displaying that he/she is 'done' with the complaint and ready to move on to other business. Final closing moves include conventional closing statements (line 60, Excerpt 5.1) and the initiation of a new sequence by the complainant himself/herself. In addition, similar to what has been observed by Traverso (2009), complaint closings are recurrently interrupted rather than progressively closed, such as by the arrival of a new participant or by the abrupt invocation of the time and the necessity to leave (cf. also Jeferson, 1984a, on marked closings of troubles talk). For reasons of space, I do not analyze complaint closings in detail.

The sequence candidate presented here is the result of a retrospective analysis. The frst expression of a potential complainable cannot be considered a complaint component from the participants' perspective since the sequence at this point has not yet developed into a complaint (but see Edwards, 2005, on speakers' general reluctance to characterize what they are doing as complaining). The 'point of no return' for complaint sequences seems to lie rather after the speaker's detailing of the problematic circumstances or transgression of a third party, as seen in the complainant's orientation to the relevance of obtaining afliative responses to this detailing.

These observations concur largely with the fndings of Traverso (2009) about L1 French speakers (see also Ruusuvuori et al., 2019, on Danish and Finnish). Rather than speaking of distinct 'phases' involved in complaining, however, I suggest that the overall sequence structure refects the orderly unfolding of the diferent interactional tasks involved in complaining. Essentially, complaining boils down to the following tasks: On the one hand, complainants need to (1) present the object of the complaint and the complainable and (2) justify the complaint; show its complaintworthiness (Drew, 1998). Complaints and the negative assessments they include are accountable acts that need to be justifed. These tasks correspond to points 1–3 in the sequence candidate. On the other hand, complaint recipients need to (1) align as recipients of a longer turn and (2) express their own stance toward the complaint, 'preferably' through displays of afliation and/or sympathy. The frst of these tasks is a prerequisite for the complaint to come about; the second corresponds to point 4. Finally, the participants need to move out of the complaint (point 5). The lack of diference over time in core actions shows the participants' convergent understanding of complaints as interactional projects and of complaining as an activity, and also concurs with the fact that indirect complaining has been described in the literature in largely similar terms across the languages and cultures investigated so far (see Chapter 3). On a methodological level, this similarity helps to warrant comparability in the comparative analysis.

But if the basic building blocks of complaint sequences are the same among both elementary and upper-intermediate/advanced speakers of French, what *changes* in the structural organization of these? As hinted at

earlier, there are diferences in the ways speakers initiate and co-construct complaints, which is something that I address in the following sections.

### **5.2 Moving into complaints**

The moral, contingent, and potentially delicate nature of complaining has consequences for the initiation of complaints. As presented in Section 3.2, research on both everyday conversations and institutional interactions has shown that speakers often move into complaining in a stepwise manner to test the grounds of the complaint before launching the activity fully. Future complainants regularly hint at the upcoming criticism before formulating it overtly and escalate negative stance expressions progressively. They may also preemptively account for and build the credibility of the upcoming complaint and portray themselves in a positive light before launching overt criticism. Complaints emerging through complaint stories are typically initiated in similar ways as other stories, through prefatory work that secures recipiency, provides circumstantial information, and helps recipients anticipate the nature of the upcoming story (Section 3.2).

The longitudinal analysis of complaint initiations in my data reveals a change over time in (1) the sequential position of complaint initiations and (2) the pre-complaint work speakers accomplish before overtly launching a complaint:


I now illustrate these diferences through empirical examples from speakers at the respective profciency levels.1

#### *5.2.1 Elementary speakers*

The large majority of complaints (24 of 31, or 77%) of elementary speakers are initiated in first position, whereas second-position initiations are less common (7 of 31, or 23% of sequences). Because many of the complaints are offered in the form of first-position tellings, they are often initiated in ways that resemble storytelling openings. Excerpts 5.2–5.4 exemplify typical complaint initiations at the elementary level.

Excerpt 5.2 is the beginning of the complaint that Malia presented in Excerpt 5.1. Before the start of the excerpt, Theo asks Malia whether she has been to a place that Mariana says she went to the previous weekend, which Malia has not. Theo assessed the place as very nice, and 2.9 seconds of silence ensues (line 1), during which the participants look down or into empty space, orienting to the prior sequence as closed. Malia then initiates the sequence that develops into a complaint about the fact that she did not do anything at all last weekend besides studying. I only present the beginning of the sequence until Malia begins detailing the complaint:

*Excerpt 5.2* 'Last weekend' (Mer1\_2016–12–07)

Malia's turn in line 2 (*e:h (0.4) le week-end (1.9) passé?*, 'uh last weekend') works as a brief circumstantial preface for a news announcement or telling about a past event. It establishes the topical relevancy of the talk after the prior sequence – which was about Mariana's weekend. The rising intonation and Malia's gaze conduct invite coparticipants to confrm recipiency (Stivers & Rossano, 2010), which they do (lines 3–4). The preface is similar to the story-prefaces produced by a less profcient L2 speaker in Pekarek Doehler and Berger's (2018) study of story openings in that it secures recipiency and frames the talk as being about a past event but does not convey any particular stance and hence does not project that it will be a complaint story. Stance expressions instead come in the next step, as Malia presents what will become the object of the complaint: that she did not do anything during the weekend (lines 5–7). As mentioned earlier, the extreme-case formulation (*rien*, 'nothing'), the prosodic emphasis, and Malia's embodied conduct of closing her eyes and shaking her head (line 7, FG.5.2) convey a negative stance. Following the coparticipants' laughter and embodied responses (lines 8–9, 11), Malia insists on the veracity of her assertion (lines 12–13) and subsequently develops the complaint (see Excerpt 5.1).

The next excerpt contains a similar circumstantial preface but also what, according to Sacks (1992, Vol. I: 359), appears "with an immense regularity" in the complaints of some people – namely, a complaint-initiation that is preceded by a positive assessment (what Sacks calls 'praise') and the contrastive conjunction 'but' (*mais*, in French). What follows ('something else', in Sack's terminology) is recognizable as a complaint-initiation in the context of the praise. If the second assertion/assessment is not already

formatted in a way that clearly indexes a negative stance, it is understandable as such against the background of what came before. The contrastive format resembles a recurrent practice for prefacing criticism and other types of dispreferred frst actions (Clayman, 2006; Golato, 2005) – namely, the use of a positive assessment or compliment that delays and mitigates the dispreferred frst action. In my data, this format is recurrently used by Aurelia and Malia (at elementary and more advanced levels) and by Cassandra (upper-intermediate/advanced level).

In Excerpt 5.3, Malia will initiate a complaint about her difculties speaking French at work, especially when it comes to speaking with the professor, who is also her boss. The laughter in line 1 closes the prior activity, in which the participants engaged in a longer repair sequence.

*Excerpt 5.3* 'Speak with professor' (Mer1\_2016–11–16)

Malia initiates the new sequence while keeping her gaze lowered (see line 1). After a restart, she ofers an announcement about something that pertains to *jour après jour* ('day after day', line 4). By delivering the end of her turn with rising intonation and gazing at Mariana, she invites recipient response, and both Mariana and Theo confrm recipiency (lines 5–6). This pre-sequence frames Malia's upcoming talk as concerning a recurrent event and recruits the coparticipants' attention for a longer turn.

Malia then asserts that she tries to study 'a lot' and listen 'a lot and a lot' (lines 8–10). The fact that her eforts pertain to studying and listening to French is understandable against the background of the prior repair sequence and considering the conversation circle setting. Malia's accompanying embodied conduct (gestures, fnger snapping, frowning, lines 9–10) upgrades the strength of the already high-grade assertions. The claim that she studies and listens a lot works as subtle self-praise, by which Malia portrays herself as an eager and studious L2 learner.

The immediately following and prosodically marked *mai:s* ('but', line 10) then projects a continuation that contrasts with Malia's attempts at studying and listening every day, and by laughing (line 11), Zarah shows her anticipation of the projected contrast.

Malia's contrasting assertion concerns what happens to her when she wants to speak to her professor (lines 12–13). The contrast takes the form of a compound TCU (Lerner, 1996), of which the frst part is the subordinate clause of a bi-clausal pattern, syntactically projecting another (main) clause. Instead of verbally completing the turn, Malia drops her hands on the table and rolls her eyes before closing them and shaking her head, while frst breathing in and then letting out a loud sigh (line 14, FG.5.3). The embodied and vocal conduct are clear displays of negative afective stance (Selting, 2010a, 2012; see also Clift, 2021; Goodwin & Alim, 2010, on eye rolls), non-verbally expressing Malia's difculties associated with speaking with her professor – which will be the object of the upcoming complaint.

As demonstrated by Iwasaki (2009), speakers regularly use embodied conduct (gaze, gestures, shifts in posture, facial expressions, etc.) to invite coparticipants into the production of turns-in-progress. When Malia produces the frst part of the compound TCU, she indeed is seeking close embodied engagement with the coparticipants. This is seen in her gaze: She looks frst at Mariana (line 12) and then, after a repair, at Theo and then at Mariana again (line 13). Mariana and Zarah's syntactically ftted collaborative completions (Lerner, 1996) *c'est très difcile* ('it's very difcult', line 15) and *ça marche pas* ('it doesn't work', line 16) show the coparticipants' close monitoring of Malia's projection – namely, the expression of a problem related to speaking. By completing Malia's turn and participating in the 'evaluative loading' (Goodwin & Goodwin, 1992: 157) of Malia's talk, Mariana and Zarah both display their understanding of and ratify Malia's course of action and afliatively support the sequential development (Lerner, 2013). In initiating a retelling of a specifc troublesome event the same day, Malia accepts the coparticipants' completions and develops the complaint (line 18 and onward).

The contrastive formulation following the circumstantial preface in this excerpt allows the speaker to do some subject-side (Edwards, 2005) precomplaint work, delay verbal expressions of negative stance, and check coparticipants' willingness to embrace the direction of the talk. Through the self-praise, Malia portrays herself as a studious French learner who complains only about especially difcult situations – such as when speaking to her superior at work – rather than being a perpetual 'whiner' (Edwards, 2005). The strong projective force of the contrastive formulation (here combined with embodied conduct inviting coparticipant participation) also allows the speaker to convey the object of the complaint without verbalizing it herself. Instead, it is the coparticipants who verbalize the problem, while Malia produces a non-lexical vocalization and embodied conduct expressing a negative stance. The contrastive formulation thus works as simple yet efective means to manage the contingencies involved in the initiation of a complaint.

In terms of complaint initiations in *second* position, all but one secondposition complaints at the elementary level are initiated as part of an answer to a question, typically an open-ended question such as a status-update

### 68 *The structural organization of L2 complaints*

inquiry inviting a longer answer (see Excerpt 5.8 for the exception). In these cases, securing recipiency for an extended turn is not an issue. Speakers may nevertheless accomplish some pre-complaint work to deal with the potential delicacy of producing a negatively valenced answer to an open-ended, neutrally formatted question. One documented practice among elementary speakers is to ofer embodied displays of stance, sometimes together with non-lexical vocalizations, before a verbal answer. Such embodied displays push back negative talk further in the turn or sequence and may work to elicit coparticipant ratifcation of the course of action.

Excerpt 5.4 illustrates such an initiation. Before this excerpt, Zarah has given a status update about herself, reporting on some difculties related to her university application. After sequence closure, Malia invites Mariana to tell some news about herself (line 3):

*Excerpt 5.4* 'The bank' (Mer1\_2016–11–02)


### *The structural organization of L2 complaints* 69

In the brief gap that follows Malia's question (line 4), Mariana closes her eyes and lets her hands fall to the table; her head is already slightly tilted down (FG.5.4). Zarah immediately starts laughing (line 5), orienting to Mariana's embodied conduct as a response signaling a non-straightforward and potentially negatively valenced answer to come. At the same time, Mariana leans forward, lowering her head even more, while smiling (line 5, FG.4.5), thus visibly further delaying a verbal response. In partial overlap with Zarah's laughter, Malia also produces a loud laughing sound but elongates the last vowel so that it sounds more like whining than sincere laughter (line 6) as Mariana raises her head and posture again to face her coparticipants. Malia's next action shows her interpretation of Mariana's embodied displays. She ofers a candidate formulation of Mariana's expressed stance as *fatigu*↑*ée:::* ('tired', line 8). The delivery of this turn, with a pitch upgrade on the last, elongated syllable, and the laughter that follows immediately afterward, indicate that Malia is in fact ofering a mocking reenactment of Mariana's yet unarticulated answer. Mariana confrms Malia's candidate (line 9), and some more laughter from Malia follows (line 10).

In line 11, Mariana recycles Malia's *fatiguée* and integrates it into the self-assessment *je suis très fatiguée* ('I am very tired'), followed by an account initiation explaining why she is tired: She went to the bank. This constitutes the frst step of developing the sequence into a complaint about being tired after having to go to the bank and not understanding anything when speaking to the staf in French. By ofering embodied stance displays instead of immediately answering her coparticipant's open-ended question, Mariana foreshadows negative talk (cf. Ruusuvuori & Peräkylä, 2009, on embodied stance expressions foreshadowing verbal ones) and recruits the coparticipants' go-ahead signals before expanding the sequence. The use of non-linguistic resources hence constitutes another way for elementary speakers to orient to the contingencies of complaining in complaint initiations.

In sum, Excerpts 5.2–5.4 have illustrated recurrent practices used by elementary speakers to move into complaining. Focusing mainly on the most common type of complaint initiations at this level – complaints initiated in frst position as part of storytellings – I have shown the general tendency among elementary speakers to initiate complaints following limited pre-complaint work such as circumstantial prefaces that situate the upcoming talk in time and place but do not project its negative valence. Although in some sequences the complainable is introduced without any orientation to delicacy (e.g., Excerpt 5.2), in other cases speakers use embodied conduct (Excerpt 5.4) or contrastive formulations with a frst positively valenced element (positive assessment/self-praise; Excerpt 5.3) to delay and foreshadow the verbal expression of negative stance and to some extent portray the speaker in a positive light. Such contrastive formulations and embodied expressions of stance seem to be efective means to recruit coparticipants' ratifcation of the speaker's course of action (Excerpt 5.3–5.4), which facilitates the development of the complaint. As I show next, with time speakers diversify their practices for moving into complaining in ways that index the moral, delicate, and contingent nature of complaints.

#### *5.2.2 Upper-intermediate/advanced speakers*

Upper-intermediate/advanced speakers initiate complaints in frst position in 13 of 36 sequences (36%) and in second position in 23 of 36 sequences (64%), thereby more frequently in second position than elementary speakers (who initiate complaints in second position in 23% of cases). Secondposition initiations occur both in answers to questions and in response to other speakers' stance displays. While some initiations at this level resemble those at the elementary level, distinct diferences can often be observed. Upper-intermediate/advanced speakers typically initiate complaints about a complainable that is topically related to the ongoing talk. The initiations, therefore, rarely involve the kind of circumstantial preface used by elementary speakers to situate the complaint in its interactional context. Since such complaints often 'compete' with other talk, participants may have to do more work to secure the conversational foor and a longer turn. As with elementary speakers, more advanced speakers sometimes launch complaints without or with only limited orientation to delicacy. In some cases, however, these speakers accomplish elaborate pre-complaint work in which they progressively escalate expressions of negative stance and preemptively work to establish the legitimacy of the complaint and of themselves as complainants (Excerpts 5.5–5.6). This refects a diversifcation of practices over time, manifested in changes in both sequence organization and in the interactional resources deployed in complaint initiations.

Excerpt 5.5 demonstrates the more elaborate pre-complaint work advanced speakers may do to prepare the grounds for the upcoming complaint and move into complaining in a stepwise manner that indexes delicacy. Before the excerpt, Aurelia (AUR) and Jordan (JOR) have discussed Aurelia's new apartment. In lines 1–3, Jordan asks why Aurelia moved from her old place. Although formatted as a polar question, the negative polarity *c'était pas bien là-bas* ('was it not good there', lines 1–2) projects an account of what was not good with Aurelia's old apartment. In line 4, Aurelia initiates her answer, indeed immediately starting to ofer an account. In the 50 omitted lines, she provides the frst reason for leaving the old apartment – namely, that it was too small. In line 56, she starts formulating the second reason, related to her old fatmate. It is within this part of her answer that Aurelia will launch a complaint about the fatmate.

*Excerpt 5.5* 'Sharing' (Lun\_2018–05–28)

### 74 *The structural organization of L2 complaints*

In line 56, Aurelia initiates the second part of her answer, projecting a second reason for moving from her old apartment (*et aussi*, 'and also'). She formulates an *if*-conditional with a hypothetical statement about what is important if one lives with other people in a small apartment: One must like sharing (lines 56–60). The impersonally formulated 'it's important' carries a moral dimension, as it invokes "normative standards of conduct" (Drew, 1998: 297) and thereby normatively reasonable behavior associated with sharing an apartment. After gazing at Jordan and leaving time for him to stop drinking and to acknowledge her assertion (lines 61–62), Aurelia positively assesses one of her other old fatmates, with whom she had a really good time (lines 63–72). The description of the other fatmate as <*incroyable>*  ('incredible', line 64) and someone with whom she could have a beer when coming back from work (lines 68–72) portrays an ideal case of a fatmate, with which Aurelia's subsequent criticism of her other fatmate will strongly contrast. Jordan displays his attentiveness and listenership by nodding at key moments in Aurelia's telling (lines 65, 67–72). Before issuing the implicitly projected contrast, Aurelia again describes some of the obligations that come with living in a small place, now using a diferent impersonal expression that invokes normative rules of morality: *il fau::t partager la cuisi:ne et tout ça* ('one must share the kitchen and all that', lines 75–76), and Jordan agrees (lines 77–78). Aurelia expands by repeating that it is okay (to be in a small apartment) if the person likes sharing (lines 79–80) because one can always fnd a solution (line 84).

Until now, Aurelia has (1) invoked normative moral expectations associated with apartment sharing and hinted at a problem with this, (2) described an ideal fatmate who surpassed such normative expectations, and (3) portrayed herself as a person who gets along with other, reasonable fatmates. Only after this, she introduces the projected negative contrast.

The contrast, too, takes the form of a hypothetical *if*-conditional (initiated with *mais*, 'but', line 85), which asserts what happens if the fatmate is someone who does not really like sharing (lines 85–86). Aurelia assesses through another generic statement that 'it's difcult to live like that' (line 87) because one has to share (line 88). As Aurelia restates the social norms associated with living together (*il faut partager*, 'one must share', line 88), she opens her hands with her palms on the table (FG.5.6) so as to underline the obviousness of the assertion (Kendon, 2004; Marrese et al., 2021). The explicit formulation of the problem, and thus of the complaint object, comes only after this: In the

end, Aurelia realized that the other fatmate, in fact, did not like sharing (lines 89, 91–92; see also headshakes expressing further disapproval). She then initiates a report with specifc examples of the transgressions made by the fatmate that accounts for and constructs the complaint (line 93 and onward).

This excerpt illustrates the intricate kind of pre-complaint work that occurs among more advanced speakers. While the basic contrastive format whereby a criticism is preceded by something positively valenced is similar to that seen among less advanced speakers, it is structurally more complex and goes further in its moral dimension. Aurelia uses the contrastive format to compare normative codes of conduct with the complained-about third party's transgressions of these. By expressing the minimal requirements for sharing an apartment as general norms (through impersonal formulations, such as *c'est important* ['it's important'] and *il faut* ['one must']), she constructs such requirements as social norms of common knowledge and thus something for which people can be held accountable in case of a breach of these norms (Drew, 1998). By describing how another fatmate lived up to or even exceeded such minimal requirements, Aurelia further portrays the reasonableness to which the complainable conduct of the other fatmate contrasts. In addition, she portrays herself in a positive light by claiming her own reasonableness as someone who gets along with other fatmates. This extended pre-complaint work thus constructs a morally defensible ideal and portrays the future complainant as a reasonable and credible person who does not complain or whine (cf. Edwards, 2005) about all types of fatmates but only about those with normatively unreasonable conduct. It also allows the speaker to move into the complaint in a stepwise manner (Ruusuvuori et al., 2019), by which the object of the complaint is frst implied through a hypothetical situation before being described in detail and explicitly formulated. Some of the interactional resources used by Aurelia to accomplish the pre-complaint work include *if*-conditionals, impersonal constructions used for invoking normative moral expectations (*il faut*, *c'est important*, etc.), and left-dislocations that allow the insertion of framing information before completing the grammatical projection (see line 89).

Similar pre-complaint practices can be seen in Excerpt 5.6. The complaint concerns Cassandra's (CAS) requirement to write a literature essay for one of her university courses. This sequence, too, begins with an information request from a coparticipant (lines 1–4).

### 76 *The structural organization of L2 complaints*

*Excerpt 5.6* 'Essay' (Mer2\_2017–06–21)

### 78 *The structural organization of L2 complaints*

Xiang's (XIA) question asks how and what Cassandra has to do for her written course work, but Cassandra treats it as an invitation to talk about her upcoming exams more generally and to evaluate her ability to complete the work. With the expression *ça dépend* ('it depends', line 5), Cassandra signals the complex, multi-unit nature of her upcoming answer, and already from the start implies that the answer will contain both a positively and negatively valenced part. Similar to Aurelia in Excerpt 5.5, Cassandra uses an *if*-conditional ('if it's in psychology', line 6) to invoke a hypothetical situation. After Xiang's go-ahead signal (line 8), Cassandra starts evaluating her ability to handle *psychology* assignments. By frst assessing the assignments as *pas simple* ('not easy', lines 9–10) and then asserting that she nevertheless manages (line 11), Cassandra portrays herself as a good student who can manage tough assignments in at least the feld of psychology. She then asserts her long-term experience as a student (lines 15, 17), implying her right to make such assessments (Pomerantz, 1984), before claiming more explicitly that she knows how it works (lines 20–22). Through these assertions, Cassandra thus constructs a nuanced picture of her prior study experiences and of her abilities as a student in a way that implies her epistemic access and deontic right (Heritage & Raymond, 2005; Stevanovic & Peräkylä, 2012) to assess the reasonableness of her coursework.

In line 24, Cassandra introduces the projected negative part of her answer with an elongated *mai::s* ('but'), describing hypothetically what would happen if she needed to write a literature essay (lines 24–25). Exploiting the grammatical projection of the *if*-clause, she adds some epistemic framing information about her lack of experience with literature essays (line 26) while rolling her eyes (FG.5.7, see Clift, 2021; Goodwin & Alim, 2010) and then looking at Xiang. The verbal and embodied conduct allows Cassandra to convey a strong contrast between the literature essay and the psychology assignments (which she has had experience with for several years) and to foreshadow the upcoming verbal expression of negative stance (Ruusuvuori & Peräkylä, 2009). The second part of the compound TCU (initiated with *si*, 'if') consists of an idiomatic expression without an exact equivalent in English: *moi je me prends la tête* (approx. 'me I'm having a hard time', line 27), which expresses the gist of what will be Cassandra's complaint (Drew & Holt, 1988) – namely, the difculties related to a literary essay she needs to write.

As Xiang does not respond but merely gazes back at Cassandra (line 28), Cassandra reformulates the hardship with the high-grade *c'est horrible* ('it's horrible') three times, all produced with marked prosody: low volume, pitch shift the second time, and smiley voice and interspersed laughter the last time (lines 29–30, 32). Through the repetition and marked prosody, Cassandra underlines the severity of the situation, afectively animates her talk, and adds a humorous layer to the complaint (Edwards, 2005; Selting, 2010a, 2012). Only after this, in partial overlap with Xiang's laughing response (line 33), does Cassandra explain that what she presented as a hypothetical situation is something that she, in fact, has to do (line 34); her two consecutive eyebrow fashes reinforce the expressed negative stance (FG.5.8, see Selting, 2012). During an extended silence (line 35), the participants gaze at each other until Xiang fnally ofers an afliative assessment in the form of a non-lexical vocalization (Couper-Kuhlen, 2012) as she shifts her gaze into empty space (line 36). Xiang's 'thinking face' (Goodwin & Goodwin, 1986) indexes her cognitive activity of taking in the serious situation described by Cassandra. After some shared laughter, Cassandra develops a long complaint about all the problems related to her assignment (not shown).

 Just like Aurelia (Excerpt 5.5), Cassandra, over several turns, constructs her own legitimacy as a complainant before explicitly introducing the topic of her complaint. In this case, the complaint is not related to the transgressions of a specifc third party but to Cassandra's unreasonable coursework. Similar to Malia in Excerpt 5.3 but in a more elaborate manner, Cassandra orients to the relevancy of portraying herself as a studious learner who is capable of coping with reasonably difcult studies. Cassandra's self-praise about her ability to handle psychology assignments and the assertions about her status as an experienced student thus work to underline her own legitimacy as a complainant, to allow her to criticize her studies without sounding like a whiner. These actions can be seen as pre-positioned accounting practices, ofered preemptively before any high-grade negative stance expressions. The result is a diferent sequential organization than in most elementary speaker complaints, in which accounts typically *follow*  strong negative stance expressions. Some of the interactional resources used by Cassandra in this pre-complaint work are the projector device *ça depend*  ('it depends'), *if*-conditionals, and the colloquial self-praise construction *je me débrouille* (approx. 'I manage/I muddle through'). Interestingly, Cassandra uses an idiomatic expression in the formulation of the complainable (see also Aurelia's attempt to do so in Excerpt 5.5, line 103), which here allows her to 'depict' the complaint-worthiness of the situation (cf. Drew & Holt, 1988) before verbalizing the complainable in precise terms.

In sum, the comparison between complaint initiations at elementary and upper-intermediate/advanced levels has revealed diferences in the sequential position of complaint initiations and in the pre-complaint work that speakers accomplish before launching their complaints. With time, L2 speakers diversify their initiation practices, and they increasingly initiate complaints in ways that orient to the moral, delicate, and contingent nature of the activity. In Section 5.4, I discuss the implications of the observed changes for the development of L2 IC.

#### **5.3 Co-constructing complaints**

As discussed in Sections 3.1–3.2, complaints have been described in the literature as co-constructed activities since they rely on several participants' cooperation for their emergence and development. The extent to which complaints are co-constructed is, however, contingent on the coparticipants' responses to the complainant's actions. A 'successful' complaint, from the complainant's point of view, leads to afliative and sometimes sympathetic

# 80 *The structural organization of L2 complaints*

responses. Coparticipants can show their afliation as *recipients* of individual complaints. In addition, if the complaint concerns an issue to which coparticipants have epistemic access, they can join the complaint as *co-complainants*, thereby contributing to the construction of a joint complaint (Section 3.2).

There is a diference over time in the level of co-construction of complaint activities, which is manifest both in the coparticipants' responses to individual complaints and in the construction of joint complaints. While most complaints eventually lead to some kind of afliative responses, there are diferences over time in the type of afliative responses and in the proportion of joint complaints:


I illustrate these diferences by means of four empirical examples. In all excerpts, the complaint concerns an issue to which several participants have epistemic access.

#### *5.3.1 Elementary speakers*

The overwhelming majority of the elementary speaker complaints (26 of 31, or 84%) lead to some kind of afliative responses. In 12 (or 39%) of the sequences, however, coparticipants ofer only limited displays of afliation before the complaint is abandoned or interrupted, indicating some problems with either the design of the complaint and/or with the recipients' ability or willingness to ofer clear signs of afliation. Afliative displays in the form of laughter, non-lexical assessments, embodied conduct, or other verbally minimal responses that only to some extent contribute to the sequential development are common.

Excerpt 5.7 exemplifes recurrent coparticipant responses to elementary speaker complaints. Before the excerpt, Aurelia asked Mia (MIA) what French courses she takes, and the participants have established that they take the same A2–B1 course but have missed each other in class so far. Having explained why she takes both the A1 and A2–B1 courses, in line 2 Aurelia initiates a negative assessment of the A2–B1 course. Although Mia clearly has epistemic access to the issue at hand, she responds with only limited displays of afliation (see somewhat afliative responses in lines 12, 31), and Aurelia rapidly abandons the complaint (line 38 and onward).

*Excerpt 5.7* 'No interest' (Lun\_2017–03–27)

## *The structural organization of L2 complaints*

### 84 *The structural organization of L2 complaints*

 Before Aurelia verbally completes her assessment of the A2–B1 course, Mia expresses understanding of Aurelia's embodiedly conveyed stance (headshakes, handshakes, and frown, FG.5.9–5.10) (lines 6–7). Having attempted to assist Aurelia in completing her turn (line 12), thereby showing some level of afliation with Aurelia (Lerner, 2013), Mia receipts the full assessment (line 14) neutrally with *uh-huh okay* (line 16). Aurelia ofers an account for her characterization of the course as not useful by invoking what she *wants* to do (and thereby implying what the course lacks) – namely, to speak (line 17). By opening her hands palm up in two beats synchronized with the prosodic stress of her talk and gazing at Mia (line 17, FG.5.11), she upgrades the strength of the assertion, portrays it as obvious (Kendon, 2004; Marrese et al., 2021), shows her afective involvement (Selting, 2010a, 2012), and invites Mia to respond (Stivers & Rossano, 2010). As Mia merely receipts this with *mm-hm:* (line 19), Aurelia repeats the assertion and the open hand gesture (line 20, FG.5.12) and holds it until Mia provides another receipt token (line 21; see Floyd et al., 2016, on the use of such forward-gesture suspensions).

Aurelia then expands, asserting what she wants to know (in French): how to write normal emails for everyday use (lines 22–23, 25). As Mia again responds with neutral alignment and confrmation tokens (lines 24, 27), Aurelia invokes what she does not want (writing a story, lines 28–29), hence again implying what is problematic with the course. She animates this assertion by enacting writing, pointing to the books behind her. She lifts her hands high (FG.5.13) while interspersing the end of her turn with some laughter, again underlining the obviousness of her claims and adding a humorous layer to the complaint (Edwards, 2005).

This time, Mia responds slightly more afliatively by repeating the response token *uh-huh* (see Norrick, 1987, for self-repetition to express afliation) and ofering a laughter-interspersed *£okay(h) hhh£*, (line 31). However, following Aurelia's next assertion (line 32), Mia remains silent, and Aurelia ofers another account (lines 34–35) that Mia again responds to with neutral alignment tokens (line 36) that treat Aurelia's talk as *informing* rather than *stance-taking*. At this point, Aurelia initiates what at frst appears to be an expansion of the account, but she restarts (line 37) and more neutrally reports on how she goes to class and does her homework and takes notes when the professor says something interesting (lines 38–40, 42, 44, 46). This works as a transition into a 'bright side' (Holt, 1993) assertion about the course, by which Aurelia fnally abandons the complaint and invokes that the course luckily is free (lines

48, 52–53, 55). The participants subsequently shift the topic and do not revisit the complaint.

In this excerpt, Aurelia ofered a series of criticism toward an object that was clearly in the epistemic domain of both participants. Through embodied conduct and prosody, Aurelia conveyed her afective negative stance and invited Mia to participate in the assessment activity (Goodwin & Goodwin, 1992). Mia, however, aligned as a *recipient* by granting Aurelia an extended turn, showing understanding of Aurelia's stance expressions, and by receipting Aurelia's turns with response tokens such as *mm-hm, uh-huh, okay*. Although Mia displayed some level of afliation (e.g., through laughter and by attempting to assist Aurelia linguistically), the majority of her responses were neutrally valenced and did not contribute to the evaluative loading of the complaint. This led Aurelia to rapidly abandon her criticism in favor of more positive talk (cf. the observation by Boxer, 1993, about L2 speakers' problematic responses to L1 complaints). While some elementary speaker complaints are more 'successful' in that they lead to more clearly afliative responses, coparticipants' contributions often resemble the ones seen in this excerpt. Besides embodied and verbal response tokens (head nods, *uh-huh, mm-hm, yeah, oui*, 'yes', *ouais*, 'yeah', *okay*, etc.), these speakers regularly respond with laughter and non-lexical vocalizations (see Section 6.1), by which they align and afliative with each other but to a weaker degree than with, for example, verbal assessment turns (Couper-Kuhlen, 2012). As discussed in Section 5.1, elementary speakers sometimes ofer slightly more elaborate afliative responses once the complainant has clearly moved toward sequence closure.

Not surprising given the typical complaint responses at the elementary level illustrated in Excerpt 5.7, most elementary speaker complaints remain individual complaints (26, or 84%, of 31 sequences). Only a few sequences develop into joint complaints (fve, or 16%, of the sequences, but in three of these, the participation of the person who initiates the complaint is marginal). The only jointly constructed complaint in which one of the focal elementary participants joins the complaint by upgrading a frst negative assessment is in what develops into a complaint about the weather (Excerpt 5.8).

Before the start of Excerpt 5.8, Zarah has told the coparticipants about her previous hometown in Sweden. Malia and Mariana have suggested that it must be cold there now, which Zarah confrms by asserting that there is even snow (line 1). In line 4, Theo assesses the temperature in Launève as very cold, thereby launching a sequence that leads to a joint complaint about the current weather conditions:

### 86 *The structural organization of L2 complaints*

*Excerpt 5.8* 'Fog' (Mer1\_2016–12–07)

## *The structural organization of L2 complaints*




Theo's assessment (line 4) is neutrally formatted but understandable as negative in the context of the preceding discussion about the cold weather in Sweden. It is also responded to as such by Malia, who agrees with *oui oui* ('yes yes', line 5) and repeats it in a full clause: *c'est très froid aujourd'hui:* ('it's very cold today', line 6) while fumbling with her sweater sleeves to partly cover up her hands. At the end of her turn, she adds the non-lexical vocalization *ohhh* to index her afective negative stance. By integrating Theo's assessment in her own turn and producing a modifed repeat (Stivers, 2005), Malia asserts epistemic independence (Heritage & Raymond, 2005) and thereby contributes not only as a recipient of Theo's potential complaint-to-be but also as a prospective co-complainant. As Zarah initiates a disagreement asserting that she does not think that it is cold in Launève, Malia continues to embody 'being cold' by lifting her hands to her mouth as if heating them up (line 9). In the 21 omitted lines, Malia counters Zarah's disagreement by pointing to the many layers of clothing that Malia is currently wearing.

### 90 *The structural organization of L2 complaints*

In line 33, Malia assesses the temperature as 'too cold'. Mariana agrees (line 34) and initiates an expansion with a negative assessment of the fog (lines 35, 37). After a repair sequence targeting the word *brouillard* ('fog', lines 37–43), Mariana produces the non-lexical vocalization *PFFFfff* accompanied by headshakes (line 44) to express her afective negative stance (Baldauf-Quilliatre, 2016) toward the weather conditions. In response, Malia ofers the high-grade negative assessment *c'est horrible*  ('it's horrible', line 45), by which she aligns and afliates with Mariana. Mariana lexically repeats Malia's turn (line 46) but elongates the second syllable so as to upgrade the assessment and claim her own epistemic independence. As Malia continues to express her agreement (lines 47–48), Mariana shrugs her shoulders, lifts her hands to the sides, and shakes them palms up (FG.5.14), further showing her frustration (and perhaps inability to do something about the situation, see Kendon, 2004; Marrese et al., 2021) about the fog until she drops her hands hard on her lap while uttering another non-lexical vocalization (line 51) to close her embodied expression of negative stance.

Malia expands the sequence by adding another piece of 'evidence' in support of the complaint: There is no sun, nothing (lines 52–53), while, in fact, they need vitamins (line 56; supposedly Malia refers to getting vitamins from the sun). By vividly gesturing to her sides with both hands palms up (FG.5.15) and then dropping them on the table, Malia indexes the obviousness of the claims and upgrades her afective involvement in the complaint to justify its legitimacy (cf. Selting, 2010a, 2012). Mariana, in turn, invokes the humidity involved in the foggy weather (line 55) and crosses her arms and shrugs (FG.5.16) to embody her reaction to the weather conditions, and she then suggests that she does not know whether she is in Launève or somewhere else (lines 60–61; see embodied turncompletion following *ou::* in line 61, FG.5.17). After a laughing response from Malia (line 62), Mariana reissues her turn but also this time embodiedly completes it with a hand gesture (line 63, FG.5.18) while Malia ofers the candidate *ou le pôle* ('or the pole', line 65) plus a stretched-out hand above the head probably suggesting 'north pole' (FG.5.18). Using limited linguistic means, Malia and Mariana thus co-construct an account for their complaint about the weather in Launève by comparing it to the weather on the north pole. Thereafter, they continue to express their disapproval and negative stances and further develop the sequence (line 68 and onward).

In sum, this excerpt has illustrated an 'early' version of what occurs more frequently at the upper-intermediate/advanced level – namely, the construction of joint complaints. Starting with a negative assessment of the temperature by one participant (Theo), the sequence developed into a joint complaint (by Malia and Mariana) about the weather through the coparticipants' second assessments, expansions with afliative negative stance expressions and evidence supporting the overall complaint,

and the joint construction of an account that worked to legitimize the complaint. Gestures served important purposes in the upgrade of assessments (see also Section 6.1) and in the production of the account. Through these actions, the participants demonstrated their epistemic independence and agency as co-complainants rather than merely complaint recipients (as in Excerpt 5.7). The fact that the complaint is about the weather is not insignifcant in this context, I would suggest, as this is something to which all participants have epistemic access and entitlement (Heritage & Raymond, 2005; Stevanovic & Peräkylä, 2012) to speak about and which is a recurrent conversation topic that seems to be interactionally available also to very early-stage speakers (see also Section 7.1). The excerpt thus demonstrates that even if it is uncommon in my data, under the right conditions elementary speakers may produce joint complaints. As I show in the next section, with time speakers increase their ability to build upon and synchronize their actions with others, which results in a higher level of co-construction and more joint complaints.

#### *5.3.2 Upper-intermediate/advanced speakers*

All upper-intermediate/advanced speaker complaints eventually lead to affiliative or sympathetic responses from coparticipants, although in four (or 11%) of the 36 sequences only limited displays of affiliation occur before complaint closure. Coparticipants regularly offer clear displays of affiliation throughout sequences and contribute actively to the construction of the complaints by, for example, offering negative assessments and supporting accounts. In 14 (or 39%) of the sequences, the complaint develops into a joint complaint. Excerpts 5.9 and 5.10 illustrate the higher level of co-construction of complaint sequences by upper-intermediate/advanced speakers *vis-àvis* elementary speakers. In these complaints, the focal participants (Malia, Cassandra) contribute to the construction of joint complaints by aligning and adding elements to the coparticipants' first negative stance expressions.

Excerpt 5.9 shows a joint complaint by Javier (JAV) and Malia about some of the professors at their current university. The participants have been talking about an interview that Malia is going to have at a diferent university in the city of Baleux, and Javier is comforting Malia by telling her not to be afraid (line 1) because the professor who will interview Malia is a person (lines 2–3) and not a monster (line 5). He contrasts this with some of the professors in Launève, who are *difciles* ('difcult', line 9), after which he ofers a more clearly positive assessment of the professors in Baleux as *plus ouverts* ('more open', line 12). Malia agrees with *oui c'est vrai*, ('yes it's true'), which she repeats (lines 14, 16).

### 92 *The structural organization of L2 complaints*

*Excerpt 5.9* 'The opposite' (Mer1\_2017–11–22)

In line 18, Javier expands the sequence with another negative assessment about some (of the professors) 'here'; the assessment is embodiedly completed through raised eyebrows and headshakes (line 18, FG.5.19), and Malia agrees (line 19). Next, Javier again expands, frst by suggesting that Malia knows who he is talking about (line 20) and then with a further negative portrayal of the professors. He pulls his hands up-down high over his head while asserting that *ils croient qu'ils sont* ('they think that they are', line 22, FG.5.20). Malia agrees with a prosodically stressed *oui:* ('yes', line 23) and thereafter ofers a more extensive contribution to the sequence (possibly in response to Javier's suggestion that Malia knows to whom he is referring). She assesses the situation described by Javier as *pas juste*, ('not fair', line 26) because the university is so small (line 27). Through this assessment and account, Malia emphatically upgrades Javier's third-party criticism and invokes a moral dimension to the reproachable behavior, suggesting that the professors' conduct is not justifable according to "normative standards of conduct" (Drew, 1998: 297) and thereby complaint-worthy. By doing so, Malia transforms the criticism made by Javier into a joint complaint.

Javier, in turn, agrees and builds on Malia's account by repeating it and suggesting that the university should allow for closer or more informal relationships between students and professors (lines 28–29, 31). Malia agrees before the end of Javier's turn (line 30), and by tilting her head and opening her right hand with her palm up (FG.5.21), she underlines the obviousness of the situation. She thereafter provides a contrast to Javier's claim about how it should be by asserting that the reality is exactly the opposite: *mais (.) c'est (.) exacteme:nt (.) contraire°(ment)°*  ('but it's exactly opposite', line 33). The prosodic delivery of this turn, with micro-pauses between every word, and the fast fipping hand gestures (FG.5.22) further enhance the afective strength of the negative contrast. Javier collaboratively completes Malia's turn with *le contraire* ('the opposite', line 34) in partial overlap with Malia, displaying his alignment and afliation (Lerner, 2013). Malia's assertion shows the participants' joint understanding of Javier's embodiedly completed assessment in line 22 about the concerned professors, and it summarizes the gist of the complaint about the professors being the opposite of familiar/informal (i.e., formal and 'above others', as implied by Javier's gesture above his head in FG.5.20). While Javier initiates an expansion (line 37), the activity is interrupted and later abandoned as Malia orients to her glasses that lay on the table and came in the way of Jordan's action of reaching for a paper sheet (lines 36–37).

Although this sequence starts with a series of criticism of a third party (or group of third parties) by only one participant, it develops into a joint complaint. After frst having merely ofered tokens of alignment, Malia joins the complaint as a co-complainant by adding new elements to the sequence in the form of a negative assessment invoking moral unfairness and an account on which Javier builds to expand the complaint. The coparticipants coordinate and fnely synchronize their contributions to show alignment and afliation, build on each other's turns, and anticipate others speakers' upcoming assessments before their verbalization.

The next complaint (Excerpt 5.10) emerges in the context of a discussion of the French grammar rules for adjectives and, specifcally, colors. It is Xiang who frst formulates a problem about the difculty of getting it right in *dictées* ('dictations'), an activity she did earlier the same day in her French course. In lines 2–8, she asserts that the problem is that if you have misunderstood (the rules for infecting colors in French), you easily accumulate 20 errors (on the dictation). At the end of her turn, she utters a sigh expressing her frustration about the situation (line 9). Cassandra then intervenes to ratify the topic as a complaint-worthy one and contribute to the construction of a joint complaint.

### 96 *The structural organization of L2 complaints*

*Excerpt 5.10* 'The color' (Mer2\_2016–11–16)


98 *The structural organization of L2 complaints* 

Cassandra frst expresses her own fear of colors appearing in the dictations (lines 10–11). She then produces an explicit criticism of a third party, likely the instructors responsible for the dictations: *ils font exprès*  ('they do it on purpose', line 14). With this criticism, Cassandra invokes not only the wrongdoings of the third party but also suggests that the transgression is deliberate and, therefore, all the more complaint-worthy

(Drew, 1998). In partial overlap, Xiang herself ofers the high-grade negative assessment *ça c'est vraiment difcile* ('that it's really difcult', line 15), with which Cassandra agrees (line 16). As Xiang initiates a report that will detail the complaint, Cassandra produces a short sigh (line 18), further displaying her own frustration about the issue.

Xiang expresses her difculties with learning 'the words' (line 20) and gives a specifc example of her disappointment earlier the same day (lines 23–31). Cassandra shows her alignment through small agreement tokens and continuers timed precisely with the transition places of Xiang's TCUs (lines 22, 27, 30). When Xiang approaches an upshot formulating the consequence of her report so far (projected by *alors*, 'so', in line 28 and completed in lines 33–34), Cassandra agrees and claims her own epistemic access to the experience: *mais je sais* ('but I know', line 32), followed by the high-grade negative assessment *c'était horrible d'étudier ça* ('it was horrible to study that', line 32), thereby aligning and afliating with Xiang while asserting her epistemic independence *vis-à-vis* the complainable. Xiang subsequently expands the sequence further, underlining the burdensome nature of the task by asserting that she was unable to learn something that she had already learned before (lines 36–37). Cassandra again strongly agrees (line 38), and as Xiang initiates what seems to be a summary assessment of how it all makes her feel (lines 39–40), Cassandra joins the complaint more actively as a co-complainant by expanding on her own experience studying the same thing.

Using an extreme-case formulation, Cassandra asserts that it was not just horrible to study the particular subject; it was the worst thing she has ever studied (line 42). At the end of her turn, she looks straight at Xiang with an upset expression (FG.5.23), showing a strong afective stance (Selting, 2010a, 2012). Xiang laughs in response (line 45), and Cassandra insists on the veracity of her assertion with *je te ju:re* ('I promise you', line 46), upgraded with prosodic stress on *jure*. She then develops a report of how she tried to memorize the words but could only do it fve times because it was so horrible (lines 49–54). While this report accounts for her assertions, it also supports the participants' joint complaint about the difculty of the French grammar rules. Cassandra ofers a detailed report of the study process, and her use of a three-part listing format (lines 60–63, see Jeferson, 1990) helps portray the situation as a long and arduous process. In overlap with the end of Cassandra's listing, Xiang intervenes by adding another element to the complaint, which she accounts for with a telling (lines 63–66 and onward).

Similar to Excerpt 5.9, this complaint becomes a joint accomplishment in which two participants fnely synchronize their respective contributions by ofering strong negative assessments and afective stance expressions, accounts, and other types of evidence in support of the overall complaint at precisely ftted moments in the interaction. Through epistemic claims such as *je sais* ('I know'), *je te jure* ('I promise you'), and retellings of past experiences, the participants construct their own independent epistemic access to the complainable while also supporting the joint complaint. The complaint involves an escalation of afectivity (Rääbis et al., 2019),

## 100 *The structural organization of L2 complaints*

whereby the speakers progressively upgrade displays of emotive involvement (as seen, for instance, in Cassandra's extreme-case formulation with accompanying embodied conduct in lines 42–43). The sequence is representative of complaints at the upper-intermediate/advanced level that concern issues to which more than one participant has epistemic access.

In sum, this section has documented a diference over time (with Aurelia and Malia) and across profciency levels (elementary vs. upper-intermediate/advanced speakers) in the degree of co-construction of complaints. This is visible in the ways in which complaints are responded to by coparticipants, with a larger proportion of sequences with (clear) afliative displays from coparticipants and joint complaints at the upper-intermediate/ advanced level than at the elementary level. In the following section, I relate these observations to the development of L2 IC.

#### **5.4 Discussion**

This chapter has sought to uncover systematic ways in which L2 speakers of French organize indirect complaints and how this organization changes over time. Comparing elementary and upper-intermediate/advanced speaker complaints, the analysis has revealed that the overall composition of complaint sequences (i.e., the basic building blocks) remains largely the same over time. In contrast, notable diferences exist in speakers' methods for moving into complaining and in the degree of co-construction of complaint sequences.

Section 5.1 identifed a 'sequence candidate' structure (cf. Jeferson, 1988) of complaint activities that consists of several recurrent actions. Complaint sequences minimally include a subtle or overt expression of a potential complainable, a detailing of the reproachable conduct or the complainable situation (often in the form of storytellings), a summary assessment, restatement, or formulation of the complaint (typically through an explicit high-grade negative assessment or a negatively valenced idiomatic expression), a recipient response that acknowledges the complaint *as* a complaint, and some kind of closing. In many cases, complaints are expanded beyond the minimal sequence structure, with additional stance expressions, tellings, or reports that further express the complaint-worthiness of the complaint to enhance the chances of obtaining afliative or sympathetic responses. The identifed sequence candidate resembles the structural observations made by Traverso (2009; see also Ruusuvuori et al., 2019) and seems to refect a shared understanding across several languages and cultures of complaining as an interactional project accomplished through diferent interactional tasks. In terms of the development of L2 IC, this observation supports the idea that L2 speakers bring certain aspects of IC with them from the L1 into the L2, such as a basic understanding of the generic organizational principles of interaction (Scheglof, 2007) or of the core components of some interactional

activities. This does not mean that speakers can directly transfer the L1 practices to the L2; instead, they have to 'recalibrate' certain aspects of their general IC to ft the L2 (Pekarek Doehler & Pochon-Berger, 2015). Such a recalibration process can be seen in the way in which complaints are initiated and in the social coordination involved in co-constructing complaints.

The proportionally more varied complaint initiations and the more elaborate pre-complaint work observed at the upper-intermediate/ advanced level testify to diversifcation over time of interactional methods for initiating complaints. Concerning the position of complaint initiations, elementary speakers' tendency to initiate complaints in frst position and the more advanced speakers' proportionally more diverse initiations (with more complaints initiated in response to questions and as upgrades of frst assessments/stance expressions) concur with a change in the overall organization of the conversations. The elementary-speaker conversations, especially at the beginning, typically take a 'round robin', one-speaker-at-the-time format, with many longer tellings, reports, and so on by one speaker with little intervention from coparticipants, whereas the upper-intermediate/advanced-speaker interactions adopt a more conventional turn-taking format. This longitudinal change is similar to what Sert (2019) documented for EFL peer interactions. With time, the participants increasingly ask each other questions, ofering opportunities for their coparticipants to initiate complaints in second position, and they volunteer more substantial interactional contributions. Complaint initiations through upgrades of other speakers' negative stance expressions presuppose fne synchronization of one's action with those of coparticipants, something that requires speakers to anticipate the syntactic and sequential trajectories of ongoing talk and to produce suitable 'seconds' (e.g., upgraded assessments) in a well-timed manner. My observations suggest that this implicates conversational and linguistic skills that are available primarily at higher L2 profciency levels (but see the 'early' example in Excerpt 5.8).

The fnding about an increase in second-position complaints over time may seem contradictory to some of the previous observations about the development of L2 IC. Berger and Pekarek Doehler (2018) found that their focal participant at frst mostly opened storytellings in second position and, with time, increasingly ofered frst-position stories. Their fndings concur with the idea that it is easier for speakers to initiate tellings in second position since openings in second position require less prefatory work than frst-position tellings. The discrepancy between their observations and mine may be due to the diferent participant frameworks, however. Berger and Pekarek Doehler investigated interactions between an L2-speaking *au pair* and the L1-speaking host family parents – a setting with more asymmetrical participant relationships both in terms of linguistic resources and institutional roles. The focal participant's increasingly more frequent frst-position stories may be due not only to her developing interactional abilities but also to her growing agency within the family. In my data, a similar concurrent change occurs, but considering the different participant framework (with L2 peers), the result is the opposite. As mentioned in Section 4.3, most of the focal participants in my study (particularly Aurelia and Malia) were energetic interactants who already, as elementary speakers, took active responsibility for advancing the conversations in their respective groups – and this can be seen in volunteered status updates leading to complaints among these speakers. With time and at higher profciency levels, other participants took similar active responsibility for maintaining progressivity of talk by building upon other speakers' stance expressions and asking each other questions, in part leading to more frequent second-position complaints.

In terms of practices for initiating complaints, the longitudinal analysis reveals a diversifcation over time in speakers' methods for moving into complaining. Although both elementary and upper-intermediate/advanced speakers sometimes launch complaints in straightforward, unmitigated ways, more advanced speakers deploy a more varied set of practices than elementary speakers to initiate complaints in ways that index the contingent and sometimes delicate nature of the activity (arguably, complaints about inanimate objects or situations may not be as delicate as complaints about specifc individuals, especially in a peer setting). More frequently than elementary speakers, they carefully prepare the grounds for their complaints by doing interactional work to build their credibility as complainants and account for the complaint-worthiness of the situation before explicitly formulating the complaint. The result is a more subtle, stepwise entry into complaining (Ruusuvuori et al., 2019) that displays a high level of sensitivity to the interactional context.

This longitudinal development pertains both to a change in sequence organization and to a diversifcation in the participants' use of linguistic resources. Whereas the initiations of elementary speakers sometimes include some kind of brief prefacing and delay before the verbal expression of a negative stance (such as the contrastive 'praise-but' formulations), the initiations of more advanced speakers regularly include longer pre-sequences in which the speaker only hints at the complainable situation. To accomplish such pre-complaint work, upper-intermediate/ advanced speakers deploy a range of linguistic resources for projection, such as conditional clauses, specifc multi-word expressions, impersonal constructions, 'frst verb' constructions, pseudo-clefts, and left-dislocations that allow them to secure a longer turn and provide elaborate framing information before overtly launching the complaint (see also Skogmyr Marian, 2021b). The fndings concur with prior work on L2 speakers' ability to accomplish delicate and dispreferred actions, such as disagreements (Pekarek Doehler & Pochon-Berger, 2011) and requests (Al-Gahtani & Roever, 2012; Youn, 2015), which has shown that L2 speakers over time diversify their interactional methods for accomplishing these actions with an orientation to the delicate or dispreferred nature of the action and in ways that are better adapted to the interactional context and the recipient.

Another observed diference between elementary and upper-intermediate/advanced speakers pertains to responses to complaints, afecting the degree of co-construction of sequences. The fact that most complaints eventually lead to some exchanges of afliation and/or sympathy shows that speakers at all profciency levels can successfully participate in complaining to some extent. But considering the interpersonal dimensions of indirect complaining and specifcally of joint complaining (see Chapter 3), speakers' ability to exchange overt displays of afliation and engage in joint complaints may have important interpersonal consequences (cf. Boxer, 1993, Günthner, 1997; Hanna, 1981; see also Berger & Fasel Lauzon, 2016, on the social-relational potentials of reciprocal emotion displays). My fndings suggest that the successful co-construction of complaints relates both to speakers' ability to introduce and format complaints in recognizable and context-sensitive manners and to their capacity to respond afliatively in ways that promote sequence expansion.

The observations about upper-intermediate/advanced speakers' ability to build up their complaints progressively may, in part, explain why certain complaints recruit afliative responses more efcaciously than others. But as indicated by Boxer's (1993) fndings and mine alike, it seems that L2 speakers at lower profciency levels have difculties responding to complaint-initiations in ways that favor the joint development of complaints. The higher level of co-construction of complaint sequences and the more frequent joint complaining among more advanced speakers testify to these speakers' growing ability for social coordination – that is, to build upon and synchronize their actions with others. Participants who wish to contribute to a sequence with more than minimal response tokens need to anticipate the syntactic and sequential trajectories of the ongoing talk and be able to instantaneously build on these in the formulation of the response. This, of course, necessitates particular linguistic skills in the L2, such as knowledge of syntactical structures and mastery of certain vocabulary. The joint weather complaint in Excerpt 5.8 showed some ways in which elementary speakers may contribute more actively to the sequential development despite limited linguistic means, such as through modifed repeats (Stivers, 2005; see also Berger, 2016, for L2 speakers' use of repetition to participate in multi-party interactions). The more advanced speakers' larger set of linguistic resources and their enhanced ability to produce timely second assessments thus seem to aford more possibilities to participate in joint complaining (see also Section 6.1).

The fndings about co-construction of complaints resonate with observations about L2 speakers' increasing capacity to accomplish humor in interaction (Skogmyr Marian et al., 2017), another conversational activity

# 104 *The structural organization of L2 complaints*

that relies on speakers' ability to fnely synchronize their actions on a syntactic and sequential level. They also relate to research on L2 speakers' growing practices for providing relevant recipient responses (e.g., Dings, 2014; Ishida, 2011; Sert, 2019), which shows that L2 speakers over time tend to diversify and more fnely synchronize response tokens and more elaborate contributions such as collaborative turn-completions. More generally, the fndings about increased co-construction of complaints highlight the fundamentally co-constructed and socially distributed nature of L2 IC (Hauser, 2019; He & Young, 1998; Kasper & Wagner, 2014), whereby specifcally the longitudinal development of interactional organization constitutes a joint accomplishment (Greer, 2019).

# **Note**

1. Excerpts 5.3–5.5 are part of the following article and have been reprinted by permission of the publisher, Taylor & Francis Ltd., www.tandfonline.com; Skogmyr Marian, K. (2021). Initiating a complaint: Change over time in French L2 speakers' practices. *Research on Language and Social Interaction*, *54*(2), 163–182, doi: 10.1080/08351813.2021.1899709. Date of publication: 10 May 2021.

# **6 Interactional resources for complaining in the L2**

This chapter examines interactional resources L2 French speakers deploy for constructing 'complaint-worthiness' and for participating in complaint sequences, and documents change over time in the use of these resources. As presented in detail in Section 3.3, complainants draw on various multisemiotic resources, practices, and actions to display a negative stance and show that their complaints are worth complaining about. I refer to these as *interactional resources for complaining* in a general sense. The chapter zooms in on two central resources – namely, (1) negative assessments (Section 6.1) and (2) direct-reported speech and reenactments (Section 6.2). The focus on these interactional phenomena ofers an opportunity to document changes in L2 speakers' use of precise linguistic and embodied resources over time, contributing to a better understanding of the development of an L2 'grammar-for-interaction' (Pekarek Doehler, 2018). I conclude with a discussion of the fndings in terms of the development of L2 IC (Section 6.3).

#### **6.1 Negative assessments**

Negative assessments1 and other expressions of negative stance are central for complaining. Speakers deploy these resources to show that complainables are worthy of complaining, to pursue complaints after insufcient displays of afliation from coparticipants, to display afliation and sympathy with the complainant, and to join other speakers' complaints (see Section 3.3). In this section, I compare the elementary and upperintermediate/advanced focal participants' use of negative assessments and non-verbal (vocal, embodied) expressions of negative stances. I show a diference over time in these speakers' lexico-syntactic practices for proffering negative assessments:

• At the *elementary level (A1–A2)*, speakers deploy a limited repertoire of linguistic formats for accomplishing high-grade frst assessments and for upgrading frst assessments. This concerns both the variety of assessment adjectives and assessment intensifers and, to some extent, the grammatical format of assessment turns. These speakers also exhibit high reliance on non-linguistic resources for displaying a negative stance, with stand-alone non-lexical vocalizations constituting an important resource for negatively assessing.

• At the *upper-intermediate/advanced level (B2–C1)*, speakers deploy a wider range of linguistic resources for assessing than at the elementary level, observable in more diverse lexical and syntactic formats for profering high-grade frst assessments and upgraded second assessments. Although these speakers also deploy non-verbal resources for assessing, they rely less on stand-alone non-lexical vocalizations and other non-verbal means to perform the same kind of evaluative work that elementary speakers do with these resources.

The diference over time has interactional consequences for speakers' participation in complaint sequences, particularly in the context of joint complaining. I now illustrate these diferences by means of empirical examples at the elementary and upper-intermediate/advanced levels.

# *6.1.1 Elementary speakers*

As shown in Chapter 5, both frst and second assessments (Pomerantz, 1984) occur abundantly in complaints and are used by both complainants and recipients. Among elementary speakers, second assessments are quite rare, however. The bulk of this section, therefore, focuses on frst assessments while briefy addressing second assessments and nonlinguistic assessments toward its end. Elementary speakers' high reliance on non-linguistic resources for displaying a negative stance generally will be demonstrated throughout the section.

# *First assessments*

Already from the beginning of their participation in the recordings, elementary speakers normally produce assessments that are linguistically marked as negative assessments. An exception to this is the least profcient speaker, Suresh, who, in the few complaint sequences in which he participates sometimes, produces assertions that are linguistically formatted as neutral but which are treated by his coparticipants as negative assessments (see Section 7.1). Assessment turns hence need not be linguistically formatted with negatively valenced components to be understood as expressions of negative stance. Most assessment turns at the elementary level are, however, linguistically marked as carrying negative valence. A distinctive characteristic of the elementary level assessments is the recurrent use of specifc linguistic formats, particularly the precise assessment *c'est très difcile* ('it's very difcult').

Excerpt 6.1 illustrates three uses of *c'est très difcile* ('it's very difcult') in the same sequence, deployed by both the complainant (Mariana) and one of the coparticipants (Malia). It also demonstrates the highly multimodal nature of many assessment turns, whereby speakers assemble verbal and embodied resources to form multimodal 'packages' (see, e.g., Goodwin, 2007; Hayashi, 2005; and Kärkkäinen & Thompson, 2018, and several contributions in Pekarek Doehler et al., 2021) for negatively assessing. Before the start of the excerpt, Malia told the coparticipants about her experiences learning French. In line 2, Mariana ofers a negative assessment of her own situation, and she initiates a complaint about her difculties with French pronunciation.

```
Excerpt 6.1 'Practice pronunciation' (Mer1_2016–10–19)
```


Mariana's turn-initiation *pour moi* ('for me', line 2) and her selfpointing frame the upcoming talk as pertaining to her own situation. After the frst part of the assessment turn, *c'est* ('it's'), she does two horizontal beats with her right hand before letting the hand fall on the table while shaking her head slightly and leaning forward (FG.6.1). The embodied conduct foreshadows the upcoming negative assessment *très difcile* ('very difcult') (cf. Ruusuvuori & Peräkylä, 2009). Following this, Mariana reports that she has a proposal to make (line 3), and she initiates a telling about her professor's instruction that she needs to practice the phoneme /ʒ/ (lines 7–10, 13, 15). Through a loud and stretched *pf*-sound (Baldauf-Quilliatre, 2016) and a repeated, enlarged version of the same embodied conduct as before (larger horizontal hand gestures, larger headshakes, FG.6.2), Mariana again expresses her negative stance and afect. This embodied conduct too precedes a negative assessment turn, and Malia's afliative assessment *c'est très difcile pour toi* ('it's very difcult for you', line 22) testifes to her anticipation of this. Mariana's assessment comes in overlap, and although it lexically repeats her prior assessment, this time she produces it with elongated vowel sounds (line 23) that upgrade the strength of the negative valence and show heightened afective involvement (Ogden, 2006; Selting, 2010a, 2012). Thus, in this excerpt the assessment turn *c'est très difcile* ('it's very difcult') is used three times by two diferent speakers to convey the speaker's own difculties and to express afliation with a coparticipant. The accompanying embodied conduct is reproduced (by the complainant) in both cases, just prior to the verbal assessment (segment) (see Scheglof, 1984, on pre-positioned gestures). This conduct hence foreshadows the upcoming negative assessment and shows Mariana's orientation to the verbal and embodied elements as belonging to the same multimodal package. Prosody and more amplifed embodied conduct furthermore work as resources to upgrade a lexical repeat of a frst negative assessment.

The recurrence of the negative assessment *c'est (très) difcile* ('it's (very) difcult'; sometimes *très* is absent) among elementary speakers can in part be explained by the fact that these participants frequently talk about the challenges of learning French and of being newcomers in Switzerland and at their workplaces – topics that warrant talk about difculties. Furthermore, research on assessments by L1 speakers of English (Goodwin & Goodwin, 1992) and of French (Pekarek Doehler et al., 2015) has shown that assessment turns very often take the format [neutral third person singular pronoun/pronoun clitic] + [copula] (+ [adverbial intensifer]) + [assessment term], such as *it's very good* in English or *c'est très bien* ('it's very good') in French. The high frequency of the precise lexico-syntactic string *c'est (très) difcile* at the elementary level is nonetheless noteworthy: 21 (or 26%) of the 81 assessment turns are initiated with *c'est* ('it is') and contain the assessment adjective *difcile* ('difcult'), most commonly with the intensifying adverb *très* ('very') preceding the adjective. A more in-depth look at the elementary speakers' use of the intensifer *très* reveals that this intensifying adverb is used in 39 (or 89%) of 44 tokens of intensifers. The two other intensifers used at this level are *trop* ('too much') and *plus* ('more'). Moreover, the lexico-syntactic format *c'est très* ('it's very') + [assessment adjective] is heavily recurrent, appearing in 28 or 35% of all 81 assessment turns (although primarily in the assessments of Suresh, Mariana, and Malia, in which the format appears in 46–51% of all negative assessments). It seems that with these speakers, the particular assessment adjective *difcile* ('difcult'), the intensifer *très* ('very'), and the precise lexico-syntactic format *c'est très difcile*  ('it's very difcult') serve as 'passe-partout assessments' for expressing negative stance at this particular moment in their L2 developmental trajectory.

Excerpt 6.2 provides another example of elementary speakers' use of *c'est très* ('it's very') + [assessment term] to ofer high-grade negative assessments. In this excerpt, Malia is telling Mariana about her first day as a PhD assistant. She has already told Zarah and Theo the same story (see Excerpt 6.11), but since Mariana came late to the conversation, she missed the first telling. When Malia told Zarah and Theo about her day, she characterized her experience as *un peu horrible* ('a bit horrible') because she thought that she would be able to speak English during her first year, but everybody spoke French to her. Instead of letting Malia elaborate on her troubles and offering her their sympathy, the coparticipants quickly suggested that it, in fact, will be good for Malia that her colleagues speak only French with her, thereby providing too early 'bright-side responses' (Holt, 1993). Malia's use of an upgraded version of the same assessment in Excerpt 6.2 (line 14) can be understood as a way to prevent receiving a similar unsympathetic response (Pomerantz, 1986) from her new story recipient:

*Excerpt 6.2* 'Started work B' (Mer1\_2016–11–02)2

Instead of describing her experience as *un peu horrible* ('a bit horrible'), Malia assesses her frst day at work as *très horrible* ('very horrible', line 14), upgrading the already high-grade *horrible* through the adverbial intensifer *très* and prosodic stress. The assessment segment is further highlighted with a horizontal hand gesture (FG.6.3). Through the assemblage of verbal and non-verbal resources, Malia thus frames her upcoming talk as strongly negative, involving a 'very horrible' experience. The high-grade negative assessment warrants an account, which Malia initiates (lines 17, 19, 27, and onward) after Mariana's display of astonishment (line 15, FG.6.4) and an objection from Theo (line 16). Considering the non-afliative responses Malia received when telling about the same experience to her other coparticipants just before, Malia's use of the high-grade negative assessment seems to work as a way to clearly disambiguate the nature of the telling already from the start to help Mariana anticipate the expected sympathetic response (similar to some story prefaces; see Section 5.2).

### *Second assessments*

In the context of complaints, speakers have been observed to use second assessments (Pomerantz, 1984) to show support for another speaker's negative characterization of a complainable (Rääbis et al., 2019), to receipt a display of sympathy from a coparticipant (Holt, 2000), and, in second complaints, to highlight one's own complaint-worthy situation in response to another speaker's frst complaint (Selting, 2012). Given this key role of second assessments in complaint sequences, speakers' ability to build on other speakers' frst assessments is hence crucial for their participation in the assessment and complaint activity (for participation

### 112 *Interactional resources for complaining in the L2*

in assessment activities, see Goodwin & Goodwin, 1992; Goodwin & Goodwin, 2000; and Pomerantz, 1984).

 The number of upgraded assessments in the elementary speaker complaints is low, and such assessments typically take the form of an exact or modifed repeat (Stivers, 2005) of the frst assessment. Thus, these speakers, only to a very limited extent, use *linguistic* means to produce an upgraded version of the frst assessment and instead resort to prosody (Ogden, 2006) or other non-linguistic means (such as non-lexical vocalizations). Excerpt 6.3, which reproduces the joint weather complaint presented in Excerpt 5.8, showcases these observations. The complaint is triggered by Theo's assessment of the temperature in Launève (line 4), which leads to a series of second assessments in the form of modifed repeats.

```
Excerpt 6.3 'Fog' (Mer1_2016–12–07)
```
As discussed in Section 5.3, Theo's turn in line 4 (*trè:s froi:d ici*, 'very cold here') is neutrally formatted, merely characterizing the temperature as *cold*  without indexing any positive or negative valence, but it is understandable as a negative assessment. Malia ofers an agreement (line 5) and the second assessment, *c'est très froid aujourd'hui* ('it's very cold today'), followed by the vocalization *ohhh* (line 6). She simultaneously fumbles a bit with her sweater sleeves as if trying to cover up her hands to protect them from the cold. Through the non-lexical vocalization and the embodied conduct, she adds an afective layer to her assessment (Reber, 2012) and indexes her negative stance toward the cold. The modifed repeat (Stivers, 2005) allows Malia to align with Theo while also asserting her epistemic independence (Heritage & Raymond, 2005). As seen in line 9, Zarah, too, recycles Theo and Malia's assessments, but in producing a disagreement: *c'est pas froid*  ('it's not cold'). In doing so, she format-ties her talk to both display relatedness with prior talk and express a contrasting stance (Goodwin, 1990).

Malia pursues her argument that it is cold by pointing to her many diferent layers of clothing (omitted lines). She then ofers an upgraded version of her (and Theo's) prior assessment: *c'est trop froid* ('it's too cold', line 33). The upgrade is done both lexically and prosodically, through the replacement of *très* ('very') with *trop* ('too') and through strong prosodic stress on both the intensifying adverb and the assessment adjective. Mariana agrees with Malia by nodding (line 34) and ofering another piece of evidence supporting the weather criticism, this time about the fog. The assessment of the fog as 'horrible' is done collaboratively with Malia over several turns, starting in line 35. After an overlap with Malia, who insists on the veracity of her high-grade assessment of the cold (line 36), Mariana restarts to introduce her opinion (line 37) and the assessable, the fog (lines 38–39). Following confrmations of understanding from the coparticipants (lines 41–43), Mariana produces a loud *PFFFfff* and headshakes showing her afective involvement and negative stance (Baldauf-Quilliatre, 2016) toward the fog (line 44). Malia, in overlap, initiates an assessment with *c'est ho- c'est horrible* ('it's ho- it's horrible', line 45). Mariana's *c'est horri::ble* (line 46) is delivered in overlap with Malia's assessment, but each element is slightly delayed. By elongating the last vowel sound, the main part of the assessment segment is produced in the clear and heard as a slightly upgraded version of Malia's already high-grade assessment. Ofering further agreement tokens (lines 47–48), Malia confrms the participants' alignment on the issue.

As mentioned in Section 5.3, joint complaints like the one shown in this excerpt are rare among elementary speakers. Since the accomplishment of joint complaints *as* joint complaints is closely tied to the participants' coordination of stance expressions (Rääbis et al., 2019), a possible reason for the few joint complaints at the elementary level is precisely the difculty of producing sequentially relevant and timely second assessments. This excerpt nonetheless exemplifes ways in which elementary speakers occasionally may accomplish both aligning and disaligning second assessments: namely, through modifed repeats – built through the addition or replacement of precise lexical elements (adverbials, negation markers) – and through prosody and embodied conduct. Whereas other-repeats have been observed as an important resource for low-level L2 speakers to participate in multi-party interactions (Berger, 2016), modifed repeats specifcally allow the participants to draw on the linguistic material of the frst assessments while simultaneously showing epistemic independence (Heritage & Raymond, 2005; Stivers, 2005).

# *Non-verbal displays of negative stance*

Besides using verbal assessment turns, speakers also produce assessments and display their stance non-linguistically. Although all speakers in my data, to some extent, deploy non-verbal (embodied, vocal) conduct to show negative stances, at the elementary level stand-alone non-lexical vocalizations with accompanying embodied conduct are particularly common means to index a negative stance or accomplish precise negative assessments (Goodwin & Goodwin, 2000; Wiggins, 2013). In Excerpt 6.4 (which shows the continuation of Excerpt 5.4), Mariana initiates a telling about her difcult experience at the bank, where she had to speak French, which made her very tired (line 11). Mariana's own display of negative stance and her coparticipants' responses rely heavily on embodiment and non-lexical vocalizations. *Excerpt 6.4* 'The bank' (Mer1\_2016–11–02)


In response to Mariana's announcement that she is very tired because she went to the bank, Theo and Malia each ofer non-lexical vocalizations of evaluative nature (lines 14–15). Theo's *o::h* (line 14), produced with falling intonation, treats Mariana's announcement as news (Heritage, 1984a) while simultaneously indexing a less than enthusiastic stance, whereas Malia's laugh particles and the following *o::::ps::* (line 15) signal a stronger negative afective stance. These response tokens serve as assessments, by which Theo and Malia show their recognition of Mariana's reported trouble and treat her announcement as newsworthy while also showing some level of afliation with her (Couper-Kuhlen, 2012). Mariana responds by producing a non-lexical vocalization herself, something in between a laughter and a sigh (line 16), which works like a sequence-closing third (Scheglof, 2007). In the meantime, Theo raises and waves his fsts as if fguratively embodying the 'fght' involved in going to the bank (FG.6.5), thereby embodiedly showing his engagement in Mariana's telling. Malia's laughter (line 17), on the other hand, orients to the humorous layer of Mariana's described situation, and her subsequent question whether Mariana's bank visit was in French (line 19) anticipates the reason for the reported trouble. In response, Mariana more explicitly expresses why the situation was difcult for her – because she did not understand (lines 22, 29), which Malia completes with the adverbial *rien*  ('nothing', line 31).

Following the coparticipants' laughter (lines 32–33), Mariana expands the sequence by reporting on the reason for her bank visit, that she had to pay the medical insurance (line 34). This information, too, is treated by the coparticipants as the expression of something negative, to which they display their sympathy. Malia ofers another non-lexical vocalization, *a::oups:*, with falling intonation (line 36), while Theo lifts and lowers both his hands palm up (FG.6.6) as if recognizing the helplessness of the situation (Kendon, 2004). Again, Mariana receipts these displays with her own non-lexical vocalization, a voiced sigh accompanied by lowered gaze and headshakes (line 39). Theo continues his embodied display of stance by lifting and lowering his shoulders (line 40), whereas Mariana ofers a verbal negative summary assessment of her telling: *c'est très mal* ('it's very bad', line 42), as she shakes her head and smiles, thereby showing her negative stance while simultaneously displaying some troubles-resistance (Edwards, 2005; Jeferson, 1984b). Following some laughter (lines 44–45, 47), Mariana closes her eyes (line 48) before gazing down and sighing again (line 49) as she leans forward and rests her elbows on the table (FG.6.7), further embodying the difcult situation and her tiredness. At this point, Malia asks Mariana a question about Mariana's obligation to pay the medical insurance, after which Mariana resumes her complaint (line 50 and onward).

# 118 *Interactional resources for complaining in the L2*

As shown in this section, elementary speakers rely heavily on a rather limited repertoire of linguistic resources for negatively assessing in complaints. The assessment segment *difcile* ('difcult'), the adverbial intensifer *très* ('very'), and the precise assessment turn *c'est très difcile* ('it's very difcult') recur frequently in these speakers' assessments. In addition, non-lexical vocalizations and embodied conduct are key resources for displaying both negative stance and afliation. Excerpt 6.4 exemplifed this by showing the reproduction of two very similar sequences: (1) the expression of a problem/difculty by the complainant, (2) afliative nonverbal (embodied, vocal) negative assessments by coparticipants, and (3) a non-verbal (embodied, vocal) sequence-closing third by the complainant. Although non-verbal conduct is also present in the complaints of upperintermediate/advanced speakers, more advanced speakers rely less so than elementary speakers on standalone non-lexical vocalizations to perform the same types of actions.

#### *6.1.2 Upper-intermediate/advanced speakers*

Upper-intermediate/advanced speakers deploy a considerably more varied repertoire of assessment formats than elementary speakers. A total of 41 diferent assessment adjectives and 8 diferent intensifers are used in 93 assessment turns, as opposed to 21 diferent assessment adjectives and 3 diferent intensifers in 81 assessment turns with elementary speakers (for type-token ratio comparison, see Section 6.1.3). The assessments of these speakers also more often take other grammatical formats than the canonical *c'est* (+intensifer) + assessment adjective, and they recurrently include phatic prefaces such as *je te jure* ('I promise/swear to you'). The diversifcation of interactional resources for assessing allows speakers to rely more heavily on *linguistic* means to do high-grade frst assessments and to adapt their second assessments to frst assessments in fne-grained ways.

# *First assessments*

Excerpts 6.5–6.7 illustrate the wider repertoire of assessment terms and adverbial intensifers found in the frst-position negative assessments of upper-intermediate/advanced speakers. Examples of new adverbial intensifers found at this level are *vraiment* and *tellement* (both approx. 'really'), which are common intensifers in ordinary spoken French. In Excerpt 6.5, Cassandra is complaining about her time as a student at a particular university institute. After producing a report about what always annoyed her when she did her bachelor's studies (lines 2–18), she ofers a high-grade assessment of her experience (line 19).

*Excerpt 6.5* 'Methodic' (Mer2\_2017–06–21)



### 120 *Interactional resources for complaining in the L2*

The assessment (line 19) takes a canonical format (*c'est*, 'it's' *+* intensifer + assessment term), with the intensifer *tellement* ('really') preceding the slowly delivered and prosodically stressed <*infexible:>* ('infexible'). Cassandra gazes at Xiang during the production of the assessment, and while uttering the adjective, she leans forward slightly and taps her right hand vertically on the table (FG.6.8) so as to upgrade and animate the assessment. Toward the end of the assessment, Xiang lifts and looks into her cup (line 19), refraining from responding. Cassandra then repeats the assessment segment (line 21), again prosodically stressing the adjective, and also repeats her embodied conduct in a more marked way by leaning forward more distinctly and redoing the hand gesture at a slower pace (FG.6.9). As Xiang still does not answer, but merely shows her attentiveness by maintaining her gaze at Cassandra (line 20), Cassandra pursues the complaint by expanding on her account (line 24 and onward).

The excerpt provides just one example of how the linguistic resources deployed for negative assessments diversify over time, with intensifers (*tellement*, 'really') and assessment adjectives (here *infexible*, 'infexible') that do not occur at the elementary level. Like what we have seen with elementary speakers (Excerpt 6.1), here Cassandra recycles the verbal, prosodic, and embodied aspects of the assessment segment while slightly augmenting the prominence of the embodied conduct in the repetition of the assessment, orienting to the assessment segment as a multimodal package comprising both verbal and embodied components.

 Another novelty in the frst assessments of upper-intermediate/advanced speakers is that they sometimes take comparative or superlative adjective forms that present the assessment as an *upgraded* or extreme-case assessment despite its sequential position as a frst assessment. Excerpt 6.6 is from the extended sequence in which Aurelia complains about Swiss society and the mentality and behavior of people in Switzerland (Excerpt 1.1). She has just told Mia about the incident in which a man stopped his car to reproach Aurelia and her friends for standing in the road. In lines 1–2, she produces what works both as a reported answer to the man and as a display of Aurelia's stance toward the event (that it was not so bad). Mia receipts Aurelia's telling with the change-of-state token (Heritage, 1984a) *a::h* and a *oui* ('oh yes', line 3) in low volume, by which she aligns with Aurelia without displaying any strong signs of afliation. Aurelia then expands the sequence to seek a more afliative response (line 6).

*Excerpt 6.6* 'Better inside' (Lun\_2018–02–26)



By claiming that the just-reported problem happens all the time (line 6), Aurelia generalizes her complaint so as to legitimize it (Drew, 1998; Mandelbaum, 1991). She then upgrades this generalization by producing a high-grade negative assessment about her *current* situation: *c'est pire maintenant* ('it's worse now', line 7) and by initiating an account (lines 7–9, continuation not shown). The assessment is done with the comparative form of the adjective *mauvais* ('bad') – namely, *pire*  ('worse'). With this assessment, Aurelia hence upgrades the severity of the described situation to add further support to her overall complaint and to pursue displays of afliation from her coparticipant (which she eventually gets).

Whereas the precise comparative adjective form *c'est plus difcile* ('it's more difcult') occurs three times in the negative assessments of elementary speakers (twice with Malia, once with Aurelia), there are no occurrences of other types of comparative (or superlative) forms of assessment segments in the elementary level data. Therefore, it seems that the more diverse negative assessment formats observed among more advanced speakers in part relate to these speakers' ability to produce comparative and superlative adjective forms.

Another longitudinal change pertaining to the grammatical formatting of negative assessments is that upper-intermediate/advanced speakers more often produce assessment turns with left-dislocations of the assessable (there is only one negative assessment with a right-dislocated assessable in the collection). Left-dislocations are frequently used in assessments in L1 French, especially in assessments of the format *c'est*  ('it's') + [assessment segment] (Pekarek Doehler et al., 2015). As demonstrated by Pekarek Doehler et al. (2015), left-dislocated referents receive interactional prominence and project the TCU as being primarily about this referent (the assessable). Although a few left-dislocations occur in the negative assessments already at the elementary level, these become more frequent with time. More advanced speakers' tendency to produce left-dislocations in their assessments indexes the increased similarity of their L2 grammar-for-interaction with those of L1 speakers and ofers interaction-organizational afordances for the expression of complaint-worthiness.

### 122 *Interactional resources for complaining in the L2*

Excerpt 6.7 illustrates this. Xiang has just criticized Liang (LIA) for not having produced the linking sound (*liaison*), and Liang admits that sometimes French speakers indeed have difculties understanding her when she omits the *liaison* (lines 1–2). In Liang's defense, Cassandra negatively assesses the Swiss (lines 4–5) and launches a complaint about them not making an efort to understand foreigners.

```
Excerpt 6.7 'Little efort' (Mer2_2017–03–29)
```
In the initiation of her turn (line 4), Cassandra frst deploys the pronoun *ils* ('they') but restarts to introduce the referent *les suisses* ('the Swiss'), thereby specifying that her upcoming talk refers to Swiss people (rather than French speakers in general). Inserting parenthetically *tu sais*  ('you know'), she prefaces the assessment with a phatic element that draws attention to the upcoming talk and serves to enhance the chance of receiving afliative responses (Fiedler, 2020). The assessment proper is introduced with another pronominal reference: *ils sont vraiment <nuls>*  ('they are really bad/stupid/ useless', lines 4–5). The self-repaired turninitiation with the left-dislocation hence allows Cassandra to insert and emphasize the correct referent of her upcoming assessment while parenthetically adding the phatic construction *tu sais* ('you know') before the assessment proper. The assessment is responded to with laughter (lines 7 and 9), after which Cassandra expands by claiming the generalizability of her assessment (line 10) and justifying this with an example (omitted lines).

A second left-dislocated assessment turn occurs in response to Xiang's receipt of Cassandra's example. Xiang asserts that the French language is very precise (lines 25–26), insisting on the importance of precision when speaking French and thereby subtly justifying her prior correction of Liang's pronunciation. Cassandra again objects, by recycling her prior assessment, frst in low voice: °*mais ils sont nu:ls hein°* ('but they are bad/ useless huh', line 28) and then preceded by short laughter, *eux aussi °<ils sont nuls hein>°* ('they too, they are bad/useless huh', line 29). Exactly what the disjunctive pronoun *eux* ('they') refers to here is not clear, but it may indicate Cassandra's understanding of Xiang's *le français* (line 25) as *les Français*, as in 'the French (people)'. The left-dislocation again permits Cassandra to reinsert and emphasize the assessable while lexically repeating the same assessment turn as just before.

### *Second assessments*

Upper-intermediate/advanced speakers produce second assessments more frequently than elementary speakers do, and these take varied lexico-syntactic forms. For example, speakers use superlative adjective forms, lexically fxed extreme-case constructions (Excerpt 6.8), as well as linguistically downgraded second assessments that work as upgraded assessments by virtue of irony (Excerpt 6.9). Aligning and upgrading assessments are key resources for constructing joint complaints and exchange displays of afliation between the coparticipants (see Section 5.3), and the more advanced speakers' larger linguistic repertoires for profering context-sensitive upgrades contribute to their increased ability to accomplish such afliative work.

In Excerpt 6.8, Aurelia uses a hyperbolic second assessment to upgrade an afliative frst assessment ofered in response to her own complaint. Aurelia has, in a longer sequence, complained about her prior fatmate (see Excerpt 5.5). In line 180, Jordan ofers a display of sympathy with Aurelia by suggesting that the last months were *pas facile* ('not easy') for her. The turn shows Jordan's understanding of Aurelia's telling-so-far and works like a 'candidate' summary assessment that presents the gist of Aurelia's difcult situation. Aurelia receipts this by confrming and upgrading Jordan's assessment.

### 124 *Interactional resources for complaining in the L2*

*Excerpt 6.8* 'Sharing' (Lun\_2018–05–28)


 Aurelia's turn-initial *phhf* (line 181) premonitors the negative valence of the upcoming turn (Baldauf-Quilliatre, 2016). She then recycles Jordan's *les derniers* ('the last', line 182), after which she restarts to confirm the accuracy of Jordan's assessment (*ouais*, 'yeah', line 182) but specifies, through a modified repeat (Stivers, 2005), that she is referring to the last *four* months. Aurelia again recycles Jordan's talk by repeating his assessment but inserting an intensifier: *c'était pas du tout facile* ('it was not at all easy', line 183), and she offers a stretch of reported speech (or possibly thought) in which she affectively supports her negative assessment (line 184). By recycling the format of Jordan's first assessment and inserting the intensifier *du tout* ('at all'), Aurelia offers an upgraded second assessment that both affiliates with Jordan's display of sympathy and strengthens the complaint-worthy nature of Aurelia's story – not only was the situation 'not easy', but it was 'not at *all* easy' – while also underlining Aurelia's primary epistemic access (Heritage & Raymond, 2005) to her experience. The construction *pas du tout* ('not at all') illustrates the more advanced speakers' more frequent use of certain types of formulaic language in the expression of negative stance, including specific idiomatic expressions and figurative language (see, e.g., Erman et al., 2016; Forsberg, 2008, on the late acquisition of some types of formulaic language in L2 development; see also Cassandra's *ils sont nuls*, 'they are bad/stupid/useless', in Excerpt 6.7).

Excerpt 6.9 exemplifes another type of resource deployed by more advanced speakers for making afliative second assessments: a linguistically downgraded assessment that, by virtue of irony, upgrades the frst assessment. Angelina has told her coparticipants about a recent visit to her Italian hometown. In line 1, she assesses the temperature in the house in Italy as *super froid* ('super cold'). Cassandra, who is also from Italy, will join her in complaining about the poor house construction in the country (note that Xiang's question in line 4 is directed to Angelina).

*Excerpt 6.9* 'No insulation' (Mer2\_2017–01–11)

Cassandra's second assessment in line 5 (*>un petit peu<*, 'a little bit') linguistically *downgrades* Angelina's assessment *super froid* ('super cold', line 1). The laugh particles preceding and following the assessment and the smiley voice during its delivery make it hearable as an ironic statement, however, which aligns and afliates with the frst assessment and gives a go-ahead for Angelina to continue (which she does by comparing the situation in Italy with the one in Switzerland; line 6 and onward). Cassandra's use of an ironic second assessment testifes to her fne-grained ability to adjust to the interactional contingencies: Angelina's frst assessment (line 1) is linguistically already marked as a hyperbolic, extreme-case assessment: *super froid* ('super cold'). The possibilities for linguistically upgrading such a high-grade assessment segment are therefore limited. Cassandra solves this practical problem in a resourceful way by mobilizing an interactional resource typically used for low-grade assessments (the downgrading construction *un petit peu*, 'a little bit') combined with paraverbal resources to accomplish an ironic statement that is recognizable as an upgraded assessment (cf. Rääbis et al., 2019). The excerpt hence provides another example of how more advanced speakers draw on their (more diverse) linguistic repertoires to produce negative assessments that are subtly tuned to the interactional context.

### *Verbally incomplete negative assessments*

As shown throughout the analysis, upper-intermediate/advanced speakers also use non-linguistic (vocal, embodied) resources to express negative stances. A diference between elementary and more advanced speakers is that the latter rely less on *stand-alone* non-lexical vocalizations than elementary speakers do to produce negative assessments and display negative stance. In contrast to stand-alone vocalizations, non-verbally completed negative assessments do not decrease from elementary to upper-intermediate/advanced levels, and this is congruent with the L1 literature that shows that verbally incomplete utterances are pervasive features of ordinary conversations (e.g., Chevalier, 2008; Hayashi, 2005; Keevallik, 2013; Li, 2016; Mori & Hayashi,

### 126 *Interactional resources for complaining in the L2*

2006; see also Skogmyr Marian, 2021a). Excerpt 6.10 (same complaint as in Excerpts 5.5/6.8) illustrates such a case. Here Aurelia has given a specifc example of a complaint-worthy behavior of her former fatmate – namely, that she, out of principle, would leave things, like a spoon, on the table for six months if she was not the one who had put it there (lines 98, 100). In line 101, Aurelia initiates a summary assessment of the complaint-so-far.

```
Excerpt 6.10 'Sharing' (Lun_2018–05–28)
```
During a long in-breath (line 101), Aurelia rolls her eyes, leans back in her chair, and starts lifting her hands in front of her (FG.6.10), embodiedly displaying a negative stance (Clift, 2021; Goodwin & Alim, 2010). She then initiates an assessment with *et c'est okay eu:h* ('and it's okay euh', line 101) while maintaining her shoulders and hands high, palms up (FG.6.11), showing her incomprehension about the situation (Selting, 2010a, 2012). Instead of verbally completing the turn, Aurelia produces the loud non-lexical vocalization *PHHFFffhhhuh* (line 101) as she lowers her hands and shoulders. The vocalization expresses an afective negative stance (Baldauf-Quilliatre, 2016), and together with the verbal initiation and the preceding embodied conduct, the action is recognizable as a negative assessment that is brought to completion with non-verbal means. Jordan displays his alignment by nodding (lines 101–102) but refrains from taking the turn, and Aurelia's subsequent *c'est un peu: ça m'a (coûté) un peu d'énergie quoi* ('it's a bit it (cost me) a bit of energy prt', line 103) works as a gloss (Keevallik, 2013) of the prior turn that ofers another occasion for Jordan to afliate. As Jordan again merely nods, albeit with larger head movements (lines 103–104), Aurelia continues with another account for her criticism of the fatmate (lines 105–106 and onward).

The practice of leaving one's negative assessment lexico-syntactically incomplete has been discussed in the literature as a way to convey a negative stance without verbalizing negative assessment terms (Chevalier, 2008; Li, 2016). Research on complaining has also shown that complainants regularly express the gist of complaints through idiomatic utterances after a descriptive telling (Drew & Holt, 1988; Ruusuvuori et al., 2019). In the present case, Aurelia deploys both of these practices by frst producing a verbally incomplete negative assessment and then glossing it through a non-literal expression. Through these actions, she showcases her ability to put to use specifc interactional (linguistic, non-linguistic) resources for context-sensitive interactional purposes in L2 interaction.

#### *6.1.3 Negative assessments: quantitative comparison*

A quantitative comparison of assessment adjectives and intensifers over time allows for observations about overall developmental tendencies (cf. Wagner et al., 2018, on the potential benefts of selective quantifcation). Importantly, the relevancy of such a comparison emerged from the initial qualitative analysis of the data. On the lexical level, upper-intermediate/ advanced speakers deploy a larger repertoire of assessment adjectives and intensifers than elementary speakers. In the 81 negative assessment turns3 of elementary speakers, 21 diferent assessment adjectives (71 tokens), and 3 diferent intensifers (44 tokens) are used. In the 93 assessment turns of upper-intermediate/advanced speakers, 41 diferent assessment adjectives (77 tokens) and 8 diferent intensifers (41 tokens) occur. Table 6.1 presents the type-token ratios (TTR) of negative assessment adjectives (NAA) and intensifers (INT) in the complaints at both profciency levels.


 *Table 6.1* Type-token ratios (TRR) of negative assessment adjectives (NAA) and


### 128 *Interactional resources for complaining in the L2*

 The type-token ratio of both negative assessment adjectives and intensifers is considerably higher for upper-intermediate/advanced speakers than for elementary speakers, going from 0.30 to 0.53 for assessment adjectives and from 0.07 to 0.20 for intensifers. These numbers show that L2 speakers proceed from a limited to a broader repertoire of lexical resources available for negatively assessing as they progress from elementary to upper-intermediate/advanced profciency levels, allowing them to increasingly diversify the lexical composition of their negative assessments (cf. Nguyen, 2019, who documented a diversifcation over time in *positive*  assessment segments in the interactional repertoire of a Vietnamese L2 speaker of English).

In terms of lexico-syntactic formatting, we see a similar pattern, with more advanced speakers deploying a larger set of lexico-syntactic formats in their negative assessments. The qualitative analysis revealed some syntactic formats found only or predominantly among upper-intermediate/ advanced speakers, such as superlative adjective forms and left-dislocated assessment turns, as well as prefacing and parenthetical constructions like *je te jure* ('I promise/swear to you') and *tu sais* ('you know'). These general observations hence support the idea that L2 speakers move from a limited to a larger set of lexico-syntactic formats for negatively assessing over time, allowing speakers to vary their high-grade assessments and upgrade frst assessments more easily to underline the complaint-worthiness of the complainable and actively contribute to joint complaints.

A quantitative analysis of speakers' use of the precise construction *c'est très* ('it's very') + [assessment adjective] sheds further light on the distribution of lexico-syntactic formats for negative assessments. This analysis was data-driven, warranted by the high recurrence of specifc lexical and syntactic resources for profering negative assessments among elementary speakers observed through sequential analysis, such as *c'est* ('it's') plus the intensifer *très* ('très', making up 39 of 44 tokens of intensifers). Table 6.2 shows the numbers and percentages of use of the lexico-syntactic string *c'est très* ('it's very') + [assessment adjective] in the negative assessments (NA) at elementary and upper-intermediate/advanced levels.

As seen in the table, the construction occurs in 35% of all assessment turns of elementary speakers (28 occurrences) and in only 1% of the assessment turns of upper-intermediate/advanced speakers (1 occurrence). Notable diferences between speakers also exist: While at the elementary level, Suresh, Mariana, and Malia deploy *c'est très* in 46–51% of their assessment turns, Aurelia uses the same construction in only 4% of her assessments. As upper-intermediate/advanced speakers, neither Malia nor Aurelia uses the precise construction in their negative assessments. As discussed, *c'est* + [assessment segment] is a highly frequent component of assessment turns in L1 French (Pekarek Doehler et al., 2015); the L2 speakers' use of the same format is therefore not surprising. But the frequent combination of *c'est* ('it's') plus the precise adverbial *très* ('very')

## *Interactional resources for complaining in the L2* 129

 among several elementary speakers and its decline over time indicate that this lexico-syntactic string constitutes a kind of *passe-partout*, 'standard solution' for the accomplishment of high-grade assessments among some speakers at the early stages of L2 learning (cf. Larsen-Freeman's, 2006, idea of 'make-do solutions' by which L2 learners dynamically put to use the linguistic resources at their disposal in response to ever-changing social contexts) and that the developmental trajectory involves diversifcation of action formats over time. In Section 6.3, I discuss the implications of these observations for our understanding of the development of L2 IC.


 *Table 6.2* Numbers and percentages of the construction *c'est très* ('it's very') in the negative assessments (NA) at elementary and upper-intermediate/ advanced levels

#### **6.2 Direct-reported speech and reenactments**

As presented in Section 3.3, research on reported speech and reenactments has shown that these often are used by complainants to depict complaintworthy behavior of others, portray themselves in a good light, and animate and show their afective involvement in complaint stories. Direct-reported speech (DRS) has been identifed as a particularly efective means for providing 'evidence' for the complaint because it allows the coparticipants to 'witness' the complained-about transgression themselves. As invoked by Couper-Kuhlen (1999), however, the introduction of DRS involves a number of interactional challenges for the speaker, such as conveying that reported talk is forthcoming, whose voice is being reported, and the interactional purpose of the reported speech. These observations open up for inquiries into L2 speakers' use and learning of DRS (for a Japanese ESL speaker's longitudinal use of DRS, see Hauser, 2013).

In this section, I analyze how and for what interactional purposes participants use DRS and reenactments in complaint sequences. While I included both indirect reported speech (IRS) and DRS in the original analysis, I focus particularly on the use of DRS in the following analysis since this type of reported speech is the most common in the data and has been identifed in the literature as a key resource for the construction of complaints. The distinction between DRS and IRS is nevertheless not always clear-cut (Clift & Holt, 2007), and this is apparent in some of the examples, especially in cases in which the participants encounter diffculties marking their upcoming reported talk as either DRS or IRS. The analysis also includes examples of reenactments (Sidnell, 2006) – that is, reports of (real or fctive) embodied actions. I only focus on instances of DRS and reenactments that convey a negative stance and thereby contribute to the construction of complaint-worthiness.

Speakers at elementary and upper-intermediate/advanced levels deploy DRS and reenactments for similar interactional purposes. At both levels, these episodes are highly embodied in nature, with speakers drawing on a range of verbal, paraverbal, and non-verbal resources to embody the reported characters and show an afective stance. In addition, the following diferences over time are observable:


I now present the qualitative fndings pertaining to elementary and upperintermediate/advanced speakers before briefy addressing some quantitative diferences.

### *6.2.1 Elementary speakers*

All elementary speakers except Suresh deploy reported speech and reenactments at some point in the data, although Aurelia and Mariana less frequently than Malia. Excerpts 6.11–6.13 demonstrate for what purposes these speakers use DRS and reenactments and illustrate typical formal properties of such resources.

# *Other-reported speech as concrete 'evidence' supporting complaint formulation*

Just like L1 speakers, the L2 speakers of my data recurrently use DRS as concrete 'evidence' for complaints – to *show* rather than *retell* past events and thereby increase the realistic character of the reported event and let coparticipants judge for themselves the complaint-worthiness of the situation. In Excerpt 6.11, other-reported talk serves to concretely exemplify the complainable. Before the excerpt, Malia assessed her frst day at work as 'a bit horrible' (see Excerpt 6.2, in which Malia retells the same story). To account for the assessment, she explains that she thought that she would be able to speak English in the frst year, but in fact, everybody started speaking French with her (lines 31–32). The DRS that follows provides specifc details supporting the claimed seriousness of the situation: Malia's professor even asked her colleagues to only speak French with her (lines 34–35, 41–42).

*Excerpt 6.11* 'Started work A' (Mer1\_2016–11–02)

Malia's introduction of the reported speech is somewhat problematic in that it leads to a repair sequence in the production of the quotative (the precise marker of DRS, see Clift & Holt, 2007) *demander* ('to ask') in past tense plus pronominal references (see hesitations and self-repair of the quotative in lines 34–35 and the other-completed repair by Theo in line 40). As she continues her turn, Malia frst ofers the subordinate marker *que* ('that') and a cut-of *je* ('I', line 41), which projects IRS to follow, but she self-repairs to DRS: *<seulement> parlez avec malia, (0.3) français*. ('only speak with Malia French', lines 41–42). Through a combination of slow delivery, prosodic stress, and horizontal hand gestures (FG.6.12–6.13), Malia animates the reported talk, portrays the professor's action as a strict order or directive rather than a request, and displays her own afective involvement in the telling (Selting, 2010a, 2012). The DRS works as the climax of the story (Holt, 2000), and Zarah's laughing response (line 44) shows recognition of this, but the

response is problematic in that it only attends to the non-serious layer of the telling and not the complaint itself. To recruit more appropriate responses, Malia lets her right hand fall on the table and lowers her gaze and her head in her left hand (FG.6.14) and produces the non-lexical vocalization *o:::h* (line 46), showing her despair, and she subsequently pursues the complaint by insisting on the complaint-worthiness of the situation.

This excerpt hence shows the use of DRS to provide a concrete, witnessable example supporting the complaint formulation. As seen in the excerpt, especially at the elementary level, the introduction of reported speech is regularly problematic, involving repair sequences and what Gardner (2007) refers to as 'broken starts': turn-beginnings with hesitation markers, pauses, and other types of self-repair produced as the speaker attempts to fnd appropriate linguistic material to construct the turn. In this case, Malia's mid-turn shift from IRS to DRS does not engender any apparent difculties for intersubjectivity, likely because of her use of marked prosody and embodied conduct together with the frst name self-reference (see Couper-Kuhlen, 1999, for an analysis of problematic introductions of reported speech). Instead, another interactional problem appears – namely, the recognizability of the DRS as the climax of a funny versus a troublesome story. The recurrence of DRS in story climaxes in both amusing and complaint stories (Holt, 2000, 2007), and the tendency of reenactments to create humorous efects (Sidnell, 2006) likely contribute to this problem, with the reported talk creating an ambiguity as to the expected responses from coparticipants.

# *DRS to support the portrayal of self as reasonable and of the third party as unreasonable*

Another use of DRS in complaint stories is to portray the complainant's own conduct as reasonable in the face of another party's unreasonable and complaint-worthy conduct. Speakers typically use marked prosody and embodied conduct to diferentiate between their own and the third party's behavior (Berger & Pekarek Doehler, 2015). Excerpt 6.12 illustrates such a use while again exemplifying a problematic turn-beginning and difculties with pronominal references. This sequence is in another complaint by Malia about speaking French at work. After receiving insuffciently afliative responses to her complaint-so-far, she expands the sequence with a retelling of an event that happened the same day, when she met her professor in the elevator and the professor told Malia that she soon needs to be prepared to teach courses in French and speak French well.

*Interactional resources for complaining in the L2* 137


Malia's turn in line 1 announces incipient IRS about something pertaining to the month of January: *elle m'a dit que: (.) en janvier* ('she told me that in January'). After some hesitation (line 2), Malia continues with what becomes recognizable as DRS. Offering the personal pronoun *tu:* ('you.sing', line 2) while pointing forward (FG.6.15), and then restarting with *elle m'a dit que tu* ('she told me that you', lines 2–3) with accompanying pointing first to the left at *elle* ('she', FG.6.16), at herself at *m'a dit* ('told me', FG.6.17) and a circling gesture with the index finger pointing forward (FG.6.18) at *que tu* ('that you', line 3), Malia embodiedly works to convey the person references in the reported speech. Having received a continuer in response from Mariana (line 4), Malia produces a loud in- and out-breath followed by a repetition of the pronoun *tu* ('you.sing', line 5) and a 'quotation mark gesture' with both hands (FG.6.19) followed by laughter, again highlighting the second person pronoun as belonging to the reported world and simultaneously orienting to her laborious efforts as a laughable. After clearer signs of understanding and some laughter from the coparticipants (lines 6–8), Malia continues the episode in direct-reported form (lines 9, 11), reporting how her professor told her that she needs to start teaching courses for the bachelor's students (lines 9, 11).

In the 25 omitted lines, Malia provides details about her upcoming teaching obligations (it is unclear whether this is part of the reported speech or not). In line 38, she clearly resumes the DRS with an upshot of the talk so far: Malia needs to be ready for the student questions and 'everything' (lines 39, 42–43). Through prosodic stress on key terms and by spreading her open hands far to the sides on *tout* ('everything', line 43, FG.6.20), Malia conveys the large scope and seriousness of her work obligations. So far, Malia has thus deployed DRS to provide a concrete example of the complaint-worthy obligations associated with her work. Next, she pursues the telling with evidence about the involved parties' personal characters, portraying her professor as a complaint-worthy person and herself as a legitimate complainant.

In lines 45–46, Malia steps out of the reported frame to ofer a metacomment (see Heinemann & Wagner, 2015) on the professor's embodied conduct during the reported talk – namely, that she said it with a large smile (FG.6.21) – thereby portraying the professor as taking enjoyment from telling Malia about her work obligations. Again changing footing (Gofman, 1981), Malia expands the reported talk with an embodied reenactment. Raising her eyebrows, pointing with her righthand index fnger, and nodding (lines 50–52, FG.6.22), Malia reenacts the professor giving her a directive: *tu dois parler bien °le français°*  ('you have to speak French well', line 52). The embodied conduct (see also Malia's stif body posture), which embodies stereotypical images of adults lecturing children, conveys the professor's strictness and stands in stark contrast to what follows – namely, Malia's reenactment of her own response.

With two cut-of syllables followed by a small °*mm°* in low volume and while nodding and pursing her lips frmly (line 55), Malia reenacts her difculties producing a response to her professor's frm directive, and she subsequently expands with a prosodically marked confrmation: *£janvier hhhhh* ↑*b(h)ien* ↑*sû::r(h)£* ('January, of course', lines 58, 60). Through this verbal and embodied conduct, Malia conveys her own compliant response and portrays herself as an acquiescent assistant in the face of a boss who takes pleasure in inficting painful experiences on Malia. The other- and self-reported speech thus works to contrast Malia's own reasonable conduct with the complained-about person's unreasonable acting and supports the legitimacy of Malia's complaint. As seen in the coparticipants' responses, the reported dialogue is successful in recruiting afliation and sympathy (line 62 and onward).

In sum, the excerpt shows the use of DRS and embodied reenactments not only to exemplify the seriousness of a claimed situation but also to portray the third party as unreasonable and complaint-worthy and the speaker as a reasonable and legitimate complainant. Prosody, gestures, shift in posture, and lexical choices serve as useful resources to signal a shift in footing in portraying the contrast between the speaker's own and the other party's behavior. The excerpt again exemplifes elementary speakers' regularly problematic introductions of reported speech, with broken starts and repairs targeting the establishment of person references that interrupt the progressivity of talk. In this case, Malia deployed a series of deictic pointing and large quotation mark gestures to disambiguate the references of personal pronouns. Interestingly, one year later (see Excerpt 6.14), Malia uses another (this time verbal) resource to accomplish similar disambiguation.

*DRS and reenactments of hypothetical situations to support criticism* 

Not all DRS and reenactments are used to retell past events (cf. Holt, 2007, on hypothetical cases of reported conduct often deployed in jokes). In my data, speakers sometimes use reported talk or embodied behavior to illustrate recurrent events or complaint-worthy conduct more generally – and sometimes it stays ambiguous whether the reported event has occurred or not (see Haakana, 2007). In Excerpt 6.13, Aurelia reenacts what usually happens to her when she goes to the store. She contrasts her own attempts to speak French (lines 1–2) with Swiss people's reaction, which she embodiedly reenacts. Note also Aurelia's use of the English quotative *like* (line 4).

*Excerpt 6.13* 'In the store' (Lun\_2017–03–27)

In overlap with Mia's anticipatory turn-completion (line 5; see Lerner, 1996), Aurelia introduces her reenactment with *tous les personnes like*  ('all the people like', line 4). On the quotative *like*, she drops her shoulders and produces a 'blank', expressionless facial expression (FG.6.23) that she holds until she starts laughing in line 7, thereby illustrating the uncomprehending reactions she gets from people when she makes an efort to speak French. As Mia laughs with her (line 8), Aurelia redoes the expression by pulling her left hand in front of her face and repeating the dropped shoulders and facial expression, this time introducing it with *comme ça* ('like that', line 7) and with closed eyes (FG.6.24). After Mia's continued displays of understanding and afliative laughter (lines 8–9, 11–13; see also Mia's embodied contribution to Aurelia's enactment, in which she reenacts the role of the imagined third party, FG.6.25), Aurelia expands by reporting verbally the responses she gets (lines 14, 16, and onward).

The excerpt illustrates how speakers may support complaints about third parties with the help of reenactments of habitual, recurring, complaint-worthy conduct (Drew, 1998). Here, Aurelia criticizes Swiss people for reacting in an ignorant or unhelpful way when she makes an effort to learn their language. The animated embodied conduct helps reinforce the contrast between her own positive behavior and the third party's negative behavior and displays Aurelia's heightened affective involvement in the telling (Selting, 2010a, 2012). The reported episode also effectively engages coparticipant participation in the telling (Holt, 2000; Sidnell, 2006), as seen in Mia's contributions to the reenacted episode. Finally, the excerpt shows that elementary speakers sometimes resort to English quotatives such as *like* in reported speech or reenactments, further illustrating the difficulty involved in introducing these types of episodes (Couper-Kuhlen, 1999) in the L2.

This section has showcased elementary speakers' use of DRS and reenactments for a range of different interactional purposes in complaining, such as to provide concrete 'evidence' in support of complaint formulations and to contrast the complaint-worthy conduct of a third party with one's own reasonable conduct. Although not shown here, they also sometimes use reported episodes to show their own despair in the face of a troublesome situation. The excerpts have exemplified the difficulties these speakers regularly encounter in introducing DRS and reenactments in French, manifested in broken turn starts, problems with establishing person references, and the use of non-French quotatives, which regularly lead to repair sequences that interrupt progressivity. The examples have shown several different quotatives deployed at the elementary level, such as the verbs *dire* ('to say'), *demander* ('to ask'), and the English-language *like*. The canonical French quotative *dire* (Moreno, 2016) occurs in 9 of 30, or 30% of the cases.

#### *6.2.2 Upper-intermediate/advanced speakers*

This section illustrates upper-intermediate/advanced speakers' use of DRS and reenactments and highlight diferences in their use *vis-à-vis*  elementary speakers. Notably, all upper-intermediate/advanced participants recurrently deploy DRS and reenactments in their complaints. Aurelia now more regularly deploys reported speech in her complaints, whereas the occurrences with Malia are slightly fewer yet recurrent. The interactional purposes of these episodes are similar to what we saw with elementary speakers: to provide a concrete example supporting a negative observation (Excerpt 6.14), to portray the speaker as reasonable and the third party as unreasonable (Excerpt 6.15), and to ofer a hypothetical illustration of a general negative tendency (Excerpt 6.16). The initiations are typically more fuent than at the elementary level, with fewer broken turn starts and long repairs that interrupt the progressivity of talk. Upperintermediate/advanced speakers deploy as many diferent quotatives as elementary speakers to initiate reported talk and actions, but the verb *dire*  ('to say') is used proportionally more often (17 or 53% of 32 cases) and no quotatives in languages other than French occur. This indicates a routinization and possible 'streamlining' of linguistic resources for introducing DRS, which concurs with fndings about ordinary L1 French conversations (see Moreno's study from 2016, in which *dire* is used to introduce DRS in 62.5% of cases).

## 142 *Interactional resources for complaining in the L2*

## *Other-reported speech as concrete 'evidence' supporting complaint formulation*

Excerpts 6.11 and 6.12 exemplifed how elementary speakers sometimes manifest difculties with establishing person references in the reported speech and may deploy embodied conduct, such as deictic pointing and quotation mark gestures, to disambiguate pronominal references. Excerpt 6.14 demonstrates how Malia, now at upperintermediate/advanced level, uses a linguistic resource to accomplish similar disambiguating work – namely, the multi-word construction *ça veut dire* ('it means') – to introduce contextualizing information (see Heinemann & Wagner, 2015) with a specifcation of meaning. Malia complains about the (lack of) indoor heating in her apartment, and she pursues the complaint with a generalized claim, that it is like a 'strategy' in the whole country not to heat the apartments. To support this negative observation, she reports what a neighbor told her: that people just need to dress warmly enough. Although the episode at frst is introduced with the subordinate marker *que* ('that', line 1), projecting IRS (as in Excerpts 6.11–6.12), what follows is DRS.

*Excerpt 6.14* 'Heating' (Mer1\_2017–11–22)

To mark the talk as DRS, and particularly the pronoun *nous* ('we', line 1) as a frst-person pronoun deployed by the third party, Malia again deploys a quotation mark gesture (FG.6.26). In contrast to the use of this gesture in Excerpt 6.13, however, here the gesture is brief and very small in scope, produced by Malia as she maintains her hands low, near the table (cf. the large gesture in FG.6.19, Excerpt 6.12). She then verbally announces a reformulation or specifcation with the construction *ça veut dire* ('it means', line 2), as she briefy points up/right with her left-hand index fnger (FG.6.27) and ofers the third-person pronoun *ils* ('they', line 2), thereby stepping out of the DRS frame and clarifying to whom the pronoun *nous* ('we') referred. Malia continues with another third-person reference, *les sui:sses* ('the Swiss', line 3), which further works as a disambiguation device, and then reports on what this third party 'thinks' (line 3) – namely, that it is 'we' (line 4) who have to wear sufcient clothing (lines 4–6). After what appears as more reported talk with a contrastive formulation about the heating (lines 10–12, 14, 17–18) and some tokens of afliation from Javier (line 22), Malia pursues the complaint by expressing her frustration in an animated fashion (line 23 onward).

## 144 *Interactional resources for complaining in the L2*

In this sequence, DRS is thus used as concrete evidence in support of the formulation of the complaint. Similar to some earlier reported episodes by Malia, she frst introduces the reported speech as IRS through the use of the subordinate marker *que* ('that'), but what follows is DRS. Here we see a new interactional resource being deployed to disambiguate person references, namely the construction *ça veut dire* ('it means') – typically used to announce an upcoming complementary specifcation (Franckel, 2017) – deployed together with brief pointing. Since this multi-word construction fgures in Malia's interactional repertoire already at the elementary level, her use of the construction here shows Malia's ability to draw on existing linguistic resources to address interactional problems that she earlier solved with non-verbal means (Excerpt 6.12). Like in Excerpt 6.12, here Malia also deploys a quotation mark gesture together with a personal pronoun, but the gesture is brief and much less pronounced than earlier (cf. Eskildsen & Wagner, 2015, 2018).

# *DRS to support the portrayal of self as reasonable and of the third party as unreasonable*

Excerpt 6.15 shows the use of DRS and reenactments to portray the complainant in good light and the third party as unreasonable. As with elementary speakers, the contrast between the reasonable 'self' and the unreasonable 'other' is accomplished through an assemblage of verbal, paraverbal, and embodied means, but the excerpt also demonstrates more advanced speakers' ability to report talk and actions in a fuent, unproblematic way. The excerpt comes from Aurelia's long complaint about Swiss people and society presented before (Excerpts 1.1 and 6.6). Aurelia retells the story of when a man stopped his car to blame Aurelia and her friends for standing in the middle of the road. Our main interest is in Aurelia's use of self- and other-DRS and how she steps in and out of the reported talk.

*Excerpt 6.15* 'Better inside' (Lun\_2018–02–26)

The DRS is introduced with the quotative *il a dit* ('he said', line 43). Precisely timed with the beginning of the quoted talk, *vous faites mal*  ('you do wrong', line 43), Aurelia starts rhythmically pointing with her right-hand index fnger toward her coparticipant Mia (FG.6.28) to embody the man's scolding. Using prosodic stress on keywords (lines 44–45), Aurelia animates and upgrades the strength of the third party's conduct, portraying his reaction as overly dramatic. Before reporting on her own response to the man, Aurelia pauses (line 45), frowns, and fips her hands palms up (FG.6.29) to display her irritation and incomprehension (Kendon, 2004; Kaukomaa et al., 2014), momentarily abandoning the reported frame to question the man's conduct from her own perspective. The self-reported talk, which presents Aurelia's excuse to the man and an objection to his scolding (lines 46–47), is produced in a prosodically more neutral way than the other-reported talk, thereby contributing to the portrayal of Aurelia's own acting as reasonable in the face of the overreacting man.

Following a side-sequence, in which Aurelia clarifes the man's positioning (lines 48–52), Aurelia resumes the reported dialogue. This piece of reported talk is accomplished in a similar way as before: Again quoting the man with *il a dit* ('he said', line 53), Aurelia deploys large pointing gestures to afectively animate the man's continued scolding (lines 53–54). At the end of the reported segment, she pauses (line 56) and produces a comparable embodied expression of frustration and incomprehension as before (FG.6.30). This time, however, she does not continue reporting on her own response to the man, but instead produces a rhetorical question (line 57) that questions the man's acting and recruits a small token of afliation from Mia (line 59), before she expands the complaint (lines 60–61 and onward).

The excerpt demonstrates Aurelia's now recurrent use of DRS and reenactments in her complaints, which she typically introduces unproblematically with the canonical French quotative *dire* ('to say') in past tense4 (here *il a dit*, 'he said'*, j'ai dit*, 'I said'). Through marked prosody and animated gesturing, Aurelia creates a contrast between the third party's overreacting, unreasonable behavior and her own reasonable conduct – thereby highlighting the legitimacy and complaintworthiness of the complaint. This contrast is further strengthened by Aurelia's embodied expressions inserted between the episodes of reported talk, which allow her to transcend the reported world and ofer her own embodied evaluations of the situation (cf. Ruusuvuori & Peräkylä, 2009).

# *DRS and reenactments of hypothetical situations to support criticism*

 Some instances of DRS and reenactments of upper-intermediate/ advanced speakers portray hypothetical situations for purposes of supporting claims of general negative tendencies, like at the elementary level. In Excerpt 6.16 (same complaint as in Excerpt 6.7), Cassandra produces DRS to exemplify the general tendency of Swiss people not to make an efort to understand L2 speakers. The excerpt again demonstrates more advanced speakers' typically fuent uses of DRS/reenactments, with an unproblematic introduction of the reported dialogue and talk at a fast pace, as well as the use of a new quotative: the verb *faire* ('to do').

### *Excerpt 6.16* 'Little efort' (Mer2\_2017–03–29)


The reported talk is initiated in an *if-*conditional presenting what happens if one (here *tu*, 'you.sing', is used as an impersonal pronoun) pronounces a word like *bordel* (literally 'brothel', but typically used for 'mess' or 'disorder') as *bordello* (the Italian equivalent, lines 12–14). The reenactment with DRS is introduced with *i font* (reduced form of 'they do') and produced with marked prosody and embodied conduct (line 15). While initiating the reenactment, Cassandra moves her head and torso up and forward in a wobbly manner and closes her eyes (FG.6.31) as she ofers the open-class repair initiator ↑*hein* ('huh', line 15) in high pitch

and with prosodic stress. When continuing with another repair-initiation, *quoi* ('what', line 15), she lifts her right hand to her ear (FG.6.32), and she maintains it there during the frst part of the expansion *qu'est-ce que t'as dit?* ('what did you say', line 17). The assembly of verbal, paraverbal, and embodied conduct excessively depicts and ridicules Swiss people's inability to understand a mispronunciation. Cassandra's actions are responded to by laughter from Miranda (MIR, line 18) and later Liang (line 21), and this leads Cassandra to summarize her telling with serious criticism – that the Swiss do not make an efort to understand (lines 22–23).

In this excerpt, the DRS and reenactment hence do not refer to a specifc past event but rather work as an example supporting a generalized claim about the complained-about party's behavior (Drew, 1998; Mandelbaum, 1991). Using excessive embodied conduct and marked prosody, Cassandra ofers a ridiculing portrayal of the 'other' that serves to underline complaint-worthiness while simultaneously creating a humorous efect. The excerpt shows the use of a quotative that occurs at upper-intermediate/advanced but not elementary level – the verb *faire* ('to do') – which announces a reported action rather than talk and which is a relatively common quotative in informal spoken French (Moreno, 2016). Cassandra's criticism (*font PAS le petit efort de te comprendre*, 'don't do the little efort to understand you', lines 14–15), by which she resumes the serious layer of her complaint after the nonserious reenactment, furthermore exemplifes more advanced speakers' tendency to draw on idiomatic expressions and fgurative language (such as the expression *faire un petit efort*, 'make a little efort') to convey negative stances.

### *6.2.3 Direct-reported speech and reenactments: quantitative comparison*

As shown earlier, elementary and upper-intermediate/advanced speakers deploy DRS and reenactments for similar interactional purposes in complaints. It is primarily the frequency of occurrence and the way these episodes are initiated that difer across profciency levels.

DRS and reenactments are used relatively rarely or never with three out of four elementary speakers. The distribution in use between participants is thus unequal, with Malia producing 25 or 68% of all DRS or reenactments (37 in total). In contrast, at the upper-intermediate/ advanced level, all speakers regularly use these resources for complaining, and the distribution between participants is more even, with all speakers producing between 31% and 36% of the 42 occurrences (Table 6.3).

### 150 *Interactional resources for complaining in the L2*


 *Table 6.3* Frequency and distribution of DRS and reenactments and of the quotative *dire* ('to say') across participants

The quantitative diference in frequency of occurrence of DRS/reenactments across profciency levels indicates that these actions may be challenging for speakers with low L2 profciency (prompting speakers to avoid them), but it does not explain why. In contrast, the qualitative analysis shed some light on *in what ways* these may be challenging, highlighting the introduction of reported talk as a particularly complex act.

Most introductions of reported speech and reenactments by elementary speakers were problematic in some way, with speakers having difculties with verb tense and word order (Excerpt 6.11), introducing DRS frst as IRS (Excerpts 6.11–6.12), and deploying English-language enquoting devices (Excerpt 6.13). A survey of the quotatives deployed by all speakers reveals that both elementary and upper-intermediate/advanced speakers rely on a range of quotative verbs and expressions to introduce DRS/ reenactments, more precisely eight diferent ones per profciency level. However, again there is a diference in distribution, with the verb *dire*  ('to say') deployed to a higher extent by more advanced speakers (in 53% of DRS) than by elementary speakers (30%, see Table 6.3). The diference in the distribution of quotatives suggests a streamlining over time in the use of quotatives, with more advanced speakers routinizing the use of a canonical enquoting device (Moreno, 2016). These fndings, to some extent, concur with Hauser's (2013) observations about a Japanese ESL speaker's resources for introducing DRS over seven months. Hauser documented a decreased use of Japanese quotatives and the emergence of the pattern 'person reference + English-language quotative' (almost exclusively the English canonical quotative *say*) over time.

While the uneven distribution of enquoting devices between speakers at diferent levels provides some indications of L2 speakers' evolving grammar-for-interaction (Pekarek Doehler, 2018), it is noteworthy that only the frst two of the presented excerpts (Excerpts 6.11–6.12) showed the participants' orientations to the elementary speakers' introductions of reported speech as interactionally problematic, leading to repair sequences with interruptions in the progressivity of talk. *That* participants introduced DRS or that particular quotatives were used rather than others, hence, did not pose particular difculties; instead, it was the establishment of person references – *who* was reported on – that proved challenging for intersubjectivity. For Malia, embodied conduct constituted a useful resource to disambiguate person references early in her learning process (Excerpt 6.12), whereas she, with time, relied on verbal means to accomplish the same interactional work (Excerpt 6.14).

#### **6.3 Discussion**

This chapter has examined some of the interactional resources L2 French speakers deploy to construct their complaints *as* complaints and how their use of these resources changes over time (Aurelia and Malia) and across profciency levels. The focus on negative assessments and on DRS and reenactments was motivated by the central role of these resources for constructing complaint-worthiness identifed in prior literature as well as by their prominence in the data under scrutiny. At the same time, these analytical objects also encompass various smaller units of interactional means observed as resources for complaining, such as negatively loaded lexical items and multi-word constructions, extreme-case formulations, marked prosody, afect-laden non-lexical vocalizations, and embodied conduct expressing a negative stance. I hope, therefore, to have demonstrated the participants' use of such resources more generally. While only to some extent addressing individual participants' longitudinal trajectories, the analysis revealed larger developmental patterns in L2 speakers' use of negative assessments and of DRS and reenactments in complaints.

The study of negative assessments showed that speakers, as they move from elementary to upper-intermediate/advanced L2 profciency, considerably broaden their repertoire of linguistic resources for assessing. This change manifests in a more diversifed set of assessment formats, both on the lexical and syntactic level, with upper-intermediate/advanced speakers deploying a greater variety of assessment adjectives and intensifers and producing syntactically more diverse assessment turns than elementary speakers. The more diverse formats for negatively assessing have interactional consequences for the construction of complaints, as they allow speakers to more readily produce high-grade assessments and upgrades of frst assessments in context-sensitive ways, for example, in order to participate in joint complaining. The fndings about longitudinal diversifcation concur with and expand on Nguyen's (2019) observations about one L2 English speaker's progressive diversifcation of positive assessment segments and Hellermann's (2008) observations about one ESL speaker's longitudinal use of closing-implicative assessments. The fndings also support prior studies on the development of L2 IC that suggest that such development crucially involves the progressive diversifcation of methods for action (see Section 2.1). They furthermore complement research on complaining more generally by systematically documenting some of the specifc purposes of negative assessments in complaint sequences (such as constructing joint complaints) – something that has not been done before (but see Drew & Walker, 2009; Rääbis et al., 2019, for very brief observations about resources for joint complaining).

Both the analysis of negative assessments and the analysis of DRS and reenactments enrich our understanding of the development of L2 grammar-for-interaction (Pekarek Doehler, 2018), an integral part of L2 IC, and contribute to the limited yet growing body of interactional usage-based SLA research (e.g., Eskildsen, 2020; Eskildsen & Pekarek Doehler, 2022). In addition to showing a general diversifcation of lexical and grammatical formats over time, the developmental trajectory for negative assessment turns observed with some participants resonates with the idea of exemplar-based learning (Eskildsen, 2012, 2015). It seems that some speakers may start out by producing negative assessments in the form of particular lexico-syntactic patterns (e.g., *c'est très difcile*, 'it's very difcult') that subsequently 'loosen up' to allow for the incorporation of other components (such as diferent adverbial intensifers). Similarly, both sub-studies of this chapter documented increased use of multi-word constructions and fgurative expressions expressing a negative stance with higher profciency levels, in line with what has been observed in usage-based studies of such linguistic resources (e.g., Erman et al., 2016; Forsberg, 2008; cf. also the 'pragmatics staircase' presented by Norrby & Håkansson, 2007, which suggests that L2 speakers' acquisition of pragmatic language use involves the progression from repetition and unanalyzed phrases to idiomatic language). In contrast, the comparison of enquoting devices deployed by elementary and upper-intermediate/ advanced speakers in the initiation of DRS revealed a pattern going in the other direction – namely, a 'streamlining' of quotatives over time, with more advanced speakers to a higher extent deploying the canonical quotative verb *dire* ('to say') than less advanced speakers, who exhibited more variation in their enquoting devices.

We hence see developmental patterns going in opposite directions when it comes to the reliance on particular linguistic constructions. At the elementary level, speakers rely heavily on certain types of linguistic constructions for expressing a negative stance and use these in diverse interactional contexts (such as *c'est très difcile*; cf. again Larsen-Freeman's [2006] idea of make-do solutions). At more advanced levels, a diferent type of linguistic constructions is found: those in idiomatic expressions and fgurative language. At the same time, a kind of longitudinal 'streamlining' of certain linguistic resources is observable, with *less* diverse linguistic formats being used for routine actions such as DRS introductions. These fndings may seem contradictory, but they are not necessarily: As suggested in a few recent studies on L2 speakers' development of grammatical resources for action (Pekarek Doehler & Balaman, 2021; Pekarek Doehler & Skogmyr Marian, 2022), while progressively diversifying certain action formats (such as for opening tasks and storytellings: Hellermann, 2008), L2 speakers also routinize and specialize certain patterns of language use for the accomplishment of particular interactional tasks. This development works in favor of interactional efciency, as the speaker can select one linguistic resource that proves locally efcient and stick to that in similar action contexts (for example, to hold the foor, see Pekarek Doehler & Skogmyr Marian, 2022; see also Eskildsen, 2012; and Kim, 2019, for successful reuses of idiosyncratic patterns for recurring actions, and Eskildsen, 2020, on the development of linguistic constructions for specifc purposes more generally). In terms of resources for introducing DRS, a larger dataset may allow for more conclusive observations regarding a streamlining of quotatives over time. In any case, the more advanced speakers' more frequent use of the quotative *dire* ('to say') concurs with what has been observed about French L1 speakers' use of this quotative (Moreno, 2016) and shows one way in which L2 speakers' grammar-for-interaction progressively approximates the ones of L1 speakers (Pekarek Doehler, 2018).

Moreover, both sub-studies highlighted the fundamentally multimodal nature of expressions of negative stance (including assessments) and of DRS and reenactments. Speakers at both profciency levels deploy prosody, non-lexical vocalizations, gaze direction, facial expressions, gestures, and change in posture, in concert with verbal resources to upgrade negative stance displays, show afective involvement, and construct contrasts between self and others, much like what has been observed in the L1 literature (see Section 3.3). Some observations regarding speakers' use of these resources are noteworthy, however. First, speakers recurrently fnely synchronize specifc embodied conduct with the delivery of negative assessment segments and treat these as multimodal packages in the repetition of assessments. This observation contributes to the growing body of research focusing on specifc constellations of linguistic and embodied resources for action formation and interaction-organization (Goodwin, 2007; Hayashi, 2005; Kärkkäinen & Thompson, 2018; see also contributions in Pekarek Doehler et al., 2021) by ofering evidence of such constellations in a new action context. Second, whereas non-verbal conduct in certain action contexts and positions remains constant over time (such as in the multimodal packages just described, non-verbal completions of

 verbally incomplete turns, and non-verbal conduct embodying characters in DRS and reenactments), my data suggest that the use of stand-alone non-lexical vocalizations and their accompanying embodied conduct to produce precise negative assessments decreases over time. The analysis of DRS introductions also provided some evidence for the progressive decrease in gesture use and gesture scope for specifc action purposes. While tentative in nature, these observations are in line with recent work in usage-based linguistics and CA (Eskildsen & Wagner, 2015, 2018; Skogmyr Marian & Pekarek Doehler, 2022) that has documented a progressive decrement in gesture use (and in gesture scope) as verbal resources take more prominent roles in particular multimodal packages in the L2. To some extent, L2 speakers' use of embodied conduct thus seems to work compensatorily, helping speakers to efectively participate in social activities (such as assessment activities) despite limited linguistic means. More systematic studies of, for example, the precise interactional purposes of non-lexical vocalizations and other non-linguistic conduct in L2 speakers' interactions over time will ofer more robust evidence regarding these issues.

# **Notes**


This chapter presents an exploratory investigation of how the development of complaint practices relates to the progressive accumulation of shared knowledge and changing social relationships. As mentioned in Section 2.2, there is a lack of CA research that directly addresses the link between the development of L2 IC and socialization processes, such as evolving social rapports. Research on complaints indicates that these are sensitive to the interactional context and to the nature of the relationship between the participants (Chapter 3). Complaints may therefore be a specifcally suitable analytical object for examining how language learning interfaces with socialization processes. My data lend themselves well to an exploratory examination of these issues, as they allow for the longitudinal study of complaint practices with speakers who are in the process of getting increasingly acquainted. Based on two longitudinal case studies focusing on the participants Suresh (Section 7.1) and Malia (Section 7.2), I illustrate how participants' increasing shared experiences and deepened social relationships over time afect the way they invoke and recipient-design their own complaints and respond to other speakers' complaints. Instead of aiming to track the participants' entire shared interactional histories (as in some of the longitudinal CA studies cited in Section 2.2.3), I trace the interactional histories of two particular complainables that recur in the participants' interactions. I discuss how the two case studies inform our understanding of the development of L2 IC and open avenues for future, more in-depth EMCA studies into the relationship between L2 development and socialization processes (Section 7.3).

#### **7.1 Case study 1: Suresh**

In Section 5.3, I showed that negative stance expressions about complainables that lie within the epistemic domain of the coparticipants ofer opportunities to engage in co-complaining, allowing participants to exchange displays of afliation. In this section, I document how Suresh, an elementary speaker of French, draws on his interactional history with his coparticipant Aurelia to produce assessments about the temperature that match Aurelia's previously expressed stance, thereby providing an opportunity for the

 participants to engage in co-complaining. The analysis focuses on Suresh's participation in the conversation group of Aurelia, Mia, and Natascha; participants with an initial estimated profciency level of A2 (Suresh's estimated level was slightly lower, A1). During the ten months that Suresh participates in this group, he contributes actively to only three complaint sequences. All three complaints are about the weather. These complaints are not about exactly the same complainable, since the weather changes from one day to another. The frst complaint is about the heat, the second about rain (not analyzed here), and the third about the cold. They are all complaints about weather conditions, however, and they serve similar interactional small-talk purposes. They all take place at moments when Suresh is alone with Aurelia, meaning that these are interactions in which Suresh is obliged to contribute actively to the conversation instead of merely listening to his coparticipants – which he often does when other participants are present.

The analysis presents two sets of related interactions in which Suresh and Aurelia partake in stance-taking about the temperature, making up a total of fve chronologically ordered excerpts (Excerpts 7.1–7.2 are about the heat; Excerpts 7.3–7.5 concern the cold). Only two of the excerpts (Excerpts 7.2 and 7.5) develop into actual complaints. I analyze how Suresh relies on his interactions with Aurelia presented in the preceding excerpts (Excerpt 7.1 and Excerpts 7.3–7.4, respectively) to ofer opportunities for Aurelia to complain.

#### *7.1.1 Profering a complaint about the heat*

Excerpts 7.1 and 7.2 demonstrate how Suresh (SUR) draws on his shared interactional experience with Aurelia to facilitate complaining about the heat. Excerpt 7.1 comes from the end of Suresh and Aurelia's third conversation with each other. This is the frst time they meet alone, and they are about to end the conversation. After establishing that they will see each other again in two weeks (lines 1–7), they both stand up (line 9) to pursue the leave-taking. Aurelia then utters a non-lexical vocalization (line 12) and assesses the temperature as *chaud* ('warm', line 14).

*Excerpt 7.1* 'In two weeks' (Lun1\_2017–05–29)

 Although Aurelia's °*il fait°* ↑*chaud* ↑*mainte*↑*nant* ('it is warm now', line 14) is neutrally valenced in that 'warm' could be interpreted as either positive or negative, the just-preceding non-lexical vocalization *ou°fhhh°*, as well as the prosodically marked delivery of the second part of the turn (with change in pitch and stress on *chaud*, 'warm'), makes the turn understandable as a negative assessment of the heat. Whether Suresh interprets it as such is not clear from the excerpt, as he does not respond and both participants leave the table (line 15), but Suresh's actions in Excerpt 7.2 indicates that he has paid attention to Aurelia's turn.

Excerpt 7.2 takes place right at the beginning of Suresh and Aurelia's next meeting, which is three (and not two, as suggested by Aurelia in Excerpt 7.1) weeks later. The participants have just sat down at the table, and Suresh initiates the conversation through a conventional how-areyou inquiry to Aurelia (line 2). After minimally responding to Aurelia's reciprocating question (line 6), Suresh produces a topic profer (line 8) that is oriented to by Aurelia as an occasion to complain about the heat.

*Excerpt 7.2* 'With heat' (Lun\_2017–06–19)


	-



Besides the omission of the adverbial *maintenant* ('now'), Suresh's turn in line 8 precisely repeats Aurelia's assessment two weeks earlier: *il fait chaud* ('it is warm'). In contrast to Aurelia in Excerpt 7.1, however, neither Suresh's turn delivery nor his accompanying embodied conduct conveys any particular stance. Aurelia nevertheless orients to Suresh's assertion as a negative assessment and an opportunity for her to express her frustration about the temperature. By rolling her eyes, raising her right hand (FG.7.1), and then starting to wave it in front of her as a fan while producing a long sigh followed by *ouais* ('yeah', line 10, FG.7.2), Aurelia ofers an embodied expression of negative afective stance (Clift, 2021; Goodwin & Alim, 2010) that transforms Suresh's assertion into an invitation to complain. Aurelia subsequently develops the sequence into a small complaint by ofering several negative stance expressions and asserting the negative consequences

of the heat (lines 13, 17, 28). Suresh, however, refrains from contributing to the complaint. Instead, he displays puzzlement (line 19) as to Aurelia's reference to stress (line 17) and does not verbally afliate or expand on her stance expressions. By not contributing to the expansion of the sequence, Suresh does not orient to his initiation as an invitation to produce a joint complaint. Instead, it seems to have been delivered as a mere topic profer, in which "a speaker proposes a particular topic . . . but does not actively launch or further develop the proposed topic" (Scheglof, 2007: 169–170), which is treated by Aurelia as a *complaint profer* that she accepts by treating the profer as a negative assessment and by expanding the sequence. Likely due to the lack of afliative displays from Suresh, Aurelia quickly abandons her attempts to get him on board with the development of the complaint, and the sequence closes upon a greeting from the assistant (who has been arranging the technical setup of the recording, line 35).

By analyzing Excerpts 7.1 and Excerpt 7.2 chronologically, we have seen two consecutive sequences (the closing of one conversation and the opening of the next) in which Suresh and Aurelia topicalize (frst Aurelia, then Suresh) and engage in stance-taking (primarily Aurelia) about the heat. In the second of these excerpts, Suresh uses an expression that is almost identic to that of Aurelia's assessment of the heat three weeks earlier to initiate the sequence. In a sense, Suresh thus 'picks things up where they last left them' and thereby orients to his and Aurelia's shared interactional history.

#### *7.1.2 Profering a complaint about the cold*

Excerpts 7.3–7.5 show a similar chronological chain of events, although here Suresh *changes his stance-taking* toward the outdoor temperature from two interactional encounters (Excerpts 7.3–7.4) to another (Excerpt 7.5), thereby profering a joint complaint about the cold that engenders mutual exchanges of afliation.

Excerpt 7.3 takes place almost four months after Excerpt 7.2. The participants (Mia, Natascha, Suresh, Aurelia) have been talking about the fact that it is still warm outside despite it being October. Natascha (NAT) has said that she thinks that the winter will be cold, and after asserting that she likes the cold, she asks Aurelia whether she likes the winter (line 1). In the talk that follows, Aurelia and Suresh display conficting stances toward the cold: Aurelia expresses her strong dislike of the winter and the cold, whereas Suresh claims that he likes the winter and cold temperatures (see particularly turns marked in bold):

*Excerpt 7.3* 'Sensitive' (Lun\_2017–10–16)

In response to Natascha's question, Aurelia asserts several times that she does not like the winter (lines 5, 8, 11). Natascha questions Aurelia's answer by suggesting that Aurelia, who comes from England, should be used to cold weather (lines 12, 15). Aurelia then asserts more explicitly that she does not like the winter (lines 20, 25) because she is sensitive to the cold (lines 29–30) and prefers when it is very warm (lines 33–34). Natascha thereafter solicits Suresh's opinion (lines 42–43), prompting Suresh to express his contrasting stance toward the winter and the cold: He asserts that he likes (the winter) *beaucoup* ('a lot', line 45) and in fact prefers the winter (lines 48–49). Following an exchange about the climate in India, Suresh's country of origin (lines 53–66), Suresh again claims that he prefers the cold: *je préfère (1.6) froid* ('I prefer cold', line 72). At this point, Aurelia objects by suggesting that it is because Suresh does not leave the house (line 76), and this leads to laughter (lines 78, 80–82), possibly to mitigate the potential awkwardness created by Aurelia's objection (Petitjean & González-Martínez, 2015). Another objection by Aurelia follows (lines 83–84), and Natascha then expresses her own opinion about the cold (line 87 and onward).

In this sequence, Aurelia and Suresh hence express conficting stances toward the winter and cold temperatures, with Aurelia strongly asserting her dislike of the cold and Suresh claiming that he, in fact, likes the winter and the cold. Although Suresh remains mostly silent during Aurelia's stance displays, he shows his attentiveness to her talk by smiling and laughing at appropriate moments (lines 12–24). Aurelia engages more actively in response to Suresh's stance expressions, rejecting these by suggesting that Suresh does not go out of the house. Suresh and Aurelia's sustained attentiveness to and long-term remembering of each other's stance expressions about the cold are visible in the next excerpt.

Excerpt 7.4 occurs four weeks after Excerpt 7.3. Suresh and Aurelia are alone this time. Now Aurelia is the one to initiate talk about the temperature right at the conversation start, and the participants again express conficting stances toward the cold. The excerpt begins as Aurelia is approaching the table where Suresh is sitting (see line 2).

*Excerpt 7.4* 'Three clothes' (Lun\_2017–11–13)

 By asking Suresh about his opinion about the cold (line 1) and holding him accountable for his stance expression four weeks earlier (line 4), Aurelia explicitly 'talks into being' their shared interactional history (Mondada, 2018; Skogmyr Marian, 2018; Voutilainen et al., 2018) and again orients to weather talk as a relevant frst activity to launch the conversation between her and Suresh (as in Excerpt 7.2). Following Suresh's confrmation of his positive stance (lines 5, 8), Aurelia expresses her own conficting opinion that she strongly dislikes the cold (lines 11, 14) because she needs to wear so many layers of clothing (lines 15, 17, 20). Suresh frst resists Aurelia's position by recycling Natascha's objection four weeks earlier that Aurelia is from England (lines 26–27, 29, 31), where it is very cold (line 34; cf. Excerpt 7.3). Upon Aurelia's insistence that she nevertheless dislikes the cold (line 37), Suresh suggests the possible explanation that this is indeed the reason for Aurelia's opinion (lines 39, 43, 46, 49–50), which Aurelia confrms (line 52).

In the four excerpts seen so far, Suresh and Aurelia have engaged in several diferent stance-takings toward the temperature. In Excerpts 7.1–7.2, which took place in the summer, Aurelia expressed her dislike of the heat. In Excerpts 7.3–7.4, which come from recordings in October and November, Suresh expressed his liking of the winter and the cold, whereas Aurelia strongly asserted her dislike of the cold. In the next fnal excerpt, Suresh produces a negative assessment of the cold, which hence conficts with his claimed opinion in the two preceding excerpts. In doing so, he again uses his shared interactional history with Aurelia to profer weather talk – and a complaint. In contrast to in Excerpt 7.2, however, this time his profer is clearly negatively valenced, and it leads to exchanges of afliation between the participants.

Excerpt 7.5 comes from Suresh and Aurelia's next meeting, two weeks after Excerpt 7.4. The participants are again alone, and they have just sat down at the table. After an exchange of how-are-yous (lines 1–4), Suresh produces a high-grade negative assessment of the cold (line 6).

*Excerpt 7.5* 'Every week' (Lun\_2017–11–27)

The assessment *mai::s c'est tro:p (.) froid* ('but it's too cold', line 6) is clearly negatively valenced and strengthened through the adverb *trop* 'too'. It contrasts with Suresh's claimed stance toward the cold in Excerpts 7.3–7.4; instead, it matches Aurelia's expressed negative stance in the same excerpts. Aurelia orients to the assessment as an invitation to complain. She immediately expresses her alignment and afliation through a loud and elongated *OUAI::S* ('yeah', line 7). This agreement falls in overlap with Suresh's *très froid* ('very cold', line 8), which is possibly a correction of *trop froid* ('too cold'), and Aurelia repeats her agreement in a prosodically similar manner (line 9). After a brief silence (line 10), Aurelia produces a small sigh, the non-lexical vocalization *ohhh* (line 11), by which she further expresses her frustration. She then expands with reference to the recurrence of the topic in the participants' conversations – 'we say the same thing every week no,' (line 12), again talking their shared interactional history into relevance – and starts laughing (line 13). Suresh also laughs (line 14) as Aurelia asserts 'but it's true' (line 15), thereby justifying and legitimizing the repeated criticism of the weather, and Suresh agrees (line 16). Next, Aurelia reissues and upgrades the negative assessment, *ou(h)ais il fait très- >trop trop trop< froid* ('yes it is very- too too too cold', line 17), using an extreme-case formulation to further legitimize the complaint (Pomerantz, 1986), and Suresh aligns with *mm-hm* and large head nods (lines 18–19).

Aurelia subsequently expands with what will become a jointly constructed account for the weather criticism – namely, that because there is no bus that goes from Aurelia and Suresh's institutes to the building where the conversation circle takes place (lines 25–27), they have to walk (line 21) and are therefore afected by the cold weather. By opening her hands to the sides and shaking her head and shoulders slightly as she suggests that there is no bus from their faculty (FG.7.3), Aurelia laminates an embodied expression of frustration upon the neutrally formatted assertion. Through confrmation requests in a declarative format, she formats her account as a 'candidate account' in which she recruits Suresh's co-engagement with the help of gestures and gaze conduct (Streeck, 2009; Stivers & Rossano, 2010) – indexical pointing up/right and then down during the reference to the descent from the science faculty (lines 21–22, 26–30) and gaze at Suresh at key moments (lines 21–22, 25–30). Suresh frst confrms Aurelia's assertion about going by foot (line 23). He then, in overlap with Aurelia's response-pursuit in line 30 and after a moment of thinking (see in-breath and gaze aversion in lines 31–33 indexing cognitive activity, Goodwin & Goodwin, 1986), asserts that there is a bus (*il y a*, 'there is', line 33), but he immediately thereafter adds that even for the bus you have to walk a few minutes (lines 33, 35–36). In doing so, Suresh aligns with Aurelia's suggested account despite disconfrming its factual grounds. He then lifts and lowers his shoulders while simultaneously opening up and closing his hands slightly and making small headshakes, embodiedly expressing the negative consequence of his assertion and a sense of hopelessness, that there is nothing to do about the situation (lines 38–39, FG.7.4–6; see Kendon, 2004; Selting, 2012). This assembly of embodied conduct closes Suresh's turn, which Aurelia receipts with a small *ahh ouais*  ('ohh yeah', line 37) and small headshakes (lines 37–40). She then, in line 40, initiates what seems to be the beginning of an upshot or summary statement (*donc e:h*, 'so uh') but abandons it, thereby aligning with Suresh's embodied expression and treating it as a sufcient characterization of the situation. She subsequently transitions into related talk about winter clothes (line 44).

In contrast to his stance-taking two weeks earlier when he asserted that he *likes* the cold 'a lot', in this sequence Suresh occasions a joint *complaint* about the cold. Similar to what we saw in Excerpt 7.2, the assertion about the temperature works as an efective conversation starter that progresses the conversation past the exchange of howare-you inquiries. Contrary to the assertion about the temperature in Excerpt 7.2, however, this one is clearly negatively valenced. Similar to before, it is Aurelia who does most of the work developing the sequence into a complaint. In this sequence, though, Suresh participates actively to a greater extent, as he aligns with Aurelia's stance expressions more explicitly through verbal and embodied means and contributes to the

construction of an account about the complaint-worthiness of the situation.

#### *7.1.3 Suresh: summary and intermediate discussion*

The analysis of Excerpts 7.1–7.5 has concretely exemplifed how participants' shared interactional histories may serve as an important resource for accomplishing context-sensitive and recipient-designed social actions (Deppermann, 2018; Norrthon, 2019) in the L2. This is particularly visible in Excerpt 7.5, where Suresh, by profering a topic that previously worked well to advance the conversation past the exchange of greetings (Excerpts 7.2 and 7.4) and by expressing a stance that Aurelia had earlier expressed herself (Excerpts 7.3 and 7.4), both actively contributes to progressing the conversation and initiates an activity in which the coparticipants can exchange displays of afliation. While for most participants in my data these actions are not particularly noteworthy, it is worth considering Suresh's low L2 profciency and his rather passive participation in most conversation circle meetings. By ofering an opportunity to engage in joint complaining, in Excerpt 7.5 Suresh shows interactional agency and demonstrates his capacity to draw on prior interactional experiences to recipient-design talk – abilities that have been observed as key components of increased L2 IC (Pekarek Doehler & Pochon-Berger, 2015). These fndings resemble those of Brouwer and Wagner (2004), who analyzed the openings of two series of professional phone calls with L1 and L2 speakers of Danish and German, respectively. The authors observed that the speakers with each call subtly adapted their opening conduct so as to more smoothly accomplish the opening sequence in coordination with their interlocutor. Suresh and Aurelia's repeated engagement in howare-you inquiries followed by weather talk develops into a similar kind of conversational routine by which the speakers longitudinally work to come together as conversation partners (cf. Greer, 2019).

 Moreover, the reappearance of precise linguistic material and action formats (assessments about the temperature, objections) throughout the studied excerpts mirrors claims made in usage-based linguistics about the high degree of recurrence of specifc linguistic patterns in everyday language use, which is what makes usage-based learning possible (see, e.g., Ellis, 2002; Eskildsen, 2020, Hopper, 1987, 1998; Tomasello, 2003). As Hopper puts it,

[S]peakers borrow heavily from their previous experiences of communication in similar circumstances, on similar topics, and with similar interlocutors . . . We say things that have been said before. Our speech is a vast collection of hand-me-downs that reaches back in time to the beginnings of language.

(1998: 157–159)

It is analytically impossible to prove that the participants' accomplishments of similar actions through similar linguistic forms at the diferent interactional encounters with one another investigated here are defnitely linked to each other. However, it is clear that Suresh and Aurelia, by repeatedly invoking the weather and topicalizing each other's stances, do establish a relevant component of what it means to be a competent speaker in their local speech community, and the reuse of interactional material from earlier encounters appears to constitute a central resource in Suresh's construction of competency in this context. This is furthermore reminiscent of Goodwin's argument about both local and long-term *accumulation* as resource for human social action, which may be particularly important in processes involving the socialization of speakers into "competent speakers in their community" (2018: 24):

The ability to reuse materials created earlier, including, crucially, materials produced by others, and, moreover, not simply to copy those resources, but transform them, creates forms of action with an unfolding, historical sedimentation of accumulative, contingent structure that has great power.

(Goodwin, 2018: 31)

Finally, Suresh's contradictory stance expressions toward the cold from one interactional encounter to another provide evidence as to participants' emic orientations to the interpersonal purposes of complaining. Suresh and Aurelia's recurrent weather talk after the exchange of greetings and how-are-you inquiries in Excerpts 7.2, 7.4, and 7.5 works as a way to enter into and advance the conversation for speakers who perhaps do not have so many things in common besides being foreign students at the same university. While the weather has previously been considered a neutral and 'safe' topic for small talk (Laver, 1975; Svennevig, 1999) that can work as a transition into other topics (Sacks, 1992), the fact that Suresh changes his expressed opinion about the cold from one interactional encounter to another so as to ensure an afliative response from his coparticipant shows that he treats this talk as an important opportunity to do relational work, similar to what has been argued for complaining generally (Boxer, 1993; Günthner, 1997; Hanna, 1981). This change also shows the fundamentally interactional nature of stance-taking. Although we have no idea about Suresh's actual opinion about the heat and the cold, this is not relevant; what is emically important to Suresh is what he can accomplish with his stance-taking in the socially situated interaction (cf. Sacks' [1975] observation that "everybody has to lie" in how-areyou inquiries). Case study 2 further advances our understanding of these issues by examining the impact of joint experiences and evolving relationships on complaint practices in the context of a rather diferent kind of complainable.

### **7.2 Case study 2: Malia**

The second case study focuses on a more personal type of complainable: a participant's difculties with learning and using French for work. As presented in Section 4.3, Malia was an upper-elementary speaker (A2) at the beginning of the recordings. During the 16 months of her participation, she advanced to upper-intermediate/advanced level (B2–C1). Malia is a doctoral assistant, and on numerous occasions in the recordings, she invokes her problems at her workplace and specifcally with the professor, who is her PhD adviser and boss. According to Malia, these problems are primarily related to Malia's French skills and the professor's expectation that Malia uses French all the time. In 23 sequences over the 16 months, Malia expresses a negative stance toward her work and/or her professor. These stance expressions often (but not always) develop into complaint sequences.

The analysis documents how the specifc complainable1 is talked into relevance, constructed and reconstructed by Malia's conversation group over three semesters. The analysis is divided into three parts, representing three phases of Malia's participation in the conversations:


The change in participants between the second and the third semesters allows for a comparison over time between Malia's complaints with, on the one hand, well-known coparticipants versus new acquaintances, and on the other hand, elementary versus intermediate/advanced speakers. I now present six chronologically ordered excerpts from the 23 sequences in which Malia and/or her coparticipants invoke the specifc complainable, distributed over the three investigated semesters. I show how the topical focus, the way the complainable is talked into being, and the nature of the coparticipants' contributions to the complaint sequences vary as a function of both their shared interactional histories, the nature of their relationships, and their French profciency.

### *7.2.1 Fall semester of 2016 (months 1–3)*

During the frst few conversations, the participants spend considerable time getting to know each other by telling about themselves and their lives. A recurrent conversation topic is the participants' attempts to learn French. This period also coincides with the start of Malia's doctoral assistantship, which she often discusses with her coparticipants. On seven occasions during the frst three months, Malia complains about her work. These complaints are all related to the difculty of using French. Five of the sequences are initiated in frst position by Malia herself. The remaining two occur in response to open-ended, neutral K− questions (Heritage, 2012) from Mariana, that is, questions that display a weak epistemic stance and hence do not index any *a priori* knowledge of or orientation to trouble. The coparticipants' participation in the sequences is typically limited. Besides when attempting to resolve issues of intersubjectivity or helping with linguistic problems, the coparticipants mostly stay silent during Malia's complaints and provide signs of afliation and sympathy through minimal linguistic or non-verbal means. Excerpts 7.6 and 7.7 illustrate the type of initiations that occur during the frst few months and the coparticipants' typical level of participation.

Excerpt 7.6 shows the very frst time Malia complains about having to learn French for work. The excerpt comes from the participants' second meeting. Malia has asked Theo and Mariana whether they have to teach any courses at the university. Mariana has confrmed that she will but that she does not know yet whether it will be in English or French. Mariana then reciprocates the question to Malia (line 2). In response, Malia invokes a 'problem' (line 5) – that her supervisor expects her to learn French so that she can teach – and she subsequently complains about the difculty of this requirement (starting in lines 35–37).

*Excerpt 7.6* 'Teach' (Mer1\_2016–10–19)

*The interactional history of a complainable* 

Mariana's *et toi* ('and you', line 2) is recognizable as a reciprocating inquiry about teaching requirements, whereby she ofers Malia an opportunity to tell about her own situation. Its open-ended format indexes the speaker's low epistemic status (Heritage, 2012) and does not manifest any knowledge about the possible answer. In response, Malia starts telling about her 'problem' (line 5). After a side-sequence in which Theo helps Malia resolve a word search targeting the word for 'supervisor' (lines 6–11), Malia reports that her supervisor has told her that she has to speak French next year (lines 12–13, 16, 18–19, 25), as this semester she only needs to go to classes in English (lines 27–31).

Following this, Malia expands with afect-laden negative assessments, accounts, and stance displays to portray her requirement to speak French as an unreasonable difculty worth complaining about. Although in the frst assessment (lines 35–36), Malia initially deploys the assessment adjective for 'calm', *tranquille* (which she self-repairs to a related non-targetlike form), her turn-initial 'but' and her next actions make the assessment recognizable as a negative one. She ofers an account reporting on her own thoughts (Haakana, 2007) or perception (lines 36–38). At the expressive *oh mon dieu* ('oh my god', line 37), Malia markedly raises her eyebrows and looks up (FG.7.7), upgrading the afective loading of the expression (Ruusuvuori & Peräkylä, 2009), and then produces a loud laughter and the direct-reported *je n'peux pas parle::r* ('I cannot speak', line 38). The reported thought is animated with prosodic modulations and with gestures that display both the obviousness and seriousness of the situation (hands to side, palm up, FG.7.8, see Kendon, 2004; Marrese et al., 2021, and clapping the left hand in the right hand, FG.7.9). The coparticipants do not join in Malia's laughter (cf. Jeferson, 1984b), but Mariana smiles and nods a bit, displaying her attentiveness to Malia's actions (line 39).

Malia expands with *e:t que je: peux (0.4) <enseigner> en français* ('and that I can teach in French', lines 40–41). The prosodic realization of the turn and the accompanying large gestures (FG.7.10) make the turn recognizable as an extreme-case formulation expressing the unreasonableness of the propositional content: that Malia, who cannot speak French, would be able to teach in French. After uttering a small whining sound and ofering accompanying embodied conduct expressing the difculty of the situation (line 43, FG.7.11), Malia launches the prosodically and embodiedly upgraded summary assessment *c'est tR:ès* ↑*di*↑*f*↑*ci::le* ('it's very difcult', line 44, FG.7.12). Through the assemblage of verbal and embodied conduct, Malia thus summarizes her supervisor's expectations as generating unreasonable difculties for Malia and therefore being complaint-worthy (Drew, 1998). At this point, Mariana ofers a token of afliation and sympathy with Malia by repeating *très difcile* ('very difcult', line 45) in low volume. After having requested (lines 46–48, 51, 57, 60–61) and received a clarifcation regarding Malia's obligation to speak French (line 62), Theo also expresses his sympathy through the nonlexical vocalization *o:hhh* (line 63) and by tilting his head slightly to the side. Malia then repeats her confrmation (line 65), and the participants expand the sequence with further talk about Malia's situation.

The sequence exhibits several features that characterize it as the frst complaint about Malia's difculties associated with speaking French at work and that index the participants' relatively novel relationship. The complaint is initiated in response to the open-ended question *et toi* ('and you'), which shows Malia's coparticipant's lack of epistemic access to Malia's professional situation and which does not display any awareness of potential problems. The coparticipants contribute to the development of the sequence by helping to complete Malia's word searches, through small acknowledgment tokens, and by asking clarifcation questions. By repeating Malia's own summary assessment and through non-lexical vocalizations, they also show their sympathy for Malia, but these displays of afliation come late and take linguistically minimal forms (cf. Section 5.3).

Two weeks after Excerpt 7.6, Malia tells her coparticipants about her frst ofcial day as a doctoral assistant (Excerpt 7.7, see also Excerpt 6.11). Theo has just given a status update on how he is doing, and Malia's *okay* in line 1 closes the sequence. Malia then volunteers a status update about herself, announcing that she started her job the day before (lines 3–4). This story preface is neutrally formatted, but by leaning her head in her left hand (FG.7.13), Malia gives some clues as to the negative valence of the story (Ruusuvuori & Peräkylä, 2009). When the coparticipants have confrmed recipiency (lines 6–7), Malia explains where the job is located (lines 9–17). She then starts reporting and embodiedly illustrating how 'horrible' her frst day was, and she again complains about the difculties associated with speaking French at work.

*Excerpt 7.7* 'Started work A' (Mer1\_2016–11–02)

When initiating the report of her frst day, Malia frst lowers her head and covers her face with her hands (FG.14) in a way that conveys a negative stance. She thereafter assesses her experience as *un peu horrible* ('a bit horrible', line 21) while fashing her eyebrows (FG.7.15) and then looking down to her right and eventually starting to smile. The oxymoronic yet strongly negative assessment explicitly frames the telling as a troubles telling, but Malia's smile as she initiates an account (line 21) adds a humorous layer to the talk (Edwards, 2005), and Theo responds accordingly with laughter (line 22). In her account, Malia presents a discrepancy between her expectations about her job and the reality: She thought that she could speak English the frst year (lines 23–24, 27), while in fact, today everyone spoke French with her (lines 31–32). This turn provides the frst concrete details about why her frst day of work was horrible and complaint-worthy – she unexpectedly had to speak French already on her frst day (cf. Schulze-Wenck, 2005). Malia then develops the detailing by reporting what her professor asked (lines 34–35) – namely, that the colleagues should only speak French with Malia (lines 41–42). Through DRS, prosodic emphasis, and horizontal hand gestures (FG.7.16–7.17), Malia animates and upgrades the afective loading of what works as the climax of the troubles telling.

As discussed in Sections 6.1.1 and 6.2.1, the complaint story climax does not get the expected afliative or sympathetic responses, as the coparticipants instead suggest that the situation is good for Malia (lines 50, 52, 54). Malia, therefore, expands with further negative stance displays (lines 44–46, FG.7.18; lines 51, 53, 55, 57), and she eventually initiates an expansion invoking her obligation to speak to everyone in French (lines 58, 60). Also, this assertion is strengthened through verbal, paraverbal, and embodied means (see extreme-case formulation, marked prosody, and gestures, FG.7.19–7.20) and completed with another non-lexical vocalization expressing frustration. When Malia moves into yet another expansion (line 61), Zarah requests a clarifcation about Malia's job (line 62), which momentarily steers away the focus from the complainable as Malia clarifes what her work as a doctoral assistant implies (omitted lines).

Malia's attempts to gain her coparticipants' afliation and sympathy continue until Theo fnally ofers a verbal recognition that he understands that the situation is difcult for Malia and suggests that it will be better in the future (lines 161–163). Malia initially receipts this with some display of skepticism, fashing her eyebrows, tilting her head while fipping her hand palm up as if indexing the hopelessness of the situation (FG.21). She then raises and crosses her fngers (FG.7.22) as if suggesting that she hopes for the situation to be better in the future, and Theo afliates by copying her gesture (FG.7.23) and verbally 'glossing' it (line 170, see Keevallik, 2013). The participants then laugh together before closing the sequence (lines 172–175). With these actions, Theo thus fnally displays some sympathy for Malia, whereas Malia herself shows troubles-resistance toward the problematic situation by demonstrating her ability to laugh at it.

Similar to Excerpt 7.6, the sequential development of this sequence shows the participants' relatively new relationship and unfamiliarity with each other. Malia's sequence-initiating announcement (lines 3–4) signals a potential common knowledge about the fact that she would start her job, but the following sequence establishing the place reference indexes the participants' low epistemic status concerning Malia's workplace. This is further underlined by Zarah's question about what Malia actually works with (line 62). Zarah's too-early bright-side response (Holt, 1993) to Malia's telling (line 50) and the relative difculty with which Malia gets her coparticipants to recognize the legitimacy of the complaint exemplify the typical lack of afective afliation by the coparticipants at this point in their relationship. This may be due to diferent factors (e.g., the coparticipants might not fully understand or simply disagree with Malia's arguments), but nevertheless, it contrasts considerably with some of the complaint responses in the following semester (cf. Zarah's afliative assertion in Excerpt 7.10 that learning French is *not* the most important thing for Malia at work, which is subtly tuned to Malia's particular situation).

To sum up the beginning phase of the recordings, Malia's complaints about her workplace and her coparticipants' responses to these refect both the participants' still novel relationship and process of 'coming together' as a group and their status as elementary speakers. The complaints are all closely tied to Malia's difculties with French. Malia's recurrent topicalization and complaints about her difculties at work indicate that she starts treating the conversation circle as a 'safe space' to tell about her difculties in a context where the coparticipants, also L2 speakers and university students/collaborators, are bound to understand her problems. In doing so, she orients herself and the coparticipants as belonging to the same membership category, and her complaints can be seen as a contribution to the process of establishing a common 'we' (Hanna, 1981). That the participants are only at the beginning of such a socialization process is visible in the coparticipants' generally limited contributions to the sequences. In most cases, Malia volunteers her complaint initiations in frst position (as in Excerpt 7.7). In the two exceptions, the coparticipants' initiating actions are linguistically minimal, merely nominating Malia to talk about an already established topic without displaying any knowledge of potential troubles (as in Excerpt 7.6). The coparticipants support the progressivity of talk by helping Malia to complete word searches, but their expressions of afliation normally consist of minimal linguistic and/or non-verbal contributions (repetitions, vocalizations, embodied conduct) that do not claim independent epistemic access (Heritage & Raymond, 2005) to Malia's situation. At the same time, the limited contributions from the coparticipants refect their status as elementary speakers. Repetitions of Malia's own assessments, non-lexical vocalizations, laughter, and embodied conduct ofer ways to express some level of afliation without relying on sophisticated linguistic resources (as shown in Section 5.3).

### *7.2.2 Spring semester of 2017 (months 4–9) – same coparticipants*

During the second semester, Malia expresses a negative stance about her workplace in 11 sequences. Not all of these sequences develop into complaints, as in some cases the negative stance displays are limited to a series of critical remarks. In the fall of 2016, Malia's complaints typically focused on her difculties with French, and references to her professor occurred primarily in accounts for the negative stance expressions, working as 'evidence' for the complaint. By contrast, in the spring of 2017, Malia starts developing complaints specifcally *about* the professor and about the professor's actions that make Malia's work conditions complaint-worthy. The complaints, therefore, become more personal. The coparticipants nevertheless start participating more actively in Malia's complaints. They show an interest in Malia's personal issues by inquiring about them (Excerpt 7.9) and by more actively contributing to the sequential development (Excerpts 7.8 and 7.10), thereby demonstrating increased epistemic access to Malia's personal situation and more explicitly expressing their sympathy with her (I will return to the issue of how the changing nature of the complainable may afect recipient participation).

One way the coparticipants contribute more actively to Malia's complaints is by ofering accounts on her behalf, by which they display their familiarity with and understanding of Malia's situation. Excerpt 7.8 illustrates this point. Here, Malia initiates a news telling about a question-answer session she had to participate in the same day at work (lines 1–5), and she develops the sequence into a complaint about her difculties with speaking French. Our focus is specifcally on Theo's turn in lines 16 and 18, in which he ofers a candidate account for Malia's negative assessment in line 13.

*Excerpt 7.8* 'Session' (Mer1\_2017–01–11)

 Following Malia's negative assessment of the event as 'very very difcult' (line 13), Theo ofers a candidate account on Malia's behalf: >*parce que tu devais-< tu- tu devais parler e:::n en français* ('because you had to- you- you had to speak in in French', lines 16, 18). This account ofers a potential explanation for why the session was so difcult for Malia. Malia confrms and builds upon the candidate account by repeating *en françai::s* ('in French', line 19) and ofering a summary assessment consisting of *et donc* followed by a loud in-breath non-lexical vocalization (line 20) and the covering of her face with her hand (FG.7.24). She thus embodiedly expresses the unreasonable difculty encountered at the question-answer session before again verbalizing it (lines 20–21). She then develops the complaint by elaborating on why it was so difcult for her (lines 22 and onward).

By ofering the candidate account for Malia's problem announcement, Theo demonstrates his familiarity with Malia's recurring work difculties related to her obligation to speak French. As shown in Section 5.2, complaint initiations are often done through high-grade negative assessments followed by accounts among elementary speakers. By preemptively producing the next relevant action following Malia's negative assessment, Theo contributes actively to the development of the sequence into a complaint, drawing on verbally more elaborate resources than he did when responding to Malia's complaints in the fall of 2016. Theo's shared interactional history with Malia hence seems to facilitate his participation in the complaint construction through the use of a practice typically deployed by more advanced speakers.

 That the complainable eventually becomes a shared concern in the group is manifested in the fact that the coparticipants start initiating talk about Malia's troubles and difculties related to work, thereby ofering opportunities for Malia to complain (cf. Ruusuvuori et al., 2019). Excerpt 7.9 takes place two months after Excerpt 7.8. It shows the frst time a coparticipant asks Malia about the situation with her professor. After the closure of the prior sequence and a longer gap (line 1), Zarah turns to Malia and asks *comment ça va avec ta prof?* ('how is it going with your professor?', line 2). In response, Malia uses high-grade negative assessments to express the difculty of her situation and subsequently develops a long complaint about all the problems she has related to her workplace and boss, particularly as it concerns the PhD registration application that her professor has to approve (the excerpt only shows the beginning of the sequence).

*Excerpt 7.9* 'How's it going with your prof A' (Mer1\_2017–03–08)

The format of Zarah's question in line 2, particularly with the use of the feminine pronoun *ta* in *ta prof* ('your professor'), shows Zarah's orientation to the topic as a matter of which both she and Malia have prior knowledge (*ta prof* referring to the specifc, known female professor). While the question in some circumstances could be interpreted as a question about the professor's well-being, in this situation, it is clearly an inquiry about how it is going with Malia's dealings with the professor. Through this 'itemized news inquiry' (Button & Casey, 1985), by which Zarah asks for news about a known troublesome situation, Zarah invites Malia to talk about troubles (Jeferson, 1988). Malia accepts this invitation by stopping her movement toward Zarah (line 3), and instead, she covers her face with her left hand (line 4, FG.7.25–7–26) and utters a sigh (line 5). Zarah's instant laughter (line 6) shows the recognizability of Malia's vocal-embodied conduct as the expression of a negative stance (cf. Excerpt 5.4, for a similar case) and Zarah's anticipation of the upcoming

answer. After removing her hand from her face, Malia provides a frst verbal answer in a low voice, *terrible* ('terrible', line 8), which she repeats (line 13) after the coparticipants' receipts (lines 9, 10, 12), again embodying trouble by leaning her head in her hand (FG.7.27). In overlap with Theo's attempts to elicit an account for why it is terrible (lines 14–16), Malia expands the sequence (lines 18–20 and onward).

This excerpt hence shows how a coparticipant talks Malia's recurrent complainable into relevance through a K+ question (Heritage, 2012) that indexes a high epistemic stance and thereby ofers her an opportunity to complain. In displaying her interest in Malia's personal issues and their development over time, Zarah orients to the longitudinal nature of the participants' relationship and to the relevance of talking about and sharing personal problems. At the same time, by inquiring about Malia's situation after a lapse in the conversation (see 4.8 seconds of silence in line 1), Zarah also shows initiative in maintaining the progressivity of talk and takes responsibility for sustaining the conversation (cf. Kim, 2017; Nguyen, 2011). Similar to what Kim observed in an L1-L2 conversationfor-learning setting (see Excerpt 2.5), a particular topic (here Malia's workplace situation) "seems to have gained an omni-relevant status" which "can be broached at any topic-bounding sequential environments" (2017: 98). Since the coparticipants by now are familiar with Malia's eagerness to talk about this topic, it works as a 'safe' solution for restarting talk after a conversational lapse. Theo's contribution to the sequence manifests itself in his attempt to elicit an account from Malia, showing his agency in advancing the relevant actions following Malia's negative stance expressions.

As a last example testifying to the participants' increased joint attention and contributions to Malia's complaints about her workplace situation and to their growing personal relationships, Excerpt 7.10 shows how Zarah displays her sympathy with Malia through epistemically strong assertions about Malia's needs. The excerpt takes place approximately 20 minutes after Excerpt 7.9. Malia has told Zarah about her difculties getting ofcially admitted as a PhD student due to her professor's unwillingness to accept her foreign university certifcates. Just before the start of the excerpt, Malia suggested that her professor's reluctance to help may be due to Malia's difculties with French. Zarah objected to this reason, referring to the fact that most of Malia's research is in English. In lines 1–3 and 5, Malia agrees with Zarah's objection.

*Excerpt 7.10* 'How's it going with your prof B' (Mer1\_2017–03–08)


Zarah receipts Malia's agreement by nodding (line 6). As Malia is about to expand (lines 7–8), Zarah initiates a formulation of her own perception of Malia's situation – namely, that because Malia lives in Switzerland, she will learn French just by talking to people (lines 9–10, 13–16). The formulation is in declarative syntax, indexing a high level of epistemic certainty. By opening up her hands palm up (FG.7.28–7.29), she underlines the obviousness of the assertion (Kendon, 2004; Marrese et al., 2021), and through circling gestures, she further animates and displays her afective involvement in the talk. She then adds *mais maintenant tu .hh tu dois commence:r (0.3) °dans° ton*  *doctora:t ton °étude°* ('but now you must start in your doctorate your study', lines 16–17), asserting her own understanding of Malia's needs. Through prosodic stress on *commence:r* ('start') and by tapping her right hand in her left hand, she enhances the strength of these assertions. The declarative format and deontically strong *tu dois* ('you must') again index a high level of certainty and entitlement (Stevanovic & Peräkylä, 2012) about something that actually lies within the epistemic and deontic domain of Malia herself. Through these actions, Zarah shows not only her understanding of Malia's situation but also her strong afliation with her and produces something to which Malia herself can agree, which Malia emphatically does (lines 19, 21).

 In overlap with Malia's second *exactement* ('exactly', line 22), Zarah embodiedly invokes the consequence of the situation with *donc* ('so', line 21) followed by the gesture of opening up her hands palms up, again expressing obviousness (FG.7.30). She then claims that that is what is most important (line 24), not the French (lines 26–27). Malia again agrees and shows her afliation with Zarah by repeating *c'est pas le français* ('it's not the French', line 28) in a prosodically similar manner as Zarah (with stress on *pas*) and by copying her open hand gesture on this last word (FG.7.31–7.32; see Couper-Kuhlen, 2012, for afliative displays through prosodic matching). Malia subsequently expands and continues expressing her strong discontentment with the situation (lines 29–30 and onward).

Whereas Zarah in earlier sequences has displayed afliation with Malia through agreement tokens and similar rather small contributions, in this sequence she expresses her sympathy by making an epistemically and deontically strong assertion about Malia's situation and needs. Zarah's assertions are based on her interactional history with Malia and her familiarity with Malia's difculties related to her PhD work and studies (which can be contrasted with Zarah's responses in Excerpt 7.7). The strong epistemic and deontic stance-taking is also indicative of the deepened social relationship between the participants as compared to the beginning of the recordings. What Zarah formulates about Malia is hardly something a stranger says to another; it is the result of a longer history between the participants and shows a level of intimacy established between them. That Malia recognizes Zarah's actions as displays of afliation is not only observable on the verbal level but also in her gestural and prosodic matching of Zarah's conduct.

This section has shown how the participants' process of coming together as a group manifests itself in the recurrent activity of talking about Malia's workplace problems. During this time, Malia's complaints become less concerned with her ability to learn and speak French (Excerpt 7.8) and more related to her problematic relationship with her professor (Excerpts 7.9–7.10). The participants now show a joint, sustained interest in Malia's personal difculties and orient to talk about these difculties as a relevant dimension of their relationship (cf. Berger & Fasel Lauzon, 2016). At the same time, the participants' profciency in French increases.

The efect of this concurrent change is visible both in the ways in which the complainable is talked into relevance and in the coparticipants' afliative displays. Now the coparticipants invite Malia to talk about her complaintworthy situation through inquiries about the complainable (Excerpt 7.9). They also help co-constructing complaint sequences by providing and eliciting accounts (Excerpts 7.8–7.9) and by making strong assertions about Malia's needs (Excerpt 7.10), showing their afliation and sympathy with Malia in explicit ways (albeit Zarah more so than Theo).

### *7.2.3 Fall semester of 2017 (months 11–15) – new coparticipants*

In the third semester, Malia meets with new coparticipants (with the exception of Zarah, who attends two of the same conversations as Malia). Malia's French profciency has now increased considerably, to the B2 level. The new coparticipants are at similar levels, ranging from B1 to C1 (intermediate to advanced level). Because the participants are new to each other, in the frst few recordings, they spend time talking about themselves and getting to know each other, like Malia did with her coparticipants one year earlier. The topics of French skills and French learning frequently arise, and Malia's complaints and negative talk about her work (fve sequences) emerge primarily in connection with talk about difculties with French. Also similar to the early examples, Malia's references to her professor occur mostly as a means to support complaints about work and French instead of as complaints *about* the professor. Thematically, these sequences are thus similar to the sequences in the frst semester.

A notable diference between the fall of 2016 and the fall of 2017, however, is that the coparticipants in the fall of 2017 more actively contribute to the development of the complaints. These sequences are thus *structurally* more similar to those in the spring of 2017 than those in the fall of 2016. The participants' limited interactional history nevertheless has an impact on the type of contributions ofered by the coparticipants. While several of the complaints are initiated in second position, only one of the sequences includes the type of K+, 'itemized news inquiry' shown in Excerpt 7.9, and this is produced by Zarah in one of the two recordings in which she participates. Instead, Malia initiates complaints and troubles talk in response to K−, neutral questions or questions that rely on knowledge inferred from the immediately preceding talk. Similarly, in responding to Malia's complaints, the coparticipants typically express their alignment and afliation based on the information provided by Malia in the same sequence.

 In this section, I analyze only one but rather long excerpt from the fall of 2017 (Excerpt 7.11). This sequence, from the participants' frst meeting, illustrates how Malia, in response to a series of neutral questions from Jordan, invokes her difculties with speaking French at work. To show sympathy, Jordan makes a comparison with himself and his own struggles with French, but his comparison does not match Malia's experience and therefore fails as an afliative move. Malia consequently pursues the complaint with specifc examples from her workplace and by invoking her professor's high demands on her until Jordan and Javier better succeed at showing their understanding of Malia's particular situation and sympathizing with her. For ease of reading, the excerpt has been divided into four parts (Excerpts 7.11a–d). In Excerpt 7.11a, after the closing of a prior sequence in which Javier has been telling about himself, Jordan asks Malia about her PhD studies (line 3). His questions (lines 3, 6, 13) lead Malia to initiate a longer telling about herself, eventually invoking her difculties associated with working in French.

*Excerpt 7.11a* 'Bachelors' (Mer1\_2017–08–23)


Through his series of biographical questions (Svennevig, 1999), Jordan invites Malia to tell about herself and her study background (lines 3, 6, 13). Jordan's questioning is typical 'frst encounter conduct' used by the more advanced participants in my data to get to know each other in the frst few meetings (among less advanced participants, initial encounters typically involve longer tellings by one speaker at a time with limited intervention from coparticipants). Jordan's assumption that Malia is a PhD student is likely based on some information he has received just prior to or implied from the conversation so far. The basic nature of the questions shows Jordan's unfamiliarity with Malia and her professional circumstances. While directing the topic to Malia's studies, the questions do not in any way orient to any troubles related to Malia's studies or work.

In answering Jordan's question about the location of Malia's master's studies, Malia adds that all three degrees were in English (omitted lines). The topicalization of language works as a steppingstone into an account for her current need to learn French fuently (lines 34–35) and into talk about her difculties at work. To exemplify her need to speak French fuently, Malia tells the coparticipants about a question-answer session with her students she had the preceding week (lines 37–40). She assesses this session as 'very difcult' (line 44), especially since the bachelor's students speak French 'very very fast' (line 53). The description of the students' talk is accompanied by embodied conduct, enhancing the verbal expressions and indexing them as negative (see shaking, waving, and rhythmical hand gestures, and facial expression conveying frustration, lines 46–47, 52–53, FG.7.33–7.34).

Malia's actions are understood by Jordan as a complaint about the students, as seen by the fact that he produces a non-serious criticism of young people (*les jeunes non*, 'the young no', line 54, see also accompanying headshakes) to align with Malia on the surface level of her actions without seriously engaging in co-complaining. Malia, however, responds seriously to Jordan's alignment (line 55) through an embodied expression of a negative afective stance (FG.7.35). She then asserts that she always has to guess what to say to her students (lines 59–74), which she assesses as sometimes 'really difcult' (line 75). Perhaps due to the lack of immediate responses from the coparticipants (line 76), Malia returns to the issue of her French learning (lines 77–78), suggesting that she thinks that she will be able to improve after having fnished her dissertation (lines 80–81, 83). She delivers the end of the turn with a smiley voice and then starts laughing (line 83), thereby showing her ability to take her difculties lightheartedly (Edwards, 2005; Jeferson, 1984b). Jordan afliates by laughing with her (line 84), and Javier ofers his sympathy by expressing his certainty that Malia will indeed improve after fnishing the dissertation (lines 85–87).

After some more laughter by Malia (line 88), Jordan produces another follow-up question to Malia, thereby supporting the continued development of the topic and displaying his further interest in Malia's situation. He uses her answer to produce a my-side story (Selting, 2012), by which he claims similarity with Malia's situation, but this story fails as an afliative move:

```
Excerpt 7.11b 'Bachelors' (Mer1_2017–08–23)
```
Jordan's question in lines 89–90 builds on Malia's assertion that she will learn to speak French after her doctorate, and it shows his interpretation that it is during the PhD that Malia will learn because she has an opportunity to speak French every day. This is not the case, however. After Malia's disconfrmation (line 92), she explains that some of her colleagues only speak English and that, although others speak French, they prefer speaking English with her because it is easier (omitted lines). Through confrmation tokens and candidate completions of Malia's talk, Jordan and Javier display their understanding (omitted lines), after which Jordan adds that there are a lot of foreigners, especially when you are a PhD student (lines 115, 117–118), and that 'it' (presumably English) is the language of the university (line 122). These last assertions work as grounds for Jordan's upcoming telling about himself, in which, in an

attempt to show sympathy with Malia, he explains why it is difcult for him to learn French.

After claiming similarity with Malia's situation (line 123), Jordan explains that he only has to start speaking French if he wants to (lines 126–127, 129–130) and that he needs to make an efort to do that while others can just speak English with him (lines 133–136). He concludes that you can speak English to everyone at the university (lines 138–139). Malia responds only minimally during Jordan's telling (lines 125, 128, 135, 137), and after Jordan's conclusive remark, she utters a quiet *oui* ('yes', line 141) while Javier ofers the slightly more afrmative *oui oui* ('yes yes', lines 142). The lack of afliative displays from Malia is understandable, given that Jordan's my-side telling about his difculties with learning French fails to capture the difculties encountered by Malia (see Selting, 2012, for successful and unsuccessful second complaint stories). For Malia, the main problem is not that she can get by with English and needs to make an efort to speak French. Quite the opposite, her complaint is about her difculties meeting the high expectations about her French. She, therefore, objects to Jordan's argument by upgrading and expanding her initial complaint about her difculties at work:


In her expansion, Malia uses adverbial intensifers (line 146), extreme-case formulations (lines 153, 154), marked prosody, a nonlexical vocalization (line 148), animated hand gestures, frowning, and rhythmical head nods (lines 154–156, 160) to portray her work demands as something out of the ordinary (and thus complaint-worthy) that requires her to speak French whether she wants to or not. Jordan responds through small acknowledgment tokens, nods, and some laughter (lines 149–150, 152, 157), and Javier eventually formulates the gist of Malia's argument in his own words, suggesting that learning French is a need that she has (line 164). In doing so, Javier afliates with Malia by acknowledging the grounds for her complaint and showing his understanding of her difculties, which Malia confrms (line 165).

As none of the coparticipants self-selects, Malia again expands the sequence by expressing the implication of her difficult situation, that she is always stressed (line 169). She then abandons her continued turn and starts laughing (line 170). Jordan joins her laughter (line 171), while Javier agrees with Malia (line 172), asserting that French always is like that, *tu arrêtes pas de te stresser* ('you don't stop stressing', line 173). In agreeing and upgrading Malia's assertion as a general fact, Javier affiliates with Malia by showing that she is not alone with her difficulties. He produces what appears to be an assessment (line 174), after which he shakes his head in silence (line 175). Finally, he contrasts the negative stance expressions with a positive assessment of the French language (line 176). This last assessment works as a bright-side contribution (Holt, 1993), by which Javier displays some resistance toward the struggles that they are all going through. In doing so, he nevertheless minimizes Malia's difficulties, and this prompts her to initiate another objection in which she insists on her problems:

*Excerpt 7.11d* 'Bachelors' (Mer1\_2017–08–23)


Malia's objection takes a conventional *yes-but* dispreferred turn-design (Pomerantz, 1984) and invokes the pressure (lines 180–182) that comes with the obligation to speak French. Jordan now agrees (line 185), whereas Javier confrms the word *pression* that Malia had been searching for in a word search (line 186). Malia then embodiedly expresses the negative afective stance associated with the pressure of having to speak French by completing the verbal fragment *et donc* with a non-lexical vocalization and a falling hand gesture ('and so', lines 187–188, FG.7.36), and Jordan aligns through repeated *ouais* ('yeah', lines 189, 191).

Having received her coparticipants' claims of understanding, Malia expands with a specifc example, reporting on what happens when her professor calls her on the phone (lines 192–194). At this point, Jordan more elaborately afliates by verbally completing Malia's embodiedly completed negative assessment (lines 195, 197, FG.7.27–7.38) with >*ouais c'est plus difcile< au téléphone* ('yeah it's more difcult on the phone', line 196). Malia agrees with a long *oui::* ('yes', line 198) while Jordan repeats his turn (line 199), further expressing his afliation with Malia. Javier also agrees (line 200), after which Malia reenacts in a humorous manner how she sounds like she is climbing up a mountain when telephoning with the professor (lines 202–203). This reenactment is successful in further engaging the coparticipants' participation (Sidnell, 2006): Jordan frst laughs loudly (line 204) and then himself adds to the story by enacting heavy heartbeats (line 206). By building on Malia's story and contributing to its development, Jordan both aligns and afliates with Malia, and Malia shows her appreciation through strong tokens of agreement (lines 207–208). She then upgrades the troubles talk by adding another difculty: that she cannot understand all the words (lines 210–211). Javier fnally builds on this to invoke yet another difculty that occurs when speaking with French-speaking people (lines 214–220). At this point, the coparticipants thus all agree on and co-construct a complaint about the difculties associated with talking French on the phone.

This long excerpt has shown typical features of Malia's complaints about her workplace and her obligation to learn French taking place in the fall of 2017 with new coparticipants. The sequence indexes the participants' novel relationship in several ways. Topically, the complaint emerges from 'frst encounter talk' after a series of biographical information-questions from Jordan about Malia's studies and professional situation. The complaint takes its starting point in Malia's difculties with French and not in Malia's more personal problems related to her relationship with her professor. Jordan's unsuccessful attempt to show sympathy by invoking his similarity with Malia further demonstrates the participants' limited familiarity with each other, as his second story fails to accurately capture Malia's situation. After several expansions with specifc examples and strong stance expressions, Malia nevertheless manages to secure afliation from the coparticipants. Similar to the coparticipants' expressions of afliation later in the spring semester (as in Excerpt 7.9), Javier and Jordan show their afliation through declarative formulations (e.g., lines 164–165) and assessments (e.g., line 199) that index the participants' epistemic independence. These assertions are not based on shared interactional histories with Malia but on their current interaction with her.

In all, Malia's complaints and negative stance expressions about her workplace and the way this complainable is talked into relevance and constructed during the fall of 2017 refect both the new participant framework and the participants' estimated level of French.2 While the lack of shared macro-level interactional histories in certain respects makes the complaints similar to those in the fall of 2016 (fewer sequences than in the spring of 2017, topical focus on French skills, initiations through neutral information questions, etc.), in other respects they resemble those in the spring of 2017 (e.g., active contributions from the coparticipants, verbally more elaborate displays of afliation). This fnding suggests that the coparticipants' generally higher level of L2 profciency allows them to participate actively in the sequences despite the lack of prior knowledge about Malia or her personal and professional situation.

#### *7.2.4 Malia: summary and intermediate discussion*

As shown earlier and summarized in Table 7.1, there is a change over time in terms of topical focus, how the complainable is talked into being, and how it is constructed by the coparticipants. This multifaceted concurrent development ties into both the participants' profciency in French and other types of socialization processes, including the nature of the participants' relationships and their shared interactional histories (or absence of such histories).


*Table 7.1* Overview of the interactional history of the complainable over time


(3) Coparticipant contributions: In the fall of 2016, the coparticipants contribute only to a limited degree to the sequential development of complaints, and they show their (limited) sympathy with Malia through linguistically quite simple means. With time, the coparticipants produce verbally more elaborate responses, including epistemically and deontically strong assertions expressing afliation. Although this change in part may refect the slightly evolving nature of the complainable, whereby, for instance, more personal complaints may be treated as more complaint-worthy and thus call for somewhat diferent recipient responses, the fact that the heightened level of coparticipant engagement persists in the fall of 2017 indicates that other factors also come into play. As seen in Section 5.3, the complaints among elementary speakers generally involve less active participation from coparticipants, whereas the complaints among more advanced speakers are co-constructed to a higher degree. Similar to the analysis of Suresh's stance-taking in Section 7.1, the analysis here sheds light on ways in which less advanced speakers may contribute more actively to complaint sequences by relying on their shared interactional history with the complainant (as in the spring of 2017). More advanced speakers (in the fall of 2017) contribute actively to the sequences without shared interactional histories, relying on their generally higher level of L2 IC.

### **7.3 Discussion**

In 2004, Brouwer and Wagner suggested that "studies of language learning have to be sensitive to the ways in which participants establish and nurse social relations" (p. 35). This argument came from the perspective of situated learning theory, conceptualizing learning as socially situated and intrinsically linked to speakers' participation in social encounters (Lave & Wenger, 1991). In most studies on the development of L2 IC, which typically do not adopt any exogenous learning theory, the analytical emphasis has been on speakers' systematic methods-for-action and their development over time without much consideration of changes in the social relationships between the participants (but see Greer, 2019; Kim, 2017; Pekarek Doehler & Berger, 2019; and Skogmyr Marian, 2018). This was also the approach I adopted in Chapters 5 and 6. In this chapter, I have attempted to shed some light on the role of socialization processes in the development of complaint practices with the help of two longitudinal case studies. Given the exploratory nature of these studies, they should be seen merely as a frst step toward more in-depth inquiries about the relationship between the development of interactional skills, shared interactional histories, and evolving social relations. The fndings of the two studies are complementary. Importantly, they converge on the

following point: The way speakers accomplish complaints is inextricably tied to the nature of the relationship between the participants and to what the participants know about each other from prior encounters. As the relationship and the participants' shared interactional history develop, so do the participants' complaint practices.

 Case study 1 shows how Suresh and Aurelia's shared interactional history served as an interactional resource for Suresh, an elementary speaker of French, to produce recipient-designed assessments about the weather, recognizable as complaint profers, to efectively advance the conversation past the exchange of greetings. While at frst (Excerpt 7.2) Suresh's minimal contributions to the sequence led to rapid abandonment of the complaint by his coparticipant, in the latter case (Excerpt 7.5), Suresh's slightly more heightened involvement ofered an opportunity for the participants to exchange aligning and afliative stances. The fndings are similar to those of Brouwer and Wagner (2004) and Greer (2019), showing how participants engaging in repeated encounters draw on their prior joint experiences to establish recipient-designed conversational routines. Suresh and Aurelia's longitudinal monitoring of and invocations of each other's stances toward the outdoor temperature and Suresh's repeated reuse of linguistic material deployed in prior encounters furthermore demonstrate the participants' own orientations to the longitudinal nature of their relationship. Suresh and Aurelia show that they remember past exchanges with each other and that it is not the frst time they engage in topically similar talk. It seems that for Suresh, who has a limited linguistic repertoire in French, the reuse of his coparticipants' talk also constituted an important resource for action-formation, ofering afordances for efective participation in the interactions (for repeated reuse of gestures for the achievement of intersubjectivity, see Eskildsen & Wagner, 2013). This observation concurs with an interactional usage-based perspective on L2 learning (Eskildsen, 2020; Pekarek Doehler & Eskildsen, 2022) that underscores the link between local interactional experiences and longterm instantiation of patterns of language use. Moreover, Suresh's expression of contrasting stance from one interactional encounter to another, designed to match his coparticipant's stance, provides emic evidence for the interpersonal and relational purposes of complaining (Boxer, 1993; Günthner, 1997; Hanna, 1981), something that is further demonstrated in case study 2.

Case study 2 shows various ways in which cumulative shared knowledge and the nature of Malia's relationship with her coparticipants afected her complaints about her workplace. By comparing, on the one hand, Malia's complaints with the same participants over time and, on the other hand, Malia's complaints with new, more advanced coparticipants, the analysis highlighted how less advanced speakers may rely on their growing interactional histories with their coparticipants to produce the kind of context-sensitive complaint contributions that more advanced speakers

do without having much prior knowledge about their coparticipants, such as verbally more elaborate ways of showing afliation and sympathy. The ability to recipient-design context-specifc actions is a key feature of increased L2 IC (Pekarek Doehler & Pochon-Berger, 2015). By, for example, starting to produce status update inquiries that target known troubles and by ofering accounts for negative assessments, the participants indeed showed their growing capacity to recipient-design their actions based on their shared interactional history (Greer, 2019; Kim, 2017). This fnding resonates well with Pekarek Doehler and Berger's (2019) observation that the growing 'relational material' that comes with more extended joint interactional histories afects micro-level interactional practices (in their case, repair practices) and should be taken into account when discussing the development of L2 IC. Moreover, when it comes to complaining (and probably other delicate activities), my fndings suggest that it is not only the participants' shared *knowledge* that plays a role but also how readily they display agency and authority in relation to each other, something that likely relates to how the participants see the nature of their relationship. For example, compared to earlier in the data, Zarah's deontically strong assertions about Malia's situation and needs in the spring of 2017 and Malia's acceptance of these showed the participants' willingness to engage as confdants rather than merely conversation partners. These observations hence support the idea that complaint practices are afected by the status of the relationship between the participants (see Chapter 3) and again underline the important role of evolving social relationships in the development of L2 IC.

As pointed out by Pekarek Doehler and Berger (2018: 575), the fact that we cannot separate speakers' changing practices for action from their evolving social relationships is "not a problem of analysis, but a problem of interpreting the fndings". By conceptualizing the development of IC as a holistic process that involves adaptation to constantly changing social circumstances, it is less interesting to try to isolate, for example, the development of linguistic competence from other aspects of language learning than to see it as an integrated whole. In line with this, the fndings of both case study 1 and case study 2 indicate that the ability to draw on shared interactional histories should be considered a key dimension of the development of L2 IC. In both studies, the participants' use of knowledge from prior interactional encounters resulted in a higher level of co-construction of complaints and enhanced participation in the interactions more generally. The speakers' ability to mobilize such knowledge in context-sensitive ways thus demonstrates increased L2 IC.

The two case studies show the role complaining (Boxer, 1993; Günthner, 1997; Hanna, 1981) and talk promoting emotional displays (Berger & Fasel Lauzon, 2016) can play in building social rapport and strengthening relationships. As seen throughout the analysis, participants recurrently talk their interactional histories into relevance (Mondada, 2018; Skogmyr

 Marian, 2018; Voutilainen et al., 2018), and they use their increasing common ground to accomplish recipient-designed and context-specifc social actions (Deppermann, 2018; Norrthon, 2019) that allow them to exchange displays of afliation with each other. In doing so, they orient to the longitudinal nature of their relationship and to the relevancy of strengthening such a relationship over time. The fndings contribute to the literature on how participants create conversational routines over multiple encounters (Brouwer & Wagner, 2004; Greer, 2019) and how such routines may help L2 speakers participate more actively in interaction (Eskildsen, 2021a; Waring, 2013; Watanabe, 2016). More broadly, they support the arguments made in language socialization research (e.g., Schiefelin & Ochs, 1986) about participation as a key driver and manifestation of increased interactional skills (see also Brouwer & Wagner, 2004; Cekaite, 2007; Lave & Wenger, 1991; Nguyen, 2011), but they do so without recourse to exogenous learning theory. While exploratory in nature, the fndings ofer a small contribution to our understanding of the role of socialization processes in the development of L2 methods-for-action.

# **Notes**


# **8 Discussion of results and perspectives**

The overall aim of this book is to enrich our understanding of the longitudinal trajectories involved in the development of L2 IC. To do so, it has presented three empirical sub-studies that addressed diferent aspects of L2 speakers' engagement in the activity of indirect complaining (Chapters 5–7). In this fnal chapter, I discuss the implications of the results of the three sub-studies for our understanding of the development of L2 IC and L2 learning more generally (Section 8.1). I also highlight the contributions of the study to our knowledge about complaining in interaction (Section 8.2). I conclude by refecting on possibilities for future research including applications of the fndings within the feld of language education (Section 8.3).

### **8.1 Understanding L2 interactional competence and its development**

The research presented in this book addressed several gaps in the L2 IC literature. No prior study in the feld has examined indirect complaining, despite the ubiquitous nature of this activity in both institutional and ordinary interactions. The focus on a complex interactional activity also provided an opportunity to shed light on a range of dimensions of L2 IC that have received only limited attention so far, such as the coconstruction of larger interactional projects (Levinson, 2013) and joint stance-taking. In addition, by adopting a multimodal analytical approach, the study set out to deepen our knowledge about the role of embodiment in the development of L2 IC, another under-researched area. Considering the social-relational dimensions of complaining, the study also lent itself to an exploratory investigation of how L2 complaint practices are afected by changing social relationships and the accumulation of shared knowledge. In three sub-studies, I therefore examined (1) the structural organization of L2 complaints, (2) interactional resources used in L2 complaining, and (3) the way in which change in L2 complaint practices intersects with larger socialization processes. Instead of summarizing and discussing the fndings of each sub-study separately here, I will present seven implications of the cumulative results for our understanding of the development of L2 IC. Before addressing these implications, however, I return to the important question of how, from a CA perspective, the observed changes over time can be conceptualized in terms of L2 development and learning.

As discussed in Section 4.1, the methodological principles of ethnomethodologically inspired CA (EMCA), which presuppose a data-driven, participant-relevant perspective, pose great challenges for longitudinal studies. To ensure a basic emic validation (Clayman & Heritage, 2021) of my analyses, I have conducted turn-by-turn, sequential analysis of the data (see Section 4.1). The emic perspective on change and development is more problematic, as participants rarely ostensibly orient to change over time or development in their interactional practices. Although Chapter 7 showed examples of speakers actually invoking their shared interactional histories, it would be difcult – and unfortunate for the feld of CA-SLA, in my view – to limit longitudinal studies to instances in which people explicitly talk about their past.

The fact that coparticipants hold each other accountable for interactionally competent conduct may nonetheless provide emic evidence for the interpretation of change in terms of the development of interactional competence (Deppermann & Pekarek Doehler, 2021; Wagner et al., 2018). In my data, certain observed changes, such as in participants' practices for introducing direct-reported speech (DRS) and reenactments, led to less repair over time and can therefore be discussed in terms of increased local recognizability (cf. Garfnkel, 1967; Garfnkel & Sacks, 1970). In other cases, it was not the recognizability of actions that was at stake for participants but rather their local efcacy. For example, high-grade negative assessments were typically efective means for elementary speakers to make their complaint initiations recognizable as such, yet with time speakers diversifed their initiation practices to be able to accomplish other complaint-relevant interactional work before launching overt criticism. The general 'permissiveness' observable in interactions among L2 speakers nevertheless challenges this idea since recipients often show greater acceptance of interactional troubles in L2 than in L1 talk (cf. Firth, 1996). Indeed, as seen throughout the analyses, linguistic errors, slow conversation pace, long word searches, and other phenomena that may be treated as accountable conduct in L1 interactions often go unaddressed by my participants and are thus constructed as orderly and 'normal' phenomena in the interactional setting at hand (although explicit orientations to language difculties and the participants' status as L2 speakers are also common, manifested, for example, in the many complaints about these issues).

So what warrants an interpretation of my fndings in terms of L2 development when it comes to change over time in conduct that was not initially treated as problematic by coparticipants? As discussed by Clayman and Heritage, CA's emic toolbox is not limited to next-turn-proof procedure: "[a] second closely related resource [i.e., emic tool besides next-turn-proof procedure] focuses on the understandings of speakers rather than recipients, as embodied in speakers' systematic deployment of interactional practices within specifc contexts" (2021: 235). Accordingly, in the context of L2 speakers, I would argue the following: As L2 speakers increase their interactional experience with the target language, they progressively accumulate new interactional resources and methods, and their *systematic, context-specifc use of these* ofers emic evidence that the speakers themselves orient to this as a qualitative change in conduct or else they would not change their systematic ways of doing things in the L2. My collection-based fndings indeed revealed such methodical, situated use of interactional practices at diferent points in time. They additionally reveal common trajectories of change across several participants, which indicates that the observed change over time actually refects developmental tendencies rather than *in situ* diferences due to local peculiarities.

A fnal point about the notion of learning is in order. So far, I have discussed my fndings in terms of *change* in interactional methods and the *development* of interactional competence. What *learning* refers to in the SLA literature has been heavily debated (see, e.g., Atkinson, 2011b, for an overview). Although many of my excerpts show learning as publicly observable behavior, the main aim of the study is, as mentioned in Section 4.1, to shed light on the long-term 'products' of learning. To do so, it has relied on the presumption that longitudinal development in L2 interactional methods refects the outcome of a learning process. This focus does not prevent the view of learning as a socially situated and observable process but merely limits the scope of inquiry to an overlapping phenomenon (see Eskildsen, 2020; and my fnal refections about the mutually constitutive nature of language use and learning in Section 8.3).

Based on the observed changes in complaint practices presented in Chapters 5–7 and in light of the just discussed interpretational issues, I now present seven consequences of the fndings for our understanding of the development of L2 IC and for L2 learning. These concern (1) the overall composition of conversational activities, (2) turn-taking management, (3) sequence and preference organization, (4) linguistic resources for action formation, (5) multisemiotic interactional competence, (6) socialization processes in L2 learning, and (7) successful accomplishment of social activities.

### *8.1.1 Stability in the basic composition of conversational activities*

The fact that the complaints at both elementary and upper-intermediate/ advanced levels comprise the same basic building blocks (Section 5.1) concurs with convergent observations about the main features of indirect complaints across several languages and cultures (e.g., Heinemann & Traverso, 2009; Ruusuvuori et al., 2019; Traverso, 2009; but note that most studies concern Indo-European languages in Western countries). Prior literature and my fndings suggest that indirect complaining boils down to a few core components: expressions of negative stance toward a particular complainable and evidence that accounts for such stance expressions and details the complaint-worthy person or situation (e.g., through complaint stories with DRS or reenactments). The expected response types to complaining are also similar across languages and L2 profciency levels, observable in participants' orientations to the relevancy of obtaining afliation or sympathy. The similarities in the basic building blocks of complaints imply that some things, such as participants' understandings of what it means to accomplish particular interactional projects or conversational activities, *remain the same* over time and across profciency levels as speakers learn an L2. To some extent, L2 speakers can thus rely on their experiences with conversational activities in their L1(s) and in other contexts as they engage in the same activity in the L2. This relates to the argument that certain aspects of IC, such as a general mastery of the generic organizational principles of social interaction (Scheglof, 2007), are part of a universal competence (Levinson, 2006) associated with the ability to participate in human interaction. L2 speakers instead have to 'recalibrate' certain aspects of this basic competence to refne their ability to efectively participate in L2 interactions (Pekarek Doehler & Pochon-Berger, 2015). When it comes to complaining, my fndings show that it is the way certain actions are accomplished and the relative frequency of occurrence of these in complaints – as well as the way in which participants coordinate their actions with each other – that change over time.

### *8.1.2 Turn-taking management: increased synchronization and co-construction*

The documented change in sequential positioning of complaint initiations (Section 5.2) and the increased co-construction of complaint sequences and joint complaining among more advanced speakers (Section 5.3) refect a change in the overall turn-taking organization of the interactions. Similar to what Sert (2019) observed for EFL peer interactions, the conversations in my data transition from what resembles a 'roundrobin format' to a turn-taking system that is similar to spontaneous L1 interactions, with faster speaker exchange and increased participation from coparticipants. The fact that upper-intermediate/advanced speakers more frequently ask each other questions and build upon each other's turns leads to more second-position complaints, increases the level of co-construction of complaints, and contributes to the higher proportion of joint complaints at upper-intermediate/advanced levels than at the elementary level.

The change in turn-taking organization can be explained by concurrent changes at several levels. On the one hand, it refects individual speakers' growing practices for coordinating their talk with that of others, for instance, by ofering recipient responses and other contributions that promote topic development. Research on these issues has highlighted precisely that speakers with low L2 profciency often have difculties providing timely and target-like response tokens and sustaining a conversational format. Over time, they diversify their repertoire of L2 response tokens, use these in more appropriate ways, and increasingly participate with collaborative turn-completions and other means that help co-constructing conversations (Dings, 2014; Sert, 2019). They increasingly also contribute to the topical development of conversations (Hellermann & Lee, 2021; Kim, 2017; König, 2019). Some aspects of the observed changes in turntaking hence pertain to a development in individual speakers' linguistic abilities, such as the learning of particular L2 response tokens (Kunitz & Yeh, 2019; Taleghani-Nikazm, 2019), and an increased ability to anticipate transition-relevant places (Sert, 2019) – or an enhanced capacity to verbally express stance (see Section 8.1.4).

On the other hand, the gradual transition toward more conventional conversational turn-taking cannot be attributed solely to individual speakers' development of precise interactional practices. It rather refects a concurrent change across several participants, resulting in a higher level of 'joint capacity' for synchronization and co-construction. As observed by Berger and Pekarek Doehler (2018), changes in conversational activities (in their case, storytelling) may also relate to changes in the relationships of the participants or to other types of socialization processes, such as decreasing interactional asymmetries between the participants (see also Pekarek Doehler & Berger, 2019). In my data, a decrease in interactional asymmetry over time resulted in more second-position complaints and joint complaining, as the participants, with the help of their growing L2 profciency, increasingly showed agency in asking coparticipants questions and upgrading other speakers' negative stance expressions. Such changes are not the accomplishment of individual participants but refect the participants' joint ability to coordinate and synchronize actions with each other.

### *8.1.3 Sequence and preference organization: diversifcation of methods*

The longitudinal changes in complaint initiations (Section 5.2) refect a progressive diversifcation of methods for launching larger courses of action and for managing delicate talk. My fndings show an increased tendency of upper-intermediate/advanced speakers to initiate complaints progressively, in ways that index the contingent, moral, and delicate nature of the activity (Drew, 1998; Heinemann & Traverso, 2009; Ruusuvuori et al., 2019). This was manifested in a change in sequence organization, whereby speakers increasingly introduced strong criticism or other overt negative stance expressions only after a longer work-up with subtle hinting at the complainable instead of immediately in the sequence initiation. These observations concur both with research documenting a diversifcation of practices for initiating longer sequences of actions, such as tasks and storytellings (Hellermann, 2008; Pekarek Doehler & Berger, 2018) and with studies showing L2 speakers' growing capacity for dealing with delicate and dispreferred actions like requests (e.g., Al-Ghatani & Roever, 2012) and disagreements (Pekarek Doehler & Pochon-Berger, 2011).

 In terms of task and storytelling initiations, more advanced L2 speakers have been observed to preface their upcoming talk through pre-sequences that prepare the grounds for the task or telling in various ways (Hellermann, 2008; Pekarek Doehler & Berger, 2018). Similar story prefaces have been documented for complaint stories in L1 talk (Selting, 2012). As for requests and disagreements, more advanced speakers, to a greater extent than less advanced speakers, preface such actions in ways that push back the dispreferred or delicate element further in the turn or sequence (Al-Gahtani & Roever, 2012; Pekarek Doehler & Pochon-Berger, 2011), similar to what has been observed in L1 talk (Pomerantz, 1984). Through these various 'pre-moves', speakers thus have the opportunity to better prepare their coparticipants for what is coming, to mitigate potentially delicate aspects of the incipient talk, and to minutely adapt their initiating actions to the coparticipants' responses, which in turn may increase the chance of obtaining afliative responses. More advanced speakers' tendency to initiate complaints in a stepwise manner that allows them to progressively escalate negative stance displays and to preemptively account for and establish the legitimacy of the complaint similarly works to enhance the possibility of obtaining afliation or sympathy from coparticipants. This development thus testifes to speakers' growing capacity for contextsensitive and recipient-designed talk (Pekarek Doehler & Pochon-Berger, 2015). In the case of complaints, the ability to adapt complaint initiations to the interactional context and the recipient may be particularly useful for the participants' engagement in such sequences in other settings and participant frameworks, where complaining might be associated more strongly with delicacy or dispreference than in peer interactions. As with the development in turn-taking management, the documented change in sequence organization also refects a rising capacity to put to use linguistic resources for specifc interactional purposes, such as grammatical constructions for projection (see also Skogmyr Marian, 2021b).

### *8.1.4 Linguistic resources for action: diversifcation and routinization*

The fndings of all three sub-studies, particularly those pertaining to negative assessments (Section 6.1) and DRS/reenactments (Section 6.2), illustrate the interplay between a longitudinal change in methods for complaining and a change in speakers' interactional uses of linguistic resources in the L2. Such a change is bidirectional, as it involves both an overall *diversifcation* in interactional uses of certain linguistic resources and a routinization and *streamlining* in the use of other resources. While this dual change has important consequences for the participants' participation in complaint activities, it suggests more general patterns of development in L2 speakers' capacity for action-formation and interaction-organization.

The analysis of negative assessments revealed a diversifcation in the use of particular linguistic resources as speakers move from elementary to upper-intermediate/advanced profciency levels of French. This diversifcation includes the use of a larger variety of lexical items (assessment adjectives and adverbs), syntactic formats (left- and right-dislocations, pseudo-clefts), and idiomatic expressions deployed for expressing a negative stance. In the context of complaining, the longitudinal development allows more advanced speakers to better adjust their assessments to the interactional context. They can vary their high-grade frst assessments and fne-tune their second assessments to align with and upgrade frst assessments in ways that enhance the chances of obtaining afliative responses and/or contribute to joint complaining. The fndings concur with prior studies showing that the development of L2 IC in part involves a diversifcation in the use of linguistic resources for accomplishing precise social actions (see Hellermann, 2008; Nguyen, 2019; and Sert, 2019, among others). While the progressive emergence and diversifcation of linguistic resources have been key concerns of much SLA research (see, e.g., Doughty & Long, 2003, and VanPatten & Williams, 2015), a crucial distinction between such research and the fndings about speakers' developing L2 grammar-for-interaction (Pekarek Doehler, 2018) must be made: From the praxeological perspective of CA-SLA, it is not the diversifcation of linguistic resources *per se* that is at stake, but speakers' capacity to put to use such resources in interaction to accomplish recognizable and context-sensitive social actions.

In addition, the analysis of speakers' use of DRS and reenactments in complaints shows that L2 learning also involves a routinization, and likely 'streamlining', in the socially situated use of interactional resources: With time and repeated use, some resources become more readily available for use in particular action contexts (routinization), to the point that they become the go-to solution at the expense of other resources, which decrease in use (streamlining). The fact that all upper-intermediate/advanced speakers use DRS and reenactments regularly while these resources are rare among elementary speakers suggests that DRS and reenactments become, with time, more routinely available resources for complaining. Moreover, the observed diference in the initiations of DRS/ reenactments across profciency levels points to a progressive routinization and streamlining of enquoting devices used to introduce reported

# 222 *Discussion of results and perspectives*

episodes. The use of linguistically less diverse (and more target-like) quotatives by upper-intermediate/advanced speakers co-occurs with more recognizable initiations that were less disruptive for the progressivity of talk (e.g., fewer broken turn starts, fewer problems with the establishment of person references). This observation suggests that L2 speakers, over time, learn how to mobilize particular linguistic resources for producing recognizable and locally efcacious initiations of DRS/reenactments (rather than merely diversifying their repertoire of enquoting devices). It adds to the limited research showing progressive routinization and streamlining in the use of linguistic constructions for precise actional and interactionorganizational purposes (Kim, 2019; Pekarek Doehler & Balaman, 2021; Pekarek Doehler & Skogmyr Marian, 2022). In some cases, speakers thus select precise interactional resources that prove efcient for particular purposes and then stick to these, just like L1 speakers may routinize linguistic formats for precise purposes, such as for opening or closing conversations (Hofvendahl, 2006).

### *8.1.5 Language and the body for action: change in multimodal practices*

 When adopting a social and praxeological perspective on L2 learning – which focuses on L2 speakers' ability to accomplish social actions rather than on their linguistic knowledge – the inherently multisemiotic nature of social interaction cannot be ignored. What members treat as competent conduct can be accomplished through diferent semiotic means; we are hence dealing with a *multimodal interactional competence*. My analyses demonstrated the multisemiotic nature of face-to-face complaining (Chapters 5–7). In addition to verbal resources, participants draw on prosody, non-lexical vocalizations, gestures, facial expressions, and shifts in gaze and posture to display a negative stance, show afective involvement, and contrast their own reasonableness with the complaint-worthy conduct of third parties, similar to what L1 speakers do (see Drew, 1998; Günthner, 1997; and Selting, 2012, among others). Moreover, my analyses shed light on some diferences in the use of embodied resources between elementary and upper-intermediate/advanced speakers. These observations ofer a small contribution to our understanding of the changes in embodied practices involved in the development of L2 IC, and they highlight the benefts of adopting a multimodal perspective in research on these issues.

The sub-study on negative assessments and embodied stance expressions (Section 6.1) revealed that the use of stand-alone non-lexical vocalizations and accompanying embodied conduct to accomplish precise negative assessments decreases over time. That is, it is not the use of embodied displays of stance *per se* that decreases, but the use of such resources on their own, in particular action contexts – such as to show afliation with a coparticipant after the expression of a complaint-worthy problem – that diminishes over time. Moreover, the longitudinal analysis of one participant's practices for introducing DRS (Section 6.2) showed similar but considerably less salient hand gestures being used to disambiguate person references over time. This observation concurs with fndings by Eskildsen and Wagner (2015, 2018), who have documented decreased gesture scope over time as speakers routinize locally efcacious patterns of language use. Functionalist research has shown that speakers gesture more in their L2 than in their L1 (e.g., Gullberg, 2011), suggesting a general negative correlation between gesture use and higher linguistic profciency. While this might very well be the case, my fndings rather suggest that it is the situated use of embodied resources to accomplish precise social actions that may change. More longitudinal research on L2 speakers' *in situ* uses of embodied conduct is needed to better understand the multimodal changes involved in the development of L2 IC (see Eskildsen, 2021a; 2021b; Pekarek Doehler & Skogmyr Marian, 2022; Skogmyr Marian & Pekarek Doehler, 2022, for recent contributions on this issue).

### *8.1.6 Socialization processes in L2 learning: shared experiences as an interactional resource*

As discussed in Chapter 7, EMCA-inspired research on the development of L2 IC has only recently started to address how the development of interactional methods relates to larger socialization processes. Longitudinal research on L1 interactions, in turn, has shown some ways in which people draw on past experiences to shape future actions (see particularly Deppermann, 2018). While merely exploratory in nature, the two longitudinal case studies in Chapter 7 illustrated how the development of certain complaint practices relates to participants' changing social relationships and their cumulative knowledge about each other as established in prior interactional experiences.

In the case of Suresh (Section 7.1), repeated topicalizations of weather conditions and the longitudinal monitoring of stance-taking toward the outdoor temperature resulted in the establishment of an interactional routine for conversation openings in his interactions with the coparticipant Aurelia. This routine allowed Suresh to take increased responsibility for managing the progressivity of talk (Kim, 2017) and adjust his stance-taking in ways that facilitated exchanges of afliation. The establishment of interactional routines and the mutual adaptation of interactional conduct to recipients in such routines have been observed in longitudinal analyses of L2 telephone openings (Brouwer & Wagner, 2004) and in news-telling sequences (Greer, 2019), showing how interactionally competent conduct is progressively established and co-constructed by the participants. Other studies have demonstrated how repeated engagement in similar interactional exchanges may lead to increased participation for L2 speakers (Eskildsen, 2021a; Waring, 2013; Watanabe, 2016). This seemed to be the case for both Suresh and for Malia's coparticipants. Over time, Malia's coparticipants took the initiative in talking into relevance Malia's personal problems and participated actively in the construction of Malia's complaints by ofering accounts and more substantial displays of afliation and sympathy than earlier (Section 7.2). Importantly, these contributions were designed specifcally for their recipients: Suresh changed his stance-taking to refect Aurelia's expressed stance from one interactional encounter to another, and Malia's coparticipants asked questions and ofered accounts that were based specifcally on their knowledge about Malia's situation. These fndings thus reveal some precise ways in which shared interactional histories help (L2) speakers accomplish context-sensitive and recipient-designed talk, a key feature of IC (Pekarek Doehler & Pochon-Berger, 2015). More broadly, the fndings support a holistic view of L2 IC development as a process that is inextricably intertwined with socialization processes.

### *8.1.7 Increased 'success' in the accomplishment of social activities*

An important overall fnding of my research is that L2 speakers at both elementary and upper-intermediate/advanced levels can accomplish 'successful' complaints, that is, complaints that lead to afliative or sympathetic responses from coparticipants. The higher level of success in the complaints of upper-intermediate/advanced speakers is nevertheless manifested in that they more frequently than the complaints of elementary speakers lead to overt exchanges of afliation, sympathy, and joint complaining (Section 5.3). The increased success of complaints over time hinges on both the complainants' ability to design complaints in recognizable and locally ftted ways and on the coparticipants' capacity to produce co-operative responses.

On the one hand, there is a development over time pertaining to action formation and recipient-design, seen, for example, in more advanced speakers' ability to introduce complaints progressively in ways that allow them to preemptively accomplish extensive accounting work that conveys the legitimacy of the upcoming complaint and the speakers' credibility as complainants before launching strong criticism. The higher level of progressivity in the complaint sequences of more advanced speakers, for instance, with less repair in the introduction of DRS, may also contribute to enhanced recognizability of these speakers' complaints. These fndings thus support the idea that the development of L2 IC involves an increased capacity over time to accomplish social actions and activities in recognizable, efective, and ftted ways, such as requests (Al-Gahtani & Roever, 2012), task openings (Hellermann, 2008), or self-selection in classroom interaction (Watanabe, 2016).

On the other hand, the increased success of complaint sequences over time can be attributed to the coparticipants' growing ability to ofer relevant and timely contributions to the sequences. Coparticipants' responses to complaint initiations and subsequent complaint components are crucial for the accomplishment of a successful complaint. As shown in Chapters 5 and 6, more advanced speakers' increased capacity to verbally express afliation in ways that actively contribute to the sequential development – such as through second assessments – leads to a higher level of co-construction of the sequences and to more joint complaints. These fndings develop Boxer's (1993) observation about L2 speakers' diffculties with efectively contributing to complaint sequences. While Boxer did not incorporate any longitudinal perspective, my analysis has shown ways in which L2 speakers eventually increase their capacity for contributing to complaints. The fnding about growing 'success' in complaining over time also highlights the fundamentally co-constructed and socially distributed nature of IC (Hauser, 2019; He & Young, 1998; Greer, 2019; Kasper & Wagner, 2014), showing that the success of a conversational activity is highly dependent on what the participants can do together. It is not enough for one participant (such as the complainant) to develop high profciency in the L2 – the coparticipants have to be able to respond in ways that recognizably align and afliate with the speaker. We can thus speak of a *concurrent* development in participants' interactional practices, which over time results in a cumulatively stronger interactional dynamic with an increased ability to successfully co-construct social activities.

### *8.1.8 Summary of implications for understanding the development of L2 IC*

The abovementioned implications for our understanding of the development of L2 IC can be summarized as follows:


of precise social actions, while in other action contexts, the use of embodied conduct remains stable.


In short, the longitudinal analysis of L2 complaint practices has shed light on numerous facets of the development of L2 IC, including changes in turn-taking and recipient responses, in the management of sequence and preference organization, and in the use of linguistic resources and embodied conduct for precise interactional purposes. It has explored some dimensions of the interrelation between L2 development and socialization processes, and it has illustrated the fundamentally co-constructed nature of IC. The development of L2 IC thus involves both the emergence of certain practices and resources, the qualitative change in some interactional methods, and a quantitative redistribution in the use of particular practices and resources. Ultimately, the fndings about complaining in L2 French help us better understand why the development of L2 interactional skills is such a complex and difcult endeavor and what it actually means to gain increased interactional competence. They also open up avenues for future studies in this feld (see Section 8.3).

#### **8.2 Understanding complaining in interaction**

Although the study focused on L2 complaint practices, the fndings also have implications for our understanding of complaining generally since they shed light on aspects of complaints that have received only limited attention in the L1 literature. These implications concern: (1) the overall composition of complaint activities, (2) complaint initiations, (3) practices for engaging in joint complaining, (4) multimodal packages for negative assessments, and (5) the interpersonal purposes of indirect complaining.

The fndings of sub-study 1 enrich our knowledge about the overall composition of complaints. Prior research on complaining has typically focused on diferent components of complaining without addressing the overall structure of the activity. Traverso's (2009) study on ordinary conversations in L1 French are one exception; Ruusuvuori et al.'s (2019) investigation of performance appraisal interviews in Danish and Finnish is another. My fndings (Section 5.1) concur largely with those of Traverso (2009), although I am hesitant to divide complaint sequences into static 'phases' as she does. The fact that the overall organization of L2 complaints is similar to L1 complaints supports the idea of a similar understanding of indirect complaining across several languages and cultures. Complaining boils down to a series of basic interactional tasks that are driven by the complainant's interactional project (Levinson, 2013) of pursuing afliative or sympathetic responses. Regardless of the interactional setting and the (type of) participants involved, complainants need to justify and provide sufcient 'evidence' for the complaint, and this is recurrently done through similar actions.

How participants move into complaining is another under-researched topic. My fnding (Section 5.2) that speakers sometimes initiate complaints straightforwardly, without orientation to delicacy (cf. Günthner, 1997; Selting, 2012), and sometimes through careful, stepwise escalation of negative stance displays (Ruusuvuori et al., 2019) adds to the idea that competent complainants can adjust their complaint initiations to local contingencies. In some cases, participants orient to straightforward initiations as locally appropriate; in other cases, progressive work-ups with elaborate pre-complaint work are deemed necessary. In addition, the fndings about frequent self-praise and similar subject-side practices that serve to build the credibility of the complainant already in the initiation of the complaint confrm and extend what has been discussed mostly parenthetically elsewhere (Edwards, 2005; Ruusuvuori et al., 2019; see also Skogmyr Marian, 2021b).

Moreover, the analyses of joint complaining (Section 5.3) shed light on some of the specifc practices people use to accomplish the precise *jointness* of such complaints. Prior research has indicated that joint complaining involves an escalation of negative stance displays and afectivity (Rääbis et al., 2019; Drew & Walker, 2009). My fndings demonstrate the importance of participants' ability to produce afliative, and specifcally upgrading, second assessments for such escalation to take place. The kind of non-verbal, embodied assessments frequently used by elementary speakers (non-lexical vocalizations, embodied conduct) seem to be less efective for engaging in joint complaining. This is likely because the 'weaker' status of non-verbal assessments compared to verbal ones (Couper-Kuhlen, 2012) makes them more difcult to further build upon in subsequent turns. A key aspect of competent complaint conduct thus involves the ability to upgrade negative frst assessments.

 Although much of the research on L1 complaining is based on telephone calls, several studies have highlighted the multimodal nature of face-to-face complaining (see Section 3.3). My observation that complainants recurrently produce high-grade negative assessments as multimodal packages (Section 6.1) has nevertheless not been documented elsewhere in the CA literature. The analyses show that speakers routinely assemble bundles of verbal and embodied resources, such as negatively loaded assessment terms, marked prosody, frowns, and other facial expressions conventionally associated with the display of frustration or indignation, as well as conduct used for reinforcement generally (e.g., hand gestures), to accomplish high-grade expressions of negative afective stance. This fnding contributes to the growing literature that investigates speakers' use of multimodal packages for precise action purposes (Goodwin, 2007; Hayashi, 2005; Kärkkäinen & Thompson, 2018; and several contributions in Pekarek Doehler et al., 2021) by documenting such packages in a new action context.

Finally, the longitudinal case studies presented in Chapter 7 provide empirical evidence for how participants' micro-level calibration of morality in interaction (Stivers et al., 2011) interrelates with their management of social relationships. The study with Suresh (Section 7.1) demonstrates that complaints about trivial matters, such as the weather, can serve important small-talk purposes in conversation openings. At the same time, as shown in Suresh's contradictory stance expression about the cold from one interaction to another, such small talk is an important avenue for doing relational work – as has been argued about complaining generally (Boxer, 1993; Günthner, 1997; Hanna, 1981). The analysis of Malia's complaints (Section 7.2) underlined the social-relational dimensions of complaining to an even greater extent. As seen in the topical development over time and in the participants' exchanges of afliation and sympathy, complaint practices are not just afected by the nature of the participants' relationship. Complaining also provides a platform for participants to show their willingness to engage in co-member conduct (Hanna, 1981) – as, for example, foreign students with similar problems and difculties who are on the same side concerning particular issues. This explains the many complaints about the French language, university courses, and Swiss society in my data, which are topics that the participants have in common and lend themselves to co-membership talk. Participants use indirect complaining to display belongingness and commiserate – in other words, to work on and *transform* their social relationships. While CA studies typically refrain from discussing the potential social-relational 'benefts' of complaining – likely to avoid the risk of ofering etic judgments – my fndings provide emic evidence for participants' own orientations to the interpersonal purposes of complaint activities. They also challenge the popular view that complaining is something harmful that should be avoided at all costs (see, e.g., the discussion in Winch, 2011).

#### **8.3 Perspectives**

This book has delimited its scope of inquiry to particular dimensions of L2 complaining, to a certain type of participants and interactional setting, and to precise points in the developmental trajectory of L2 complaint practices. Research topics and methodological optimizations for future studies on L2 complaining emerge logically from these delimitations, and I will therefore not discuss such specifc research inquiries in detail. In this fnal section, I instead consider a few larger theoretical and applied

implications from the empirical fndings that deserve more scientifc attention in the future.

The present work has portrayed human social interaction as inherently multimodal, the mastery of which involves a *multimodal interactional competence*. Throughout the analyses, we have seen how speakers treat linguistic structure, prosody, non-lexical vocalizations, facial expressions, hand gestures, gaze, and body posture as relevant resources for establishing joint understanding and for organizing discourse in face-to-face interaction. Assemblages of such resources – such as a sigh and a lowering of one's head into the hand – work as recognizable and sometimes routinized multimodal packages for action. The indications about a change over time in L2 speakers' embodied conduct highlight the crucial need for more research on the multimodal dimensions of the development of L2 IC (see Markee, 2019). Although many existing CA studies take embodied conduct into account in their analyses, few studies have systematically addressed longitudinal changes in L2 speakers' situated embodied practices (for notable exceptions, see Section 8.1.5). We still know very little about what happens with L2 speakers' use of their bodies as they increase their linguistic resources and their cumulative experiences with L2 interactions. More attention to multimodal practices for both action formation and for interaction-organizational purposes is therefore needed in order to develop a holistic understanding of L2 development and learning.

The study has ofered theoretical insights that can be used for applied purposes in the feld of language education. So far, few attempts have been made to use empirical research on the development of L2 IC to develop pedagogical policy and practice. Salaberry and Kunitz's (2019b) edited volume, which addresses both the teaching and testing of IC in a variety of languages, constitutes an important efort to bridge theory and practice in this feld (see also Barraja-Rohan, 2011; Huth, 2020; Kunitz et al., 2021; Piirainen-Marsch & Lilja, 2022; and Wong & Waring, 2010, for the use of CA fndings generally to teach L2 interaction and Sandlund et al., 2016; Roever & Kasper, 2018; Youn, 2015; and the contributions in Salaberry & Burch, 2021, on CA research for language testing purposes). Some of my fndings lend themselves particularly well to pedagogical applications. For instance, the observations about jointly constructed stance-taking activities highlight the need for pedagogical attention to interactional uses of frst and second assessments in the L2 (as opposed to, e.g., teaching of how to express one's liking/disliking *in abstracto*). This is not the place to discuss detailed pedagogical implications or to ofer precise teaching or testing recommendations, however. More substantial eforts, both fundamental and applied, are needed for the development of empirically grounded and ecologically valid L2 pedagogy. Only a solid theoretical base on the micro-level workings of interactional competence and its development over time, as well as concerted eforts across a chain of experts (Pekarek Doehler, 2019), will help bridge the existing gap between theory and practice. The cumulative results of such work will ofer grounds for a more coherent, sociologically grounded epistemology of L2 learning and teaching (Wagner, 2019) and may help develop the authenticity of high-stakes language policy documents such as the CEFR (Council of Europe, 2020).

My exploratory analyses of the intersection between the development of L2 practices for complaining and the longitudinal accumulation of shared knowledge and evolving relationships have illustrated how interactional competence is shaped by and refects our concrete everyday experiences with others. I have only indirectly addressed language learning as a social activity. But the longitudinal analyses in Chapter 7, and particularly the case study with Suresh, have shown ways in which speakers recurrently orient to and rely on specifc prior experiences to shape future actions (see also Eskildsen & Wagner, 2013). In this regard, the study afords a window into how people's experiences conspire to shape their L2 development (Ellis, 2019) and sheds light on some ways in which language learning and use are mutually constitutive (Firth & Wagner, 2007; Pekarek Doehler, 2010). Interactional competence emerges from repeated participation in real-life social encounters. People remember prior talk and actions, and they redo these with some adjustment to present contingencies. Additional CA research on how shared interactional histories and evolving social relationships afect interactional practices of both L1 and L2 speakers will contribute to a more holistic understanding of interactional competence and its development over time. Similarly, the combination of CA and usage-based linguistics (Cadierno & Eskildsen, 2015; Eskildsen, 2020; Eskildsen & Pekarek Doehler, 2022) deserves more attention in the future, as this might help advance our understanding of the mutually constitutive nature of L2 use and learning.

As a fnal point, this book has illustrated how the general, lifelong learning and adaptation processes that are ubiquitous features of human social life (Goodwin, 2018; Tomasello, 2019) manifest themselves in precise features of social interaction. L2 learning constitutes only one of the many adaptation processes in which we engage in our daily lives, and 'learning how to ft in' (Atkinson, 2019) is not a skill that is specifc to L2 speakers. My study showed people getting together to talk and improve their spoken French. In doing so, they ended up complaining – not because they are L2 French speakers, but because they are humans and that is what humans do. Malia's workplace complaints (see Section 7.2) highlighted the intertwined nature of linguistic and social-cultural adaptation processes: Malia's workplace integration and her identity construction as a competent professional was inextricably tied to the development of interactional skills in the L2. In a sociological study of stress management among American graduate students around the middle of the last century, Mechanic (1962: 221) concluded that "perhaps the most important question we can ask about human behavior is how man

continues to persist and maintain 'health' and 'balance' in the complex circumstances of modern life". It would be safe to say, I believe, that modern life has continued to complexify since then. Given today's geographical and social mobility, one challenge for many people today is surely how to adapt to new linguistic and cultural environments. Based on the present study, I would suggest that fnding people with similar life situations and discussing problems – and even complaining about them – is one of the ways in which we do such adaptation work. Future research will hopefully shed further light on how language learning intersects with other types of human adaptation processes.




*English language learning, teaching and testing in global contexts* (pp. 132– 158). Multilingual Matters.



Keevallik, L., & Ogden, R. (2020). Sounds on the margins of language at the heart of interaction. *Research on Language and Social Interaction*, *53*(1), 1–18.



P. Seedhouse, S. Walsh, & C. Jenks (Eds.), *Conceptualising learning in applied linguistics* (pp. 105–126). Palgrave Macmillan.




# **Appendix**  Transcription conventions

The following conventions are based largely on Jeferson (2004), with the addition of a selection of Mondada's (2019) conventions for annotating embodied conduct.


# 248 *Transcription conventions*


To facilitate reading, embodied conduct is sometimes described in double brackets, for example: ((all participants nod))

Talk that is particularly important for the analysis appears in **bold**.

# **Index**


167, 189, 216, 221, 227, 228; linguistic resources used in 17; negative 31–33, 35, 36, 37, 59, 61, 84, 89–90, 91, 94, 95, 105–129, 151–152, 159, 167, 170, 179, 187, 208, 216, 221, 222, 227–228; and non-lexical vocalization 36; passepartout 17, 109, 129; positive 17, 31, 32, 64, 65, 74, 91, 205; second 32–33, 35, 38, 39, 103, 111–114, 123–125, 221, 227; summary 7, 31, 33, 58, 117, 124, 125, 179, 180, 189; upgrading of 32, 85, 90, 111, 112, 113, 120, 123, 124, 128, 151–152, 170, 221, 227

Atkinson, D. 43

Berger, E. 13, 14, 17, 38, 64, 101, 210, 213, 219 Boxer, D. 37–38, 39, 103, 225 broken starts 130, 133, 138 Brouwer, C. E. 19, 172, 211, 212 CA-SLA 3, 4, 11, 43–44 Chomsky, N. 11 circumstantial detailing of complaint 28, 34, 59, 61 circumstantial prefaces 62, 64–65, 69, 70 Clayman, S. E. 46, 216 Clift, R. 34 closing-implicative assessments/turns 17, 30, 58, 152 closings, complaint 29–30, 58, 60–61 co-construction of complaints 79–80, 103–104, 196, 211, 213, 218–219, 225; elementary speakers 80–91; and negative assessments 31; and recalibration of interactional competence 101; upper-intermediate/

advanced speakers 80, 91–100, 218; *see also* interactional histories; joint complaints collaborative turn completions 15–16, 104, 219 co-membership, and complaining 210, 228 Common European Framework of Reference (CEFR) 2 communicative competence 11 comparative adjectives 120–121 complaints/complaining 2, 4, 23, 48, 226–228; closings 29–30, 58, 60–61; co-construction of 79–100, 103–104, 196, 211, 213, 218–219, 225; and collaboration between coparticipants 14; and co-membership 210, 228; complaint worthiness of situation 24, 26, 30, 33, 34, 58, 79, 94, 95, 100, 102, 111, 121, 124, 125, 128, 130, 132, 133, 149, 172, 179, 185, 187, 211, 218; complex nature of 3, 7; composition of 226–227; core features of 23–25; cumulative evidence about 39–40; development 27–28; direct 23, 39; indirect 23–24, 25, 39–40, 217–218; initiations 25–27, 37–38, 59, 62–79, 101, 102, 189, 210–211, 219–220, 227; interactional resources for 31–37; interpersonal purposes of 228; joint 29, 32, 40, 59, 79, 80, 85–100, 103, 112–114, 123, 128, 152, 160, 171, 172, 218, 219, 221, 225, 227; in L2 interaction 37–39; naming of complaint 60; positive dimensions of 2–3; prosodic and other nonlinguistic resources 35–37; recipient responses to 28–29; research gaps 40–41; second 30, 60, 62, 67–68, 70, 101, 102, 211, 218, 219; similarities with other conversational phenomena 24–25; and storytelling 13, 24, 29; structural organization of 25–30, 54–104; subjective side of 27, 31, 227; successful 224–225 confrmation 84, 114, 124, 138, 167, 171, 179, 202 context sensitivity 21, 36, 103, 123, 152, 172, 213, 220, 224 contrastive formulations 31, 32, 62, 64–65, 66–67, 69, 74–75, 78 conversation analysis (CA) 3–4, 7, 10, 19, 38, 42, 154, 217; CA-SLA 3, 4, 11, 43–44, 221;

ethnomethodologically inspired 11, 19, 42–43, 44, 48, 216; longitudinal 45–46


Edwards, D. 26, 27, 36


emic perspective 4, 10, 43, 44, 173, 212, 216, 217, 228

	- openings 13–14, 26–27, 62, 64, 220

quantitative analysis 46


supplementary material 47; setting and participants 46–47; transcription and anonymization 47–48


102; interactional building blocks 54–62, 100, 217–218; length of complaint sequences 54; sequence candidate 30, 54–55, 59–61, 100 summary assessment 7, 31, 33, 58, 117, 124, 125, 179, 180, 189 superlative adjectives 120–121, 123 sympathy 7, 23, 27, 103, 211, 220, 224–225; and complaint closing 30, 60; elementary speaker complaints 117, 138; and second assessments 32, 111; upper-intermediate/ advanced speaker complaints 58, 91, 124, 175, 179, 186, 187, 192, 195, 197, 200, 203 synchronization 21, 95, 99, 101, 103, 104, 153, 219 telephone opening sequences 19–20


upper-intermediate/advanced speakers: co-construction of complaints 80, 91–100, 103, 218; complaint initiation 62, 70–79, 101, 102, 219–220; DRS and reenactments of 130, 141–151, 221; first assessments 118–123; interactional histories 174–211, 212–213; negative assessments 91, 94, 95, 106, 118–129, 179, 185; second assessments 123–125; sequence organization 219–220; use of negative assessment adjectives and intensifiers 127–128; verbally incomplete negative assessments 125–127; *see also* interactional histories

usage-based linguistics 152, 154, 172, 212, 230

verbally incomplete negative assessments 125–127

