Metadiscourse in multi-party conflict talk Hermine Penz Metadiscourse has been recognized as a pervasive feature of discourse which allows us to reflect and comment on discourse, explicitise intended meanings, yet also redefine and reformulate or structure it. Many studies have pointed out that it can be used to guide our interpretations and that it plays an important role in achieving understanding. Very few studies, however, have focused on the role of metadiscourse in conflict talk. This study aims to contribute to filling this gap by investigating the uses and functions of metadiscourse in small group discussion in the context of higher education. The results show that metadiscourse plays an important role in all phases of conflict, i.e. its initiation, development and resolution. In the speech activity of small group discussions communicative activities of all types (verbal, non-verbal, written) are labelled and thus explicitised. In addition, metadiscourse plays an important role in commenting on the discourse structure. 1. Introduction Our everyday interactions are characterised by both conflict and cooperation as they constitute an inherent feature of social life. Both are achieved interactively and are dependent on the use of metadiscourse as this allows people to guide others with respect to how their talk is to be interpreted and how it is structured. While conflict has been researched intensively in other disciplines, the study of conflict in discourse has not attracted the attention of scholars as much, which however, might partly be due to the difficulty of obtaining data. The concept of metadiscourse has recently attracted more interest from researchers, in particular as a feature of discourse. This article, however, sets out to study metadiscourse in interactive conflict talk among a small group of students who are engaged in solving a task for a course in linguistics, which is a type of academic discourse. After providing a brief overview of the concept of metadiscourse, the paper discusses conflict talk before moving on to the AAA - Arbeiten aus Anglistik und Amerikanistik Band 42 (2017) · Heft 2 Gunter Narr Verlag Tübingen Hermine Penz 302 analysis of the data. The data consist of four conflict episodes which occurred within a thirty minute segment of talk. In this segment the conflict builds up and becomes more serious in the course of the interaction, yet is resolved at the end. The study investigates the role of metadiscourse in the initiation, development and termination of the conflict in the small group interaction. 2. Metadiscourse Metadiscourse is a unique feature of human language as it allows us to talk about language itself. This ability has been termed reflexivity by Hockett (1960), who views it as one of the design features which distinguish human communication from animal communication. Caffi (1984: 449) states that “reflexivity can [also] be found in the speakers’ ability to communicate on the communication they are involved in, and precisely in the fact that they not only can define it, but also confirm, refute, or modify their own definitions or the definitions given by their partners.” A similar view is expressed by Jones (2016: 185), who argues that “[r]eflexivity is the quality of discourse that allows us to turn back at […] what we have said and done and talk about it, to reflect upon it, to reformulate it, and to reframe it through […] metadiscourse.” This phenomenon has been discussed since antiquity and linguists have used various constructions with the prefix meta, such as metalanguage (Jakobson 1990/ 1976), metadiscourse (Vande Kopple 1985, Crismore 1989, Hyland 2005), meta-talk (Schiffrin 1980), meta-text (Mauranan 1993), and metacommunication in this connection. A distinction has frequently been made between metalanguage and object language as the former denotes a language which is about language and is at a different level than object language (cf. Mey 2001: 173). However, this concept is blurry as it is difficult to distinguish metalinguistic features from other functions of language. Broadly speaking, two traditions can be identified in the study of metadiscourse. The broad approach, mainly represented by researchers who study written discourse, includes comments about the ongoing text/ discourse and any instances where the presence of the writer or the reader is made explicit. The writer’s presence is displayed by metalinguistic elements that signal his/ her attitude towards the text or by indications of how the text is organised. The reader’s perspective relates to metadiscourse which guides the addressee in structuring interpreting and evaluating the text. In essence, the broad approach is discourse about the textual and interpersonal functions of language in Halliday’s (1994) terms. Most scholars who study written metadiscourse take the broader view. The narrow approach mainly restricts itself to the textual function (e.g. Mauranen 1993, Schiffrin 1980). Since the focus of this article lies Metadiscourse in multi-party conflict talk 303 on spoken interaction I will primarily discuss research on spoken metadiscourse in this brief overview of the research literature. Schwitalla’s (1979: 141) research represents one of the earliest studies of metadiscourse in general and of spoken language, in particular. He uses the term metacommunication and describes this as a verbalization of events which are communicatively relevant in the ongoing discourse. According to him this applies to the following areas: (a.) defining, describing and making illocutions and perlocutions explicit; (b.) defining and evaluating linguistic interactions, their rules and (temporal) conditions; and (c.) defining topics and requesting that particular topics should be dealt with (or should be excluded). Interestingly, he also includes nonverbal behaviour and activities (e.g. why are you looking at me like that? ) in his definition of metacommunication. Schiffrin’s (1980) classification of metadiscourse is based on studying everyday conversation with regard to where metadiscourse occurs and why it occurs at particular points in the interaction. She concludes that it is used for both referential and expressive purposes. Schiffrin chooses the term meta-talk and establishes three categories: (a.) metalinguistic referents (linguistic entities referred to such as words, phrases, sentences; discourse deixis, such as the former, the latter, the next point; demonstrative pronouns such as this and that when they point to items in the text (whereby they also become discourse deixis), e.g. let me say this; (b.) operators (true, false, right, wrong, verbs like mean and discourse connectives such as like or for example); (c.) verbs (verbs of saying, such as say, tell, ask, assert; verbs indicating that something will be done by a piece of talk such as clarify, define; metalinguistic verbs which refer to a speech event, e.g. argue, joke). The most researched context for studying metadiscourse is academic discourse, both in school and university settings. Lee and Subtirelu (2015) did a comparative analysis of metadiscourse in EAP classroom interaction and university lectures. Studies of spoken academic language in a variety of activity types were carried out by Mauranen (2007), (2010). Academic conference talks were analysed by Luuka (1994). The volume on the use of metapragmatics edited by Bublitz and Hübner (2007) includes studies of a wide range of different contexts such as everyday language, educational discourse and professional discourse, among these studies are those by Ciliberti and Anderson (2007), who compare the functions of different types of institutional discourse. Penz (2007 and 2011) investigated intercultural project discussions in education and Graf (2015) researched metadiscourse in coaching interactions. Bublitz (2001: 1332) states that metadiscourse may refer to any level and any aspect of discourse. These include the channel (e.g. acoustic problems in understanding), the roles of speaker and listener, speaking rights and turn-taking, word meaning, sentence and utterance meaning, topic of talk, frames and text types, conversational maxims, textual ele- Hermine Penz 304 ments, discourse organisation, etc. In addition to addressing the ongoing discourse, it can be applied prospectively and retrospectively in organizing and structuring the discourse and negotiating meaning. The functions of metadiscourse can generally be seen in describing, organising and commenting on the ongoing discourse, often serving the purpose of achieving understanding and supporting acceptance. These functions have, for example, been identified in cooperative intercultural group discussions in which participants lacked common linguistic and cultural backgrounds (cf. Penz 2007, 2011). Watzlawick et al (1967: 52) have emphasized the relationship aspect of metacommunication as it “refers to what sort of message it is to be taken as, and, therefore, ultimately to the relationship between the communicants.” However, they take a broader view of metadiscourse than most scholars. Most researchers stress the positive role that metadiscourse plays in communication as a means to achieve understanding. However, speakers frequently employ metadiscourse about the illocutionary or perlocutionary force of other speakers’ talk for the purpose of expressing aggression or tension in interactions, while metadiscursive labelling of their own talk tends to enhance (the relationship aspect in) communication by reducing tensions/ conflict (cf. Schwitalla 1979: 2, see also Watzlawick et al. 1967). Jones (2016: 185) has pointed out that metadiscourse goes beyond commenting on what they themselves or what other people have been talking about, yet applies to other discursive processes which essentially organize, comment on, take a stance on, and communicate intentions, among them, framing, contextualizing, identity construction and positioning, signaling relationships (by means of politeness strategies), etc. What is more, he argues that metadiscourse also connects our everyday interactions (‘small c’ conversation) with larger social discourses, i.e. the discourse that has been going on in a particular social group/ society about a particular topic or debate (‘Big C’ Conversations). 1 Metadiscourse is also a potential means of holding people accountable for their utterances and actions, yet in situations where power is distributed asymmetrically, it can also be used for the opposite, i.e. “to make us less answerable, to silence debate, harden stereotypes, increase inequality”, and so forth (Jones 2016: 186). This is illustrated by examples of police abuse of African-American citizens in the US, where the more powerful police (and their legal representatives) have frequently been able to re-interpret their excessive use of force. Even in cases where police violence was videoand/ or audio-recorded, the former was justified as officers’ attempts to protect themselves by decontextualizing and fragmenting the events and recontextualizing them within the framework 1 See Gee (2011) for the distinction between ‘small c’ conversations and ‘Big C’ Conversations. Metadiscourse in multi-party conflict talk 305 of the language of police work, terming the different phases escalation, deescalation and assessment periods, which is a type of metadiscursive reinterpretation of the language and actions of the events within the cultural storyline of American blacks as violence and fear associated with this group (cf. Jones 2016: 193-194). In her study of parliamentary debates in the UK, Ilie (2003) demonstrates that politicians use metadiscourse on various levels (often simultaneously) to control, evaluate and to negotiate the effects of their own and of the other participants’ ongoing talk. It is used by speakers to roleshift, to redefine terms and concepts, to challenge facts/ statistics and to target multiple audiences and to maximize or minimize merits and accountability. Metadiscourse can emphasize, minimize and maximize but also cancel or even reverse ongoing discourse. Martínez Guillem (2009) proposes a broader notion of meta-discourse which includes commenting on the ongoing discursive situation, but is extended to the incorporation of external, interdiscursive and/ or sociocognitive aspects of discourse as she is interested in connecting the ongoing discourse in a plenary session of the European Parliament with its wider socio-cultural context. The use of metadiscourse has been found to vary substantially across different genres and different cultures (cf. Mauranen 1993, 2010, Hyland 2005, Ädel 2006, Penz 2011, Graf 2015). These differences are not surprising in view of findings that indicate that metadiscourse may provide some insights into underlying interactional and pragmatic norms of the language use of a community. In this paper, metadiscourse is defined as ‘discourse about the ongoing discourse’ (cf. Mauranen 1993). However, this definition is extended to include non-verbal elements of communication and pieces of written language used in generally oral small group discussions, as discourse here is viewed from a more multi-modal perspective of meaning making which includes more than the spoken channel. 3. The concept of conflict talk Conflict appears in all spheres of social life and each of us has experienced it on many different occasions and in varying contexts. Conflict is not to be considered an individual’s behaviour, but social conflict always requires two or more individuals who oppose one another (cf. Uhlinger Shantz 1987). In this sense conflict is similar to cooperation, as this is just as crucially dependent on interaction. However, unlike cooperation, conflict has been studied much less in the field of linguistics, even though it can be considered a ‘normal’ part of social interactions. Uhlinger Shantz (1987) further argues that conflicts are social episodes distributed in time, meaning that they have a beginning and an end. Hermine Penz 306 Kakavá (2001: 33) defines conflict as “any type of verbal or nonverbal opposition ranging from disagreement to disputes.” Eisenberg/ Garvey (1981: 150) employ the term adversative episode, which they describe as the “the interaction which grows out of an opposition to a request for action, an assertion, or an action.” In other words, conflict is seen as a sequence that starts with an opposition, develops and ends with its resolution. In his study of conflict among children, Maynard (1985) distinguishes between three phases of conflict, i.e. the antecedent or arguable (the event/ move which causes the conflict), the opposition, and the reaction towards this event. However, he argues that the opposition to a move need not necessarily lead to a dispute or a conflict episode as this only occurs when there is a counter to the opposition. Several other studies have also proposed a tripartite structure of conflicts, which consists of a statement produced by A, a counterstatement (or disagreement with A) by B and a counterstatement (disagreement with) by B (which could possibly be an insistence on the original statement (cf. Gruber 1996, Muntigl/ Turnbull 1998, Nguyen 2011). Hosida and Aline (2015: 231) use the following terminology for this three step structure: “(1) an initial assertion that contains an arguable, (2) an opposition turn, and (3) a counter-opposition turn.” Once the conflictive sequence has been initiated, it may be continued, but also terminated, the latter of which may be accomplished in various ways, which have been described by different scholars. Conflicts may be closed by compromise, submission, concession, withdrawal, stand-off, dominant third party intervention, or humour (Nguyen 2011; Norrick / Spitz 2008; Sharma 2012; Vuchinich 1990) (see Hosida / Aline 2015: 232).” In their analysis of children’s conflicts Eisenberg and Garvey (1981) found that the form of the initial opposition in an adversative episode affects the subsequent strategies of the participants in the interaction, i.e. conflict (resolution) is highly interactive. When looking at oppositions that were responses to a verbal act, Eisenberg and Garvey (1981) found that speakers used one of the five following ways of negating: (1.) using a simple negative, e.g. No, (2.) supplying a reason or justification for disagreement, (3.) making a countering move, e.g. an alternative proposal, (4.) postpone agreement, or (5.) evade or hedge (p. 158). In their analysis, a striking 91.7 % of the adversative episodes were continued when the opposition was a simple No. Research on conflict has mainly focused on two main aspects, i.e. (1.) the origin/ development of conflicts with respect to the situations, reasons, motivations and aims, and (2.) actual and potential ways of solving these conflicts (cf. Gruber 1996: 17). With respect to the origin of conflict, various categorisations and typologies have been proposed. For example, distinctions have been made between conflicts of interests and conflicts of values (cf. Aubert 1973, discussed in Gruber 1996: 19). The former are based on a situation of shortage in which two or more actors Metadiscourse in multi-party conflict talk 307 want the same thing of which there is not sufficient for all. The latter originate in disagreement in relation to the normative status of a social object. In contrast to conflicts of interest, these can rarely be solved by compromise as the involved parties frequently believe that compromise regarding values can never be achieved (cf. Gruber 1996: 19). Although scholars do not always agree on which point to consider as the starting point of an episode of conflict, they all treat conflict as a phenomenon growing out of the interaction of two or more people. Yet, they concentrate their analysis on individual conflict episodes within an interaction, each of them independent from the others. It should be pointed out, however, that interactions in which several conflict episodes have occurred (or when participants have a communicative history of previous conflicts) these previous conflicts may have an influence on the new conflicts arising in the interaction. Conflicts are not necessarily negotiated directly by the immediately involved parties, yet may be dealt with in terms of meta-conflicts: in this case third parties are drawn into the conflict in varying capacities - either as (outside) intermediaries or as co-participants (cf. Gruber 1996: 19, drawing on Bühl 1973). In addition, it has been demonstrated that the situational and institutional context of (group) interactions plays a role in the way conflict sequences develop and which resources are used in the process: in the small group interactions in higher education students resorted to the resources available to them to support their decisions and to challenge other speakers, e.g. discussion task guidelines (cf. Sharma 2012, in Hosoda/ Aline 2015: 234). 3.1 Disagreement and conflict The opposition which follows a statement/ arguable and the counteropposition, which have been viewed as the main structural constituents of conflict, frequently occur in the form of disagreement or correction (cf. Goodwin 1983). Pomerantz (1984: 70) considers disagreement as “dispreferred actions,” which are realised with some delays, repair features, repetitions, etc. However, other researchers have found that in certain contexts, e.g. courtroom situations (cf. Atkinson/ Drew 1979), in casual conversations among Greeks (cf. Kakavá 1993), or in conversations among Chinese and German interactants (cf. Kotthoff 1993) non-delayed disagreements were common. Disagreements are viewed in the research literature as unwanted actions on the one hand but also as a reflection of intimacy (cf. Schiffrin 1984; for an overview see Angouri 2012). Angouri (2012) distinguishes between marked and unmarked disagreement, the latter of which frequently occurs in task-based interactions. In problem solving activities, disagreement is considered to be an inherent component and thus does not carry negative meaning. Hermine Penz 308 In multi-party conflict participants align with other participants in the interaction in the course of conflict talk (Goffmann 1981, Goodwin 1981, 2007, Nguyen 2011). In his study of conflict in a pharmacy patient consultation Nguyen (2011: 1769) demonstrated that alignment in multiparty conflict may be done indirectly without distancing oneself from another party. Participants may also invoke the authority of a non-present outsider. This strategy has been found to serve two functions: it may aid a participant in avoiding directly opposing another and can also be used to place the responsibility for solving the conflict on a person who is not directly involved. Although linguists have studied conflict in detail with respect to their origins, development and resolution, the role of metadiscourse in the process of conflict talk has rarely been studied. Exceptions are Schwitalla (1979), who discussed metadiscourse in political interviews, and Jones (2016), who points towards the crucial role of metadiscourse in accepting or rejecting accountability in connection with racial incidents in the US. Schwitalla’s study of conflictual political interviews found that interactants pursued the conflict mainly by means of metacommunication. Metadiscourse was frequently employed to refer to utterances of the interview partners, yet hardly ever in relation to speakers’ own talk. Disagreement and conflict among speakers almost always affects their relationship, as the content and relationship levels frequently interact with each other (cf. Gruber 1996). This paper endeavours to highlight the role of metadiscourse in various phases of conflict talk in small group discussion in higher education. 4. Data and method The analysis of conflict talk is based on a thirty minute segment of a twohour tape that was recorded during a small group discussion among students in an academic setting in higher education at a US university. The purpose of the group meeting was to do an assignment for a class in linguistics, to be more specific, in lexical semantics, which I participated in and taped as a graduate student in 1989. The conversation was transcribed and then analyzed. In addition, two participants were interviewed regarding their intuitions about the group discussion to avoid the danger of presenting the analyst’s viewpoint as the only viable interpretation, which would represent deceptive objectivity (cf. Tannen 1984). The task of the group and the topic of the discussion was to decompose verbs that are given in the context of a sentence into their semantic primitives (their basic meaning elements) and record the results as well as the problems encountered and the main points of the discussion. In the 30 minutes segment selected for analysis, the decomposition of the verb want in the context of the sentence Sam wants a new bicycle for his birthday was dis- Metadiscourse in multi-party conflict talk 309 cussed. The following two versions of the decomposition were eventually produced on the board: 1. BE EXPERIENCER ORIENT DESIRE TOWARDS THEME State Ind. State Thing Path Thing or: BE SAM ORIENT DESIRE TOWARDS bike State Ind. State Thing Path Thing The fact that the discussion centres on a very specific topic, i.e. the semantic decomposition of verbs, requires the use of a highly specialized vocabulary and could therefore be seen as a discussion employing language for specific purposes. In addition, the linguistic terminology employed represents a type of metalanguage “a specialized form of language or set of symbols used when discussing or describing the structure of a language” (Cambridge Dictionary online). Yet, this particular type of ‘language about language’ will not be attended to as the focus of the analysis at hand is ‘discourse about discourse’. The group interaction is also characterised by a substantial amount of non-verbal and non-oral communication, which are both essential parts of the discussion (the latter includes things such as people writing on the board or people comparing their notes, etc.). Unfortunately, most of this non-verbal and non-oral communication is not captured by the tape. The five participants were graduate students of sociolinguistics at the time of the discussion. Three of them were native speakers (2 NS of American English, 1 NS of British English), two participants used English as a lingua franca (ELF); their native languages were Greek and German, respectively. According to Seidlhofer’s (2011) definition of ELF, though, the conversation cannot be defined as an ELF interaction as the native speakers clearly dominate. However, intercultural issues, such as expectations about how group work should be carried out, might have played a role implicitly, although they are not verbalized at all. Pseudonyms are used for all names, except for mine, which is represented as H. The students’ ages range from mid-twenties to early thirties. John and Pam are American, Jane is British. Anna is Greek and I am Austrian. John and Anna had known each other for two years and were very close friends. John, Anna, Jane and I had known each other as classmates. We also worked together in the first group project for the class in lexical semantics. Pam joined the group as a new member, yet had been friends with Anna for some time. The conflict during this group work severely affected the relationship of some participants: as a result of this group discussion Jane and Pam did not talk to each other for two weeks although they became friends afterwards; Anna had some arguments with John on the same evening as a result of this discussion while Jane reported that she had the impression that she had somehow offended Anna. Hermine Penz 310 The analysis of the data is purely qualitative and takes a broadly Conversation Analytic approach combined with interactional pragmatics 5. Metadiscourse in extended conflicts: Analysis In the identification of metadiscourse ‘discourse about discourse’ is taken to refer to all communicative activities in the group discussion and the connected context of discourse of the classroom. In the data at hand this includes the spoken interaction during group work, the (process of) writing up the results and the (spoken) instructions of the teacher even though these were uttered in the classroom context a few days before. However, these comments are crucially relevant for the task at hand, and the group discussion can be seen as an extension of the classroom discourse, even though in the strict sense the teacher’s words would be considered intertextual as they occur outside of the immediate textual world of the interaction. The extracts selected for the analysis are discussed in sequence as the conflict unfolds. In extended conflict episodes, the question of how a conflict develops and is finally terminated is clearly relevant. What is more, in this process the role of metadiscourse in conflict should become clearer than it would by singling out individual oppositional sequences. Metadiscourse in extract 1 is mostly used to express the participants’ perceptions of how the group work should proceed, relating to communicative activities which include the non-verbal, spoken and written modes. In the conversation preceding the following extract, John has just volunteered to write the decomposition of the sentence as it is being discussed on the board. Extract 1 : 232 Pam: (moves as if to get up to the board) 233 John: What are you doing? 234 Jane: Just thought that would make you happy. 235 Pam: I was just looking 236 I was thinking we could put-like maybe your thing down 237 and then put my thing down. 238 John: Oh I-I was just gonna talk 239 and we can write as we go. 240 Jane: Sure. 241 Pam: I was just kind of saying put maybe 242 Jane: [No, that’s gonna take too long]. 243 Pam: [/ xxxxxxxxxxxxxxxxxxxxxxxxxxxx/ ] 244 Jane: let’s just discuss it..at one time. 245 Pam: I was just thinking, 246 if we had something written down, 247 then we could say, Metadiscourse in multi-party conflict talk 311 248 now we don’t think..[that we think ] 249 Jane: [John is doing that] 250 just to have something that’s written down, 251 you know what I mean? 252 Jane: Well, we are doing that. 253 John: I think, we sh-it’s better that [we just..] 254 Jane: [just put it on] 255 Pam: [O.k.] 256 John: I mean, I am, you know, 257 you could make, let’s see 258 Jane: Uhm, you can make [all comments] you want 259 John: [Is this be? ] 260 Pam: [Yeah When John starts writing and commenting on what he is doing Pam makes a non-verbal move (which can be interpreted as a move to start going to the board). John metadiscursively reacts to her non-verbal action. His question can be interpreted as an opposition/ arguable. Pam clarifies, also using metadiscourse, that she had thought each participant would write their version of the verb decomposition on the board, one after the other (lines 235-237), a response which is a clear indication that she had actually intended to write her version on the board next. In this utterance Pam uses metadiscourse to explain how she understood the group to work. In his subsequent turn John presents his view of the procedure and further clarifies it, i.e. start with the discussion and writing (the results) on the board in the process. Jane agrees with this suggestion, while Pam continues with what is understood as her insistence on proceeding differently (a counter-opposition). Jane immediately rejects this proposition by arguing with lack of time (while overlapping with Pam, lines 242-243), asserting in her next turn that they should discuss one example at a time. When Pam provides a lengthy argument for her view in her next utterance (lines 245-251), Jane comments that they are doing what Pam wants anyway and asks John to continue by saying Just put it on, finishing with an invitation to Pam to provide comments (line 258). In this conflict episode the main antagonists are John and Pam, yet Jane also opposes Pam. She aligns with John as she supports him explicitly by agreeing with him (lines 240) and, implicitly, by opposing Pam (in line 242) (cf. Goffman 1981). It could even be argued that in the course of the interaction the conflict between Pam and Jane becomes equally strong, yet is deflected in the end. In extract 1, the conflict starts with an opposition, the reactions towards which from all the participants are at a moderate level as they all employ mitigating features (mostly included in the metadiscursive utterances) to keep the conflict at a low level in this early stage of the discussion. The opposition is expressed in the form of a question. Although questions are used to express disagreement, they can be considered on Hermine Penz 312 the lower end of the scale when considering other forms of disagreement, such as negative statement plus negative particle no (cf. Pomerantz 1984). In addition, a number of mitigating devices are applied by all speakers, such as maybe, just, conditional, etc. The reaction to John’s opposition and the following interaction are at about the same level in terms of seriousness of conflict, i.e. opposition and reaction directly seem to influence each other. All the participants employ what Garvey and Eisenberg (1981) call “moves to avoid creating offense,” i.e. they employ linguistic forms that can serve this function. Later in the discussion another dispute arises, mainly between Jane and Pam. Anna is taking Pam’s side because she also disagrees with Jane. Strong opposition causes a strong reaction. In this conflict episode (see Extract 2), the disagreement components employed are very strong. According to Pomerantz (1984), delayed disagreement, i.e. disagreement prefaced by silence, hesitating devices, or weak agreement, is often the preferred type, yet this is hardly ever used by any of the participants. The extract also illustrates that the dispute occurring mainly between Jane and Pam is more serious than the previous one. Another phenomenon that surfaces again is that in a multi-party conversation, people immediately take their stance. By agreeing with one of the initial opponents they at the same time disagree with the other. Thus it is possible to disagree with somebody without doing so directly. Metadiscourse in Extract 2 first relates to finding the right term for an element in the decomposition process (which is a type of code gloss) and then changes to applying metadiscourse to oppose the way the written representation of the task is continued and the ensuing counteropposition, followed by a reference to the notetaking skills of the participant who expressed the initial opposition in this passage. Metadiscourse here is used to negotiate a term, which however does not play a role in the conflict. The other instances of metadiscourse surface as a strong opposition (here used to refer to others’ actions/ utterances) and to refer to the teacher. The ensuing conversation centres mainly on the issue of how the notation of the decomposition should be done, that is, the question of whether and where to put brackets or parentheses. The teacher has used brackets in class in order to indicate the relationships between various elements of the decomposition. John, who is still writing on the board, asks whether he needs to close the bracket after what he has just written. Extract 2 : 354 Jane: But because you’re changing the whole kind of statement 355 you need to have another of those 356 half..circle things. 357 What do you call them? 358 Anna: Where? Metadiscourse in multi-party conflict talk 313 359 John: Parenthesis. 360 Anna: Parentheses you mean? 361 Jane: Parenthesis, yeah. 362 O.k., this is the way it needs to be done. 363 This is a parenthesis. 364 Pam: You don’t put that there. 365 I’m sorry. 366 Jane: Yes you do. 367 Pam: Uhuh [I take= 368 Anna: [No, she didn’t 369 Pam: = I take great notes, you have to [understand that] 370 Jane: [Yeah, she did.] 371 Jane: Because we’re getting outside of be state into 372 orient [state.] 373 Anna: [No, no,] no, no 374 Pam: No, she doesn’t= 375 John: Orient state has two arguments = 376 Jane: [O.k. 377 Pam: = Uh look at how she did= 378 John = thing and cause. 379 John: [Yeah 380 Pam: = she did cause. 381 Anna: [No, I don’t, no. 382 Pam: You don’t 383 John: So we get a thing which is an individual. 384 Anna: Afterwards, afterwards. 385 Pam: We can talk to her about [this. 386 Anna: [No, no afterwards, 387 no, she didn’t uh..uh 388 Jane: O.k. under orient goes state, 389 whether or not we use the parenthesis. After using metadiscourse to find the right term for the notational element (lines 356-361), Pam (in line 364) opposes Jane’s assertion in the previous line by using a negative statement without hedging or prefacing. The metadiscourse she uses in her opposition refers to the writing that John produces on the board. Jane opposes Pam in return, restating her position and emphasizing it in line 366. Pam then starts to justify her view and Anna overlaps with her disagreeing with Jane and by doing so, supports Pam’s position. Pam continues to justify her view by referring to her note-taking skills, emphasizing her statement by adding the metadiscursive label you have to understand that (line 369), thus appealing to a higher authority (the teacher’s) to support her view. This is jokingly acknowledged by Jane in line 370. Jane, in line 371, also gives a reason for her opinion, which is met by very strong opposition from Anna, who disagrees by saying no four times, overlapping with Jane. Pam in the following turn again makes a negative statement, starting with no and disagrees with Jane. She again refers to the authority of the teacher continuing to explain how the teacher did the decomposition according to Hermine Penz 314 her notes (in lines 377 and 380). In line 381, Anna latches on to Pam’s utterance, again disagreeing with Jane in line 382, but then mitigates her opposition by suggesting consulting the teacher about this. While Pam’s opposition becomes weaker here, Anna’s disagreement with Jane becomes stronger. She emphasizes her position in line 386 by repeating afterwards and prefacing it with a double negative, continuing with her disagreement in the following line. Jane then drops the issue in question by talking about how to proceed further in the decomposition (lines 388-389). The above discussion demonstrated that strong opposition causes a strong reaction. It appears that the interaction of the initial opponents is influenced by each other. However, it is interesting to note that Anna’s very strong opposition, which is directed toward Jane, is not met by a strong reaction on her part. It seems that Jane is focusing on her interaction with Pam much more than with Anna, who joins as a second opponent to Jane, i.e. after Pam. However, it is very obvious that apart from the discussion of the term parenthesis, metadiscourse is mainly employed by Pam in this extract, in one instance to express her opposition You don’t put that there, in the other cases to appeal to the authority of the teacher, which is a way to strengthen her position (lines 369, 385). After some more discussion and argumentation about how to proceed further with the decomposition, another serious dispute arises between Pam and Jane. It reaches its peak in Jane’s refusal to continue contributing verbally, yet ends a few turns later with Jane conceding. Metadiscourse here is largely connected with words of saying and the process of writing up the example on the board. Extract 3: 475 John: All right, you tell me what next. 476 Jane: O.k. close bracket after orient. 477 John: Big parenthesis? 478 Pam: You do NOT close the bracket. 479 Jane: / xxxxxx/ keep it open 480 Pam: = you have an open parenthesis. 481 Jane: keep it open, o.k. after orient just comma. 482 John: So we got two things here? 483 Jane: I am not [saying any more,] 484 John: [/ xxxxxxxxxxxxxx/ ] 485 YOU tell him. 486 John: Orient, that’s got a thing. 487 Pam: Yeah. 488 John: Like that? 489 Jane: No. 490 Pam: No. 491 Jane: Why don’t you go up there and do [it then? ] 492 John: [Fine.] 493 Anna: You have to open parenthesis. Metadiscourse in multi-party conflict talk 315 494 Jane: [Yeah I mean 495 I can’t do [it right either], so... 496 Anna: [you have to ] 497 open the parenthesis and then 498 Jane: Can you show it on the board? 499 Pam: [Yeah. In Extract 3, John is still writing on the board and asks how he is to continue. His utterance is clearly metadiscursive as it includes tell as a word of saying. Jane instructs him to close the bracket after the word orient. Her utterance is directly opposed by Pam in line 478. In the following line Jane concedes and adopts Pam’s view. They both continue in their respective subsequent utterances to instruct John to not include parentheses. In other words, Jane and Pam agree on the content level. However, when John asks another question in line 482, Jane produces metadiscourse to address a clear opposition to Pam: she states her refusal to continue giving advice to John and tells Pam to proceed. In the subsequent turns Jane actually continues to contribute, as does Pam. Yet when Pam disagrees again with John (as does Jane), Jane asks Pam to go to the board and proceed with the decomposition in the form of the whquestion Why don’t you go up there and do it then? (line 491), to which Pam consents when Jane utters her request a second time (line 498). What is interesting in this extract is that Jane’s opposition to Pam appears to be caused by antagonism which is based on the ‘how’ or the fact that Pam takes a fairly dominant role rather than on the ‘what’, i.e. the contents. At the same time, Jane’s opposition could be read as a way of voicing her frustration about the difficulty of the task, which also surfaces in the subsequent turns, where she finally invites Pam to continue writing the decomposition on the board, by which the conflict seems to be terminated. Metadiscourse is heavily deployed in the development of conflict and its resolution in the following extract involving John and Anna as the main antagonists. A variety of functions appear in this interaction, as it is used to structure the discourse, e.g. that’s the first one, the second one, to label speech actions (preceding, ongoing, subsequent), most of which play an important role in initiating and terminating the conflict. A combination of bringing up a proposal that was rejected on several occasions throughout the group work and therefore caused disagreement, and the fact that this also seemed to bring the discussion back to the example that Anna repeatedly refused to discuss before, culminate in a violent verbal dispute which occurs at the end of the discussion analysed. Prior to the conversation in the following transcript, Anna has written her version of the decomposed sentence under Jane’s. They have found that there were only two differences and seemed to be happy with their accomplishment. Hermine Penz 316 Extract 4: 1390 Anna: O.k., that’s the first one 1391 The second one. 1392 The economy [frightens ] 1393 John: [Can we maybe] I don’t know, 1394 go about this in a slightly different way? 1395 Jane: Yes. 1396 John: Because to me we got caught up in orientation, 1397 I mean in notation. 1398 Yeah 1399 Anna: John, please. 1400 John: What? 1401 Anna: Let’s drop that. 1402 John: What? 1403 Anna: Please. 1404 Jane: No, no, no, no = 1405 John: I really= 1406 let him say what he wants to say. 1407 John: = think that uhwe should try to capture 1408 some of the hierarchical relations. 1409 Jane: [Right. 1410 We don’t have to do it like a tree thing, 1411 like this. 1412 Jane: [Uhm 1413 But just..forget all the bullshit 1414 about little parentheses 1415 Jane: [Right 1416 and let’s capture the major relations. 1417 [Uhm 1418 and we can write it out, 1419 Jane: / xxxxxxxxx parentheses/ 1420 John: just to capture it. 1421 Jane: [That’s great. 1422 so that we agree on the semantics, 1423 Jane: [Yeah 1424 and then we can go back 1425 and then Pam is our expert on notation, 1426 and then maybe you can convert that, 1427 Pam: [Uhm 1428 but I would feel 1429 we would benefit from talking 1430 Pam: [Uhm 1431 from the most abstract semantic function, 1432 Pam: [Uhm 1433 down to one of these things and then so on. 1434 Anna: We have spent half an hour= 1435 Jane: Ve[ry good] 1436 Anna: = [talking] about this and disagreeing, 1437 and we haven’t come to a conclusion. 1438 And [I don’t think] 1439 Jane: [That’s not true.] Metadiscourse in multi-party conflict talk 317 1440 it- I want, I don’t think 1441 we’re gonna reach any conclusion... 1442 A: nd we might go out tomorrow to class on 1443 Tuesday in class, 1444 and she will present a whole different framework 1445 Jane: [Yep 1446 and then we will say, 1447 o.k., we spent one hour talking about this, 1448 while we can go on with the rest, 1449 because I don’t know 1450 what we are going to find with the rest. 1451 So please, let’slet’s look at the others.= 1452 John: (sighs) 1453 Anna: =And if we have time, let’s go back. 1454 How about that? 1455 John: How [does that] disagree 1456 Jane: [I agree] 1457 from what we are doing? 1458 Anna: No, because we’re going back to the same thing, 1459 and we have found-found all these arguments, o.k. 1460 Let’s move to the second one. 1461 Pam: [Uhm 1462 because 1463 John: [That’s what I am proposing. 1464 Jane: What the hell are you.. 1465 we are going to the second. 1466 H: Do the second one, [but in a different way.] 1467 Anna: [I thought you said it ] 1468 for the first one. 1469 John: No, no. 1470 Jane: Anna, there is nothing wrong with disagreeing 1471 on how this is meant to be mapped out. 1472 John: No, I agree, 1473 that we were beating the first one on a dead horse. 1474 Jane: What’s wrong with that? 1475 John: = and that’s why I was trying for the second one. 1476 Anna: O.k., o.k. 1477 John: I was suggesting that we take= 1478 Anna: O.k. 1479 John: = a step back approach. 1480 Anna: Sorry, I thought you meant the [first one,] that’s why 1481 John: [No, no, no.] 1482 Jane: I want EVERYBODY to calm down, 1483 Anna: [O.k. 1484 this is supposed to be fun. 1485 Anna: No, I thought youhe meant for the first one 1486 John: Yeah. 1487 No, I mean, we’d [started in a sense] 1488 Jane: [the first one’s dead] 1489 number one. Hermine Penz 318 1490 I was trying the....thing, 1491 and I got messed up and stuff. The extract above starts with metadiscourse relating to the structuring of discourse: in line 1390 Anna utters a closing statement with respect to the first example of the task. She considers it being dropped and is ready to go on to the second one, which she starts to read aloud. John interrupts her and suggests (in lines 1393-94) slightly changing the approach to the given task. Again, he uses metadiscourse to refer to the process of how they should go about their task, to which Jane agrees. He continues arguing that everyone had become too confused by the bracket notation. Anna pleads with him to drop this (lines 1399, 1401), thus expressing her opposition. Metadiscourse is employed by her to label the speech activity Let’s drop that. John is not sure what she means and asks for clarification. Jane takes a turn, opposing Anna by saying no four times and insisting that she should let him say what he wants to say, which is again an instance of metadiscourse. In lines 1407-1433, John suggests in great elaboration that the hierarchical relationships between the various elements of the decomposition should not be indicated with brackets, but with an alternative, and provides a number of reasons for his suggestion. In his elaboration he employs a high amount of metadiscourse to refer to the structure of the discourse. John receives positive verbal feedback from both Jane and Pam. Anna reacts with an outburst of anger, when he finishes his lengthy turn overlapping partly with Jane, who positively acknowledges John’s proposal with an evaluative very good (lines 1434-1435). Emphatic expression and pitch of voice clearly indicate that Anna is extremely upset. In her subsequent argument (containing her opposition to John’s suggestion), she complains that half an hour had been spent on one example without coming to any conclusion (lines 1434, 1436-1437). Jane strongly disagrees with her in line 1439. Anna continues her enraged argument saying she does not think they would reach any conclusion and eventually find that the teacher would present a totally different solution. She pleads that they proceed with the other examples. Her extended argument can be viewed as an accumulation of metadiscourse, a large part of which is made up by discourse labelling of preceding and subsequent actions, such as we have spent half an hour talking about this and disagreeing and we have not come to a conclusion, followed by metadiscourse labelling subsequent speech activities, e.g. I don’t think we’re gonna reach any conclusion, etc., finally pleading that they should discuss other examples. John reacts to Anna’s opposition by questioning how her proposal disagrees with his, supported by Jane. Both employ discourse labelling in their requests. After John, Jane and H confirm (lines 1463-1466) that they were proposing the same procedure, Anna concedes that she had interpreted this to apply to the first example, which can be interpreted as Metadiscourse in multi-party conflict talk 319 an explanation and a justification for her outburst. From this point onwards, conflict resolution progresses. Again, a great amount of discourse labelling is employed by all participants to name communicative actions and intentions in this process. Metadiscourse is particularly salient in this last extract, where the participants’ steps, activities and intentions in discourse are made explicit both in the initiation, development and resolution of the conflict. Raising the level of explicitness appears to be an important component of conflict episodes of the type discussed. It also seems to be the case that this level of explicitness is not just maintained by the parties who are directly involved in the conflict but by all the participants who contribute their views in the process. 6. Discussion and conclusion As has been mentioned in Section 2 already, the use of metadiscourse is also dependent on the genre. The data analysed in this article can be defined as task-oriented academic discourse as the main purpose of the group discussion is to accomplish a task set by their course teacher. Metadiscourse in this setting is concerned with all the relevant communicative aspects such as spoken and written discourse, as well as nonverbal interactions which occur throughout this speech activity (cf. Levinson 1992 for the concept of speech activity). Similar to other studies of task-based group discussions, metadiscourse is frequently applied to clarify procedural aspects relating to the ‘how’ in discourse. In this respect similarities can be found with intercultural project discussions (see Penz 2007, 2011). It has been demonstrated that metadiscourse plays an important role in all phases of conflict (initiation, development and termination) and is employed by all participants. In many cases, metadiscourse is resorted to for expressing opposition, which is the starting point of the conflict. Similarly, this also applies to counter-opposition. In the course of the group discussion questions of discourse structuring also play an important role. These are also frequently dealt with metadiscursively. The dynamics of conflict episodes in the group also turn out to be interesting: in this multi-party interaction other participants may align with the initial opponent, or a conflict may shift to other parties. Since high explicitness has been found to be a characteristic feature of task-based and work-related interactions in cooperative interactions, it is not surprising that conflict resolution is also largely accomplished through metadiscourse. An interesting finding, which applies mainly to the most extended conflict episode (Extract 4), is that in multi-party interactions where one or two participants employ metadiscourse extensively, the other members of the group will join, thus resulting in accumulated metadiscursivity. Hermine Penz 320 With respect to the outcome of the group discussion it is interesting to note that the conflicts affected the relationships of the people involved. This became apparent in the interviews after the first analysis and has already been discussed in Section 4. Even though Angouri’s (2012) research has shown that (even strong) disagreements may not necessarily have a negative effect in the activity type of task-related group work, the results in the present study clearly influenced people’s relationships negatively. This article has only contributed a small piece to the puzzle of how metadiscourse is applied in conflict episodes. I hope that future research will extend this endeavour and study metadiscourse and conflict in additional genres and different cultural contexts. References Ädel, Annelie (2006). Metadiscourse in L1 and L2 English. Amsterdam: John Benjamins. Angouri, Jo (2012). “Managing disagreement in problem solving meeting talk.” Journal of Pragmatics 44. 1565-1579. Archibald, Alasdair/ Alessia Cogo/ Jennifer Jenkins (eds.) (2011). Latest Trends in ELF Research. Newcastle upon Tyne: Cambridge Scholars Publishing. Atkinson, Maxwell J./ Paul Drew (1979). Order in Court. Atlantic Highlands, NJ: Humanities Press. Atkinson, Maxwell J./ John Heritage (eds.) (1984). Structures of Social Action: Studies in Conversation Analysis. Cambridge: CUP. 57-101. Aubert, Vilhelm (1973). “Interessenkonflikt und Wertkonflikt. Zwei Typen des Konflikts und der Konfliktlösung.” In: Bühl (1973). 178-205. Brinker, Klaus/ Gerd Antos/ Wolfgang Heinemann/ Sven F. Sager (eds.) (2001). Text- und Gesprächslinguistik/ Linguistics of Text and Conversation. KSK 16.2. Berlin & New York: de Gruyter. Bublitz, Wolfram (2001). “Formen der Verständnissicherung in Gesprächen.” In: Brinker/ Heinemann/ Sager (2001). 1330-1340. Bublitz, Wolfram/ Axel Hübler (eds.) (2007). Metapragmatics in Use. Amsterdam. John Benjamins. Bühl, Walter (ed.) (1973). Konflikt und Konfliktstrategie. Ansätze zu einer soziologischen Konflikttheorie. Neuwied: Sammlung Luchterhand. Caffi, Claudia 1984. “Some remarks on illocution and metacommunication.” Journal of Pragmatics 8(4). 449-467 Cambridge Dictionary Online. (n.d.) [Online] http: / / dictionary.cambridge.org/ de/ worterbuch/ englisch/ metalanguage (accessed 14 November 2017). Ciliberti, Anna/ Laurie Andersen (2007). “Metapragmatic comments in institutional talk: A comparative analysis across settings.” In: Bublitz/ Hübler (2007). 143-166. Crismore, Avon (1989). Talking with Readers. Metadiscourse as Rhetorical Act. New York: Peter Lang. Dittmar, Jürgen (ed.) (1979). Arbeiten zur Konversationsanalyse. Tübingen: Niemeyer. Metadiscourse in multi-party conflict talk 321 Dobrić, Nikola/ Eva-Maria Graf/ Alexander Onysko (eds.) (2016). Corpora in Applied Linguistics Current Approaches. Newcastle upon Tyne: Cambridge Scholars Publishing. Drew, Paul/ John Heritage (eds.) (1992). Talk at work: Interaction in institutional settings. Cambridge: CUP. Eisenberg, Ann R./ Catherine Garvey (1981). “Children’s use of verbal strategies in resolving conflicts.” Discourse Processes 4. 149-170. Gee, James P. (2011). An Introduction to Discourse Analysis. Theory and Method. 3 rd ed. London & New York: Routledge. Goffman, Erving (1981). Forms of Talk. Philadelphia, PA: University of Pennsylvania Press. Goodwin, Charles (1981). Conversational Organization: Interaction between Speakers and Hearers. New York: Academic Press. Goodwin, Charles (2007). “Interactive footing.” In: Holt/ Clift (2007). 16-46. Goodwin, Marjorie H. (1983). “Aggravated correction and disagreement in children’s conversations.” Journal of Pragmatics 7. 657-677. Graf, Eva-Maria (2015). The Discourses of Executive Coaching. Linguistic Insights into Emotionally Intelligent Coaching. Habilitationsschrift Universität Klagenfurt. Grimshaw, Allen D. (ed.) (1990). Conflict Talk. Sociolinguistic investigations of arguments in conversations. Cambridge: CUP. Gruber, Helmut (1996). Streitgespräche. Zur Pragmatik einer Diskursform. Opladen: Westdeutscher Verlag. Gruber, Helmut (1998). “Disagreeing: Sequential Placement and Internal Structure of Disagreements in Conflict Episodes.” Text 18. 467-503. Gunnarsson, Britt-Louise/ Per Linell/ Bengt Nordberg (eds.) (1994). Text and talk in professional context. Uppsala: ASLA. Halliday, Michael A.K. (1994). Introduction to Functional Grammar. 2 nd ed. London: Edward Arnold. Hockett, Charles F. (1960). “The Origin of Speech.” Scientific American 203(3). 88- 96. Holt, Elizabeth/ Rebecca Clift (eds.) (2007). Reporting Talk: Reported Speech in Interaction. Cambridge: CUP. Hosoda, Yuri / David Aline (2015). “Single episode analysis of extended conflict talk sequences in second language classroom discussion.” Journal of Language Aggression and Conflict 3(2). 231-262. Hyland, Ken (2005). Metadiscourse. Exploring Interaction in Writing. London & New York: Continuum. Ilie, Cornelia (2003). “Discourse and metadiscourse in parliamentary debates.” Journal of Language and Politics 2(1). 71-92. Jacobson, Roman (1990/ 1976). “The Speech Event and the Functions of Language.” In: Waugh/ Monville-Burston (1990). 69-79. Jones, Rodney H. (2016). Spoken Discourse. London [etc.]: Bloomsbury. Kakavá, Christina (1993). Negotiation of disagreement by Greeks in conversations and classroom discourse. Unpublished PhD thesis. Georgetown University, Washington DC. Kakavá, Christina (2001). “Discourse and Conflict.” In: Schiffrin/ Tannen/ Hamilton (2001). 650-670. Kotthoff, Helga (1993). “Disagreement and concession in disputes: On the context sensitivity of preference structures.” Language in Society 22. 193-216. Hermine Penz 322 Lee, Joseph J./ Nicholas C. Subtirelu (2015). “Metadiscourse in the classroom: A comparative analysis of EAP lessons and university lectures.” English for Specific Purposes 37. 52-62. Levinson, Stephen, C. (1992). “Activity types and language.” In: Drew/ Heritage (1992). 66-100. Luuka, Minna R. (1994). “Metadiscourse in academic texts.” In: Gunnarsson/ Linell/ Nordberg (1994). 77-88. Martínez Guillem, Susana (2009). “Argumentation, metadiscourse and social cognition: organizing knowledge in political communication.” Language in Society 20(6). 727-746. Mauranen, Anna (1993). Cultural Differences in Academic Rhetoric. Frankfurt a.M.: Peter Lang. Mauranen, Anna (2010). “Discourse Reflexivity - A Discourse Universal? The Case of ELF.” Nordic Journal of English Studies 9(2). 13-40. Mauranen, Anna (2007). “Discourse reflexivity and international speakers - how is it used in English as a lingua franca? ” Jezik in slovstvo 52(3-4). 1-19. Maynard, Douglas W. (1985). “How children start arguments.” Language in Society 14. 1-29. Mey, Jacob L. (2001). Pragmatics. An Introduction. Oxford: Blackwell. Muntigl, Peter/ William Turnbull (1998). “Conversational Structure and Facework in arguing.” Journal of Pragmatics 29(3). 225-256. Nguyen, Hanh thi (2011). “Boundary and Alignment in Multiparty Conflict Talk.” Journal of Pragmatics 43. 1755-1771. Norrick, Neal R./ Alice Spitz (2008). “Humor as a resource for mitigating conflict in interaction.” Journal of Pragmatics 40(10). 1661-1686. Penz, Hermine (2007). “Building common ground through metapragmatic comments in international project work.” In: Bublitz/ Hübler (2007). 263-292. Penz, Hermine (2011). “‘What do we mean by that? ’ Metadiscourse in ELF Project Discussions.” In: Archibald/ Cogo/ Jenkins (2011). 185-201. Penz, Hermine (2016). “The uses and functions of metadiscourse in intercultural project discussions on language education.” In: Dobrić/ Graf/ Onysko (2016). 95-120. Pomerantz, Anita (1984). “Agreeing and Disagreeing with Assessments: Some Features of Preferred/ Dispreferred Turn Shapes.” In: Atkinson/ Heritage (1984). 57-101. Schiffrin, Deborah (1980). “Metatalk: Organizational and Evaluative Brackets in Discourse.” Sociolinguistic Inquiry: Language and Social Interaction 50. 199-236. Schiffrin, Deborah (1984). “Jewish argument as sociability.” Language in Society 13. 311-335. Schiffrin, Deborah/ Deborah Tannen/ Heidi E. Hamilton (eds.) (2011). The Handbook of Discourse Analysis. Oxford: Blackwell. Schwitalla, Johannes (1979). “Metakommunikation als Mittel der Dialogorganisation und der Beziehungsdefinition.“ In: Dittmar (1979). 1-143. Seidlhofer, Barbara (2011). Understanding English as a Lingua Franca. Oxford: OUP. Sharma, Bal Krishna (2012). “Conceding in Disagreements during Small Group Interactions in Academic Writing Class.” Classroom Discourse 3. 4-28. Tannen, Deborah (1984). Conversational Style. Analyzing talk among friends. Norwood, NJ: Ablex. Uhlinger Shantz, Carolyn (1987). “Conflicts between children.” Child Development 58. 283-305. Metadiscourse in multi-party conflict talk 323 Vande Kopple, William J. (1985). “Some Exploratory Discourse on Metadiscourse.” College Composition and Communication 36. 63-94. Vuchinich, Samuel (1990). “The Sequential Organization of Closing in Verbal Family Conflict.” In: Grimshaw (1990). 118-138. Watzlawick, Paul/ Janet Beavin Bavelas/ Don D. Jackson (1967). Pragmatics of Human Communication. New York & London: W.W. Norton & Company. Waugh, Linda R./ Monique Monville-Burston (eds.) (1990). On Language. Roman Jacobson. Cambridge, MA & London: Harvard University Press. Appendix Transcription conventions, adapted from Tannen (1984: xix): .. noticeable pause or break in rhythm (less than 0.5 pause) ... half second pause (an extra dot is added for each half second pause) underline marks emphatic stress CAPS marks higher intensity . marks sentence final intonation ? marks yes/ no question rising intonation - marks a glottal stop, or abrupt cutting off of sound, as in “uh-uh” / xxx/ indicates transcription impossible / words between slashes/ within slashes indicate uncertain transcription = continuation of utterance [ lines in brackets ] overlapping speech [ ] two people talking at the same time (italicised words in brackets) comment on quality of speech, context and non-verbal communication Hermine Penz Department of English Studies Karl-Franzens-University Graz Austria