A Conversation Analysis Approach to French L2 Learning

Introducing and Closing Topics in Everyday Interactions

Clelia König

• Publisher: Routledge
• Year: 2019
• eBook ISBN: 9780429631801
• Edition: 1
• Topic: Langues et linguistique
• Subtopic: Linguistique

This book offers a critical examination of second language (L2) learning outside institutional contexts, with a focus on the way second language learners introduce, close, and manage conversational topics in everyday settings. König adopts a Conversation Analysis for Second Language Acquisition (CA-SLA) approach in analyzing oral data from a longitudinal study of L2 learners of French, au pairs in Swiss families, over several years. With this approach the author presents insights into the ways in which L2 learners introduce and close conversational topics in ongoing conversations and how these strategies evolve over time, setting the stage for future research on this little documented process in second language acquisition. This volume contributes toward a greater understanding of L2 learning "in the wild, " making this key reading for students and researchers in second language acquisition, applied linguistics, and French language learning and teaching.

1
Data, Methodology, and Research Questions

1.1 Conversation Analysis: Methodological and Analytical Issues

The analytical approach of CA is characterized by a detailed focus on specific interactional phenomena. CA investigations are based on transcriptions of naturally occurring interactions as they happen in different everyday contexts. This means that the investigations data are highly dependent on the details of the interactional context in which talk is embedded. CA analyzes interactions as they take place in the interactants’ lives, in a panoply of different contexts. The researchers record naturally occurring conversations as they happen every day: in the family, at work, at school, on the street, in a car, with a doctor, on the phone, and so on. Finally, CA focuses on the structural organization of conversation and on the methods that the interactants use when they talk to each other. People, then, are seen not merely as exchanging messages through talk-in-interaction but also as collaboratively accomplishing social actions.

The basic starting point for CA researchers is the question of how interactants do what they do at that specific interactional moment. This question is central to the methods (Garfinkel, 1967)—i.e. the systematic procedures—that interactants employ and develop during social actions with their interlocutors. This attention toward the speakers’ methods correlates with CA interest in the language-in-use: the language as it is used in specific occasions by specific interactants (cf. ten Have, 2007).

1.1.1 Conversation Analysis’s Methodological Way: The Ethnomethods in the Data

CA is based on naturally occurring data, for whose obtainment ethical and formal constraints have to be respected. For example, the participants have to agree to be recorded and to have their data used for scientific purposes. Moreover, the researchers are obligated to anonymize the participants’ names and every other piece of information related to the time and place of recordings. The data can be audio-recorded or video-recorded. This depends on the type of data. For example, phone calls are only audio-recorded; it also depends on the willingness of the participants to be video-recorded as well as on the aim of the study. In the next chapter, I present my data: I have audio recordings of family conversations between au pairs and their host family members in the French-speaking part of Switzerland. In this case, the families did not allow the research team to video-record their children.

Once the data is recorded, it is transcribed in a second step. The transcription is already a first data analysis, because it is up to the researcher to decide what to write when something is unclear or when there is a pause. Depending on how an interaction is rendered in the written form and on how many details are included in the transcript, the transcripts vary in their definiteness and precision (for an introduction on how to transcribe talk-in-interaction, see Jenkins, 2011).

If it seems obvious to include as many details as possible, there is a limit to consider: the transcript must be readable and usable. This means that, depending on the phenomena that are being investigated, only the relevant transcription conventions are used. For the present work, the basic reference for preparing the transcript was Gail Jefferson’s 1984 paper, in which she presented and discussed a list of transcription conventions (a part of which is included in Appendix I). However, since the focus of the investigation was not exclusively on the prosody of the speech, I have transcribed only in a limited way the prosodic contours of the participants’ turns. Instead, I was intrigued by pauses, signs related to turn-taking, and the general organization of the conversation, so I have put these features in the foreground while transcribing.

1.1.2 Participants’ Methods and Interactional Phenomena

Using detailed transcriptions as just outlined, CA investigates language as it is employed in people’s conversations in different settings. Language, then, is not regarded as a fixed entity that a speaker uses in a right way or a wrong way. On the contrary, it is flexible, adaptable to the practical purposes that the interactants are accomplishing in their conversations. For these reasons, CA analyzes the methods (cf. Garfinkel, 1976: 11) through which interactants accomplish social actions in conversations. Through this approach, the ethnomethodological roots of CA become observable: the “ethnomethods” are the central analytical object of CA research. They are “ethno” because they are endogenous for the interactants themselves, i.e. these methods are known in common in the community that the speakers belong to. A community is not only a regional, political, religious, or linguistic grouping of individuals. Rather it is constituted every time people come together and share a conversation. A community is therefore defined always ad hoc by the participants themselves during the interaction. The interactants do not have to declare their rules for their talk-in-interaction every time they gather. They make the rules each time.

CA investigations are directed either towards members’ methods for accomplishing specific actions during interactions—such as repair, (dis) agreement, conversation opening and closing, topic management, (dis)alignment and (dis)affiliation, task opening and closing—toward specific resources (oh, mhm, and), or toward structural principles governing conversational sequences (organization of preference, turn-taking, construction and expansion of adjacency pairs, sequence organization, etc.). The next example illustrates the analytical focus on actions, resources, and structures.

(1) Excerpt 1.2 from ten Have (2007: 4; adapted from Frankel (1984: 153))

Example 1 stems from a conversation between a doctor and his patient. The patient asks the doctor a question in line 01 and does it by proposing a positive assessment regarding the effects of the chemotherapy she has to undergo. The doctor does not answer this question for a long time (2.2 seconds of pause) to which the patient orients as something that does not align with her positive assessment (Pomerantz, 1984). The pause in line 03, however, gives the patient the possibility of self-selection for another turn. She takes it and reformulates her previous assessment in line 04 in the exact opposite direction, i.e. she “reverses” (ten Have, 2007: 5) her assessment (It will?). Now the doctor confirms this in line 05: I’m afraid so.

This short example contains several points of interest for CA. First, it shows the action of agreeing/disagreeing. With her first question, the patient projects a response that aligns with the positivity of her stance. The long pause that follows, however, suggests some trouble from the doctor’s side to actually align with his patient’s stance. With her second question, then, the patient reverses the polarity of her stance, i.e. from a positive (and hopeful) to a negative one. At this point, the doctor can align to her turn. In this case, we see the organization of preference at work: the doctor could have answered the patient’s question with a disaligning and disagreeing turn. However, he prefers not to answer in this way and offers the patient the possibility to reformulate her question so that she can answer with an aligning turn. However, no video data are available for this interaction so it could be that some other activities are being carried on by the participants to which we as analysts do not have access.

Second, as far as resources are concerned, ex. 1 illustrates the use of affiliating responses. This example is taken from a doctor–patient consultation, which contains exchanges about private and intimate problems that may arise from a session of chemotherapy. As demonstrated by Selting (1994), Stivers (2008), Stivers et al. (2012), and Steensig (2013), in specific cases, especially at the end of storytelling, the interlocutors are offered a final turn with which they can not only align but also affiliate. An affiliation is observable in the sense that the interactant shows a sort of “empathy” toward the other speaker. In ex. 1, the affiliation of the doctor is seen with his last turn: the doctor does not respond to the patient’s first question’ instead, he lets her reformulate it so that her next answer is not only aligning but also affiliating. This is observable if one thinks that a merely aligning answer could have been something like “Yes, it will.” On the contrary, the doctor proffers a turn that displays empathy toward the difficult and personal topic they are discussing.

Finally, the investigation of conversational structure highlights the organization of preference. Preference is a structural feature of conversation; i.e. when a first pair part is uttered (like the first question in lines 01–02), specific type of second pair part is projected and made conditionally relevant. However, there is not just one possible second pair part that can be delivered (in this case, the second pair part could be the answer to the question or a silence). The second pair part shows the type of alignment its speaker takes toward the previous turn. There are essentially two types of alignment: a positive one, i.e. the second speaker structurally aligns to the first pair part of the first speaker (like in question-answer or greeting-greeting), or a negative one, i.e. the second speaker structurally disaligns from the first pair part of the first speaker (e.g. invitation-decline or agreement-disagreement). It is important to state that

this is not necessarily an alignment with, or distancing from, the speaker of the first pair part (although it may be that as well) but the project of the first pair part, and the course of action it is designed to implement.

(Schegloff, 2007: 59–60)

1.1.3 The Investigation of Linguistic Phenomena in Everyday Talk-in-Interaction

So far, three underlying features of the study of talk-in-interaction have been addressed: actions, resources, and structure. However, several studies (see among others: Auer, 1996; Couper-Kuhlen, 2004; Ford & Thompson, 1996; Ford et al., 1996, 2001; Lerner, 1991; Selting, 1996; Sorjonen, 1996) have concentrated on specific linguistic phenomena in interaction and have the following aim:

[describe] linguistic phenomena as resources for the construction and organization of practices and actions/activities in interaction.

(Kern & Selting, 2013: 1)

The attention brought to linguistic resources derives from their being recurrently found in conversation and being strategically used by the speakers in their social actions. This means that interactants can rely on them as resources for achieving a variety of practical purposes during the interaction. Selting and Couper-Kuhlen explain in their introduction to the volume “Studies in interactional linguistics” (2001) that it is worth looking at the linguistic, prosodic, and grammatical resources that the speakers employ in conversation. They support this point of view addressing one of the core machineries of interaction: turn-taking. The authors argue that interactants manage turn-taking on the basis of TCUs (turn constructional units), which are linguistic units, since they can be words, phrases, clauses, or sentences (p. 5). Moreover, the point of completion for TCUs is established in the same way as other linguistic features, namely through their grammatical, prosodic, and pragmatic cues (p. 6). In so doing, they consider that linguistic observations have their rightful place in the research domain of CA and can moreover be helpful in achieving a deeper understanding of the most basic principles and machineries of conversation.

Within this linguistic approach to interactional data, known as interactional linguistics, a specific focus has been on grammar in interaction. In what is arguably the centerpiece in the field of interactional linguistics, Ochs et al. (1996: 3) formulate how

three genres of inquiry converge here—one grounded in functional approaches to language concerned with its role in communication and cognition, one grounded in linguistic anthropology and the cultural underpinnings of language, and one grounded in conversation analysis and the interactional matrix of language structure and use.

Grammar is no longer seen as a fixed group of rules that can be used in a right way or a wrong way. Instead, grammar in interaction is an open class of rules that the speakers adopt on the spot, i.e. while they are speaking, to adjust their turns and their actional trajectories. The studies in this specific research domain have, for example, challenged the tra...

Table of contents

• Cover
• Half Title
• Series Page
• Title
• Copyright
• Dedication
• Contents
• Acknowledgments
• Introduction
• 1 Data, Methodology, and Research Questions
• 2 Topic Analysis in First and Second Languages
• 3 L2 Acquisition and Interactional Competence
• 4 Topic Management in French L1 and L2 Interactions
• 5 Topic Introduction
• 6 Topic Closure
• 7 Nature and Development of L2 Interactional Competence
• 8 Conclusion(s)
• Appendix I: Transcription Conventions
• Index