Analysis of Negotiation Episodes in Foreign Language Learner Interactions

In second/foreign language learning, it is hypothesised oral interaction facilitates language acquisition. Studies show that one aspect of interaction that promotes second/foreign language acquisition is through the process of negotiation, and one factor that influences interaction is the task or activity that learners are engaged in. Hence, it is the aim of this study to investigate if foreign language (FL) learners would engage in different negotiation strategies when completing different communication task types. Nine FL learners from a tertiary EFL class in groups of three participated in the study. They completed three different communication task types (i.e., information gap, jigsaw and decision-making) over a period of four weeks. Data for the study comprised transcribed recordings of the groups’ oral interaction. This study qualitatively examined FL learners’ oral interactions when engaged in three different types communication tasks. Data collected were analysed for instances of negotiation. Findings revealed that the different communication tasks elicited negotiation episodes as the participants engaged in task completion. They applied similar negotiation strategies during peer interaction. However, further scrutiny revealed that there were differences in terms of the depth of negotiation for the different communication task types. The results of this study exhibit evidence that language communication tasks do promote meaningful interactions among tertiary EFL learners. Data clearly showed widespread negotiation episodes during task completion. The knowledge on the types of communication tasks that can promote meaningful interaction and negotiation episodes can assist language practitioners to make informed decisions on tasks that are suitable for their learners.


INTRODUCTION
The growing interest in classroom interactions reflects a theoretical shift in perspectives on learning and instruction.Learning is seen not only as a "constructive process that takes place in the mind of the learner but also as a process of meaning-making and enculturation into social practices" (Kumpulainen & Wray, 2002, p. 3).These views of learning have pedagogical implications.The views have changed the traditional interaction patterns of many classrooms as well as the roles of teachers and learners as communicators and learners.Learners are no longer passive receivers of knowledge.They are more involved in learner-centered learning activities and collaborative work whereby lessons are executed in

LITERATURE REVIEW
In second language (L2) learning, it is hypothesised that oral interaction facilitates second language acquisition (SLA) (Kurhila & Kotilainen, 2017;Long, 2017;Luan & Sappathy, 2011;Martin-Beltrán, 2017;Sato & Ballinger, 2016).What happens during interaction is that learners will usually receive feedback on the form and meaning of their messages, whether their messages are clear or otherwise.If their messages are not clear, they would be informed in many ways by the other speakers.For example, the other speakers may ask questions, seek clarification and request for explanation.Hence, the learners need to adjust their messages so that they become clear, comprehensible and can be understood by other speakers.This process of adjustment leads to modification of speech to enhance the comprehensibility of their messages.In other words, they are pushed in their production of language.In doing so, simultaneously, they may also push the other speakers/learners to do the same i.e. modify their speech.The whole sequence of receiving input and producing output is vital for language acquisition and development (Crossley et al., 2016;Ma et al., 2017;Philp & Iwashita, 2013).
What the speakers do when they receive input and produce output is that they are actually engaged in the process of negotiation.Negotiation is one aspect of interaction that promotes SLA.Researchers have found that negotiation plays an important role in SLA as it provides opportunities for interactional modifications to occur and this increases input comprehensibility (Long & Robinson, 1998).There is a large body of research based on the concept of negotiation (e.g.Foster & Ohta, 2005;Jeong, 2011;Long, 2011;Luan & Sappathy, 2011;Mayo & Ibarrola, 2015;Pica, 1991;Rouhshad et al., 2016;Yi & Sun, 2013) and the areas researched are wide and extensive.For instance, studies ranged from examining effects of age on negotiation (García Mayo & Imaz Agirre, 2016;Mayo & Ibarrola, 2015), examining different modes of interaction and negotiation (Rouhshad et al., 2016), investigating effects of training learners to use various meaning negotiation strategies (Tai et al., 2016) as well as identifying communication strategies in negotiation of meaning employed by learners when engaged in computer mediated communication (Omar & Kaur, 2017) among others.
The study on the negotiation for meaning has been closely linked to Long's Interaction Hypothesis (1983).The Interaction Hypothesis suggests that interaction provides learners with opportunities to receive the target language as input, produce output, and through interactional modifications or adjustments, learners direct their attention to mismatches between their interlanguage and the target language.It is through interaction, learners receive comprehensible input and feedback from their interlocutors, and make modifications to their output.Findings from research on interaction suggest that second language learning and development occur when learners have the opportunities "to negotiate for comprehensible input, receive feedback, and modify their output" (Mackey, 2007, p. 100).In other words, when learners interact, they engage in interactional processes.These processes, which include negotiation for meaning, feedback, recasts and production of output that take place during interaction have been claimed to facilitate L2 learning.
In language classrooms, tasks become the driving force for language use (Baharun, 2015;Baharun et. al, 2016;Baharun & Zakaria, 2017;Eslami & Kung, 2016;Fernandez Dobao, 2012;Fondo Garcia & Appel, 2016;Kaivanpanah & Miri, 2017;Lan et al., 2016;Nassaji & Tian, 2010).Tasks offer interaction opportunities to learners as they function as stimuli for generating talk among learners.Communication tasks have often been associated with conveying messages and meaning across.They also have been used by L2 researchers to investigate learner interactions generated by learners when they engage in task completion.It is found that when communication tasks are used in language teaching, learners engage in episodes of negotiation of meaning and this is believed to have facilitated second language acquisition (e.g.Luan & Sappathy, 2011;Nassaji & Tian, 2010;Yi & Sun, 2013).Nonetheless, despite the fact that for more than 20 years, tasks have become well established as a unit of design in a communicative curriculum to encourage learners to engage in realistic communication during task performance and completion (Crabbe, 2007), it is believed that there are still areas that could be explored and investigated as tasks are important tools "for providing communicative opportunities that are likely to lead to implicit learning" (p.124).The literature review and the argument provide grounds for the researchers to study different communication task types in order to identify those that would offer learners maximum opportunity to engage in interactive interaction during task completion.

METHODOLOGY
This section describes the research design of the study.It presents a detailed description of the research process undertaken.It includes the focus of the study, the participants, the method of data collection and data analysis.

FOCUS OF STUDY
This study aimed to investigate if FL learners will engage in different negotiation strategies when completing different communication task types.It examined FL learners' oral interactions when engaged in three different communication task types.Each task type had a stimulus that became the point of reference and discussion for the participants.The participants worked in teams of three.They were required to engage in peer interaction with one another to comprehend the input given to them, to work collaboratively and to make group decisions to complete the tasks given.The data for the study comprised transcribed recordings of group work interaction in class.
Three types of communication tasks, namely the information-gap, jigsaw and decision-making were used.The tasks were chosen because of their different characteristics and capacity to elicit episodes of meaning negotiation (see Ellis, 2000;Pica, 2005;Pica et al., 1993).As learner negotiated interactions were the focus of this study, thus the three task types were selected to be used and investigated.
For the information-gap task, the participants were given a set of jumbled-up pictures and a set of written instructions.Input was not equally distributed among them.By this it meant that among them, only one participant had the instructions while the rest had a set of jumbled-up pictures.Thus, the one holding the instructions was required to provide the information to those having the jumbled-up pictures.Those holding the pictures had to request for information, ask questions for confirmation and clarification purposes, and get the relevant information.This was to enable them to arrange the pictures accurately.According to Pica et al. (1993), due to the fixed roles assigned to each interactant, "mutual opportunities for working toward comprehension, feedback, and interlanguage modification are limited" (p.21).
For the jigsaw task, each participant in the team was given a different portion of input taken from the total information.Since none of the participants had total or similar access to the information needed for task completion, they needed to exchange and share the input with one another.The participants needed to sit together in their own teams, engage in peer interaction and collaboratively complete the tasks.As pointed out by Pica et al. (1993), due to its unique characteristics, the jigsaw task "can be considered the type of task most likely to generate opportunities for interactants to work toward comprehension, feedback, and interlanguage modification processes related to successful SLA" (p.21).
As for the decision-making task, all the participants had total and similar access to the information needed for task completion.They all shared the same input (a set of jumbled-up pictures and a set of pictures with a set of related words) so they started out with shared access to the information.However, as highlighted by Pica et al. (1993), for the completion of the decision-making tasks, "interaction is not necessary (-) in order for the participants to carry out the task, as one participant can work individually using the information" (p.22) and due to this, with this task type "comprehension, feedback, and production opportunities become increasingly reduced" (p.22).Details of the task types used in the study are presented in Table 1.

Jigsaw
Based on the descriptions of the three task types, it is hypothesised that learners would interact with one another when engaged in task completion for all the task types.However, due to the unique characteristics of the different task types, it is believed that there would be differences with the jigsaw being the task that would provide most opportunities for negotiated interaction and hence, most number of negotiated episodes.While it is important to investigate the relationships between task types and learners' performance and achievement statistically, Nakahama, Tyler and van Lier (2001) argue that by "attending only to the overall numbers of repair negotiations masks important discourse dynamics and therefore masks important learning opportunities beyond the ideational or informational level " (p. 388).This means that it is equally important to analyse learner interactions qualitatively to understand the dynamics of the discourse as learners engage in task completion.Shimamura (2015) also discusses the "necessity of qualitative approaches to classroom research" (p. 1).Thus, in this study, learner interaction was qualitatively explored to identify the kind of negotiated episodes and negotiation strategies that the participants employed as they engaged in different task types task completion.Specifically, the present study aimed to answer the following research question: • What are the negotiation strategies exhibited in learners' oral i nteraction when they engage in different communication task types?
Input on the kind of negotiated strategies exhibited by learners would inform the researchers on the dynamics of the discourse when they engage in completion of different communication task types.

THE PARTICIPANTS
An intact class of 18 (20-22 years old) EFL participants from a public university in Malaysia participated in the study.The participants were chosen based on several criteria.Firstly, they came from an intact class and were all majoring in the same programme, Bachelor of Communication.In terms of their educational background, they all came from religious secondary schools and shared the same mother tongue or first language which was Bahasa Melayu (Malay language).In relation to the English language, they had studied English as a subject in school for at least 11 years since primary education (6 years) until they finished their secondary education (5 years).As for their MUET achievements, ten of the participants had obtained a Band 1 while the rest were Band 2 achievers.Hence, their levels of English language proficiency based on MUET results were not vastly different from one another.
The 18 participants were divided into groups of three: Red, Blue, Brown, Green, Yellow and Purple.Pseudonyms were given to protect the identities of the participants.

DATA COLLECTION
The participants from the teams worked in groups to complete the assigned tasks.They dealt with task completion without any explicit teaching involved.Three different communication task types (i.e.information-gap, jigsaw and decision-making) were used.On the days of data collection, tasks were given to the teams.They were asked to sit in their teams and attempt the tasks.Following the instruction given, they sat together and attempted to complete the tasks.Their oral interactions generated during task completion were audio-recorded and transcribed verbatim for analysis.However, due to the issue of attendance, data from only three groups, nine participants in total, were taken as they were found to be most complete and comprehensive.They were the Red, Blue and Brown groups.
No specific time was given to the teams to complete the tasks.However, in total, the three teams took approximately 214 minutes to complete the tasks as presented in Table 2. Data collected were analysed for instances of negotiation.They were qualitatively analysed to identify the types of negotiation strategies produced by the participants in different communication task types.The data collected were coded based on Long's definitions (1983) of confirmation checks, clarification requests and comprehension checks.These categories were chosen as they were established categories and have been used in a number of studies investigating episodes of negotiation found in interaction (e.g., Long, 1996;Nakahama et al., 2001).However, for this study, adaptation was made based on the data gathered which was situationally defined.It was observed that when the participants engaged in peer interaction, they were engaged in other strategies as well.They asked for help, asked for and provided explanation, and argued and expressed disagreements.Hence, the list was expanded to include asking for explanation and providing extended answers or responses, and expressing disagreement.Instances of negotiation which included these negotiation strategies were coded and analysed to give evidence to negotiated interactions found in the oral discourse as the participants engaged in task completion of the selected communication task types.Details of the categories used for coding of data are presented in Table 3. F: Now this is sea pollution.A: Sea pollution, not this?(asking for clarification) F: This water pollution.This also water pollution.

Asking for help
Asking for help occurs when a participant wants to ask for help/assistance from the other participants regarding a linguistic form, meaning of words, translation of words, spelling of words etc.

W: So terrorism is what? (asking for help -meaning of words)
A: Drought? (asking for help -meaning of words) W: Logging? (asking for help -meaning of words) A: Logging? (asking for help -meaning of words) W: But this land fields?(asking for help -meaning of words) F: No. Asking for explanation Asking for explanation is an attempt by a participant to request for further explanation/information from the other participants regarding what is being discussed, meaning of words, etc.

S: Rrr…Sorry I don't understand.
What what what do you mean by painkiller?(asking for explanation) Giving extended explanations An extended explanation is an attempt by a participant to provide more information/explanation when providing answers to questions.
F: What's meaning by hunger?(asking for explanation) S: Hunger is…hunger is for people who are hungry.There's hunger…there's…the people… (giving extended explanation) Verification of data transcript procedure was employed to address the issue of reliability.To establish this, the researchers carried out inter-rater reliability to refine and finalize data coding.The coded transcripts done by the rater were compared to the ones coded by the researchers.The level of agreement of coding for the information-gap transcript was 96%, 94% for the jigsaw transcript and 92% for the decision-making transcript.According to Miles and Huberman (1994), 90% is an acceptable level of coding reliability.Hence, based on Miles and Huberman's (1994) level of coding reliability, it can be established that the transcripts had been coded appropriately.

RESULTS
Data from this study revealed that all the different communication task types used generated negotiated interaction when the participants interacted to complete them as identified in their peer interactions.During their task engagement, they produced negotiation episodes and these episodes could be analysed in terms of negotiation strategies.It was observed that during task completion, the team members engaged in similar negotiation strategies.A qualitative analysis of the data revealed that even though the participants engaged in similar negotiation strategies during peer interaction, there were differences in terms of the depth of negotiation for the different communication task types.The following section provides detailed discussion.

EPISODES OF NEGOTIATION
When the participants experienced problem in understanding or breakdown in communication, they engaged in negotiation process to ensure shared and mutual understanding.As the participants negotiated, they used a variety of strategies such as asking for and giving explanations, asking for confirmations, asking for clarifications and asking for help.Excerpt 1 below shows the strategies used by the participants during the informationgap task completion.The information-gap task used was called 'How to Use a Nasal Spray.'For this task, one participant from each team was chosen to form an expert group.In the expert group, the chosen individuals from each team were given a set of instructions on how to use a nasal spray.They studied the instructions and discussed issues they did not understand.Having done those, they returned to their original teams.Meanwhile, the other participants in the teams were given a set of jumbled up pictures related to the instructions.Even though the participants in the teams were provided with the relevant information for task completion (i.e. the set of instructions and the set of jumbled up pictures), the one with the instructions held crucial information.The other participants in the team were required to request for information from him/her and he/she needed to supply it.Interaction was vital for the teams to complete this task.In their teams, the participant who had the information gave instructions to the others in his/her team on how to use a nasal spray.The others tried to rearrange the jumbled-up pictures so that they get the correct sequence.They were encouraged to ask for clarifications and explanations should they need more information so that they could arrange the pictures accurately.As shown in Excerpt 1, the participants argued over the picture they had.Man expressed his disagreement (Line 56, in bold); however, after Wan had given her explanation (Line 62, in bold), Man finally agreed.Even though there were negotiation strategies observed in this example, the participants did not engage in intensive negotiation.Many of those identified were directed to gathering information from the participants who held the information and ensuring comprehension for task completion.
When the participants engaged in task completion for the other two communication task types, jigsaw and decision-making, they also used similar negotiation strategies.However, when closely examined, it was found that the depth of the negotiated interaction episodes was different.For example, even though the participants used similar negotiation strategies (i.e.asking for and giving explanations, asking for confirmations, asking for clarifications and asking for help) when gathering information during jigsaw task completion, they also engaged in intensive negotiation as seen in Excerpt 2.
The jigsaw task type used was called 'Newspaper Jigsaw'.For this task, each participant in the teams was assigned to different letters (i.e.A, B, and C) and was instructed to form new groups based on the letters assigned to them.These groups were known as the expert groups.Each new expert group was given two paragraphs from the original article used in the task.In the expert groups, the participants discussed the paragraphs in terms of meaning of words and their contents.They were later instructed to paraphrase/summarize the paragraphs.Having done that, the participants returned to their original teams.In their own teams, the participants began their discussion.As each participant in the team had different portions of the article to complete the task, they needed to exchange the information with one another.They began their discussion by taking turns to request and supply the information to the rest of the participants in their teams.Their goal was to try to figure out what the article was about and to eventually produce a brief written summary of the article.Having one convergent goal and one acceptable outcome agreed by all team members, they needed to interact by requesting and supplying information to one another to achieve that goal.
Excerpt 2 (from jigsaw task type interaction) shows how Jani began by providing the participants in her team with information they needed.The other two participants, Hartini and Mohd, constantly stopped her to ask for more explanation and confirmation to ensure that they understood accurately the information given by Jani.Jani patiently responded and provided the information and explanation required.It was interesting to note that to explain further to her team members, Jani rephrased her explanation, clarified and provided examples 89,91,93,96,in bold).The modifications made gave evidence of joint meaning-making among the participants.It appeared that they had used each other's knowledge to solve the problems they faced and intensive meaning negotiation episode could be observed.To sum up, in spite of these findings, it is reasonable to conclude that using the different communication task types could elicit episodes of negotiation from the participants.However, the distinctive characteristics and nature of the tasks may have contributed to the way the participants interacted with one another during task completion and the kind of negotiated interaction generated by the participants as seen in their peer interactions.

DISCUSSION
In order to examine the negotiation strategies in the participants' oral interaction as they engaged in the different communication task types, this study employed a theory-driven approach using the Interaction Hypothesis (Long, 1983), as the theory that underpin the study.In terms of the production of negotiation episodes, it was observed that all the communication task types used in this study generated negotiation episodes.Episodes of confirmation checks and clarification requests as well as asking for help, asking for and providing explanation, and arguing and expressing disagreement were identified in the participants' peer interactions.However, when closely examined, even though the participants used similar negotiation strategies when engaged in negotiated interactions, the depth of the interactions differed from one task type to the other.
For example, for the information-gap task type, the focus of their negotiation was on task management.Thus, many of the negotiation episodes were directed towards gathering information from the participant who held it.The episodes were direct to the point and did not involve intensive negotiation.In contrast, even though many of the incidences of negotiation identified in the jigsaw peer interactions were directed to gathering information, there were some instances of real negotiation involved.This was when the participants engaged in meaning-making activities during which they sought for explanation and confirmation from other participants.
When compared to the decision-making task type, it was observed that the participants used similar negotiation strategies.However, their peer interactions were very much characterised by active discussions and the negotiation episodes were very intensive.Episode of intensive negotiation could be observed when the participants engaged in active discussion.Further, when providing responses to their team members, they took the effort to provide extended explanation, give justifications and share opinions.
Based on the results, interestingly findings from this study conflicted with Pica's et al. (1993) typology of communication task types which proposed that the information-gap and jigsaw tasks were the task types that would encourage oral interaction among L2 speakers (p.21).Instead of the information-gap and jigsaw tasks, in the present study, it was observed that the decision-making task type encouraged oral interaction and elicited intensive negotiation episodes among the participants.The different characteristics and nature of the tasks may have contributed to the way in which the participants interacted with one another during task completion and the kind of interaction generated by the participants as seen in their learner interactions.
The findings from the current study however concurred with those conducted by Lan et al. (2016) in terms of the kind of interaction learners produced when engaged in different task types in Second Life (SL).SL is one of the most popular multiuser virtual environments which combines network connectivity and virtual reality.This combination provides users with a virtual yet real environment in which they can interact socially with others via resident avatars (James, 2005).Lan et al. (2016) investigated whether certain task types in SL can elicit more intra-group interaction than the others, focusing on information-gap and reasoning-gap task.They found that both task types encouraged interaction among learners.However, it was interesting to note that for the information gap task type, the participants "solely focused on exchanging the collected information with their team members, piece by piece, without making further efforts on integrating what they had observed" (p.68).This may indicate that the information-gap task elicited interactions which were straightforward and did not contain much negotiation.In contrast, with the reasoning-gap task type, the participants "figured out the answers to the problems through inter-peer informationexchange and meaning negotiation" (p.69).
Findings from this current study also share similar results with a study conducted by Fondo Garcia and Appel (2016) in terms of the influence of task types on learners' performance.Fondo, Garcia and Appel (2016) used four different tasks with eight tasks in total, to investigate "how, and to what extent different task designs influenced the performance of the participants in terms of communication strategies, length of conversation and immediate feedback provided during on-task time" (p.145).The tasks they used were taken from Pica et al. (1993) typology of communication task types; the information-gap, problem solving and opinion exchange.Results from their study found that the different task types had different effect on learners' performance.These concur with findings from the current study.
To conclude, despite the findings obtained in this study, to claim that one task is better than the other in promoting interaction among learners is rather premature.The results in this study have to be interpreted cautiously due to the small size of the sample and the variety of tasks.However, they do have practical implications for language teaching, syllabus design and research.

CONCLUSION
Theoretically, this study expands the notion of interaction as presented earlier to include exploratory interaction as it is believed that when examining learner interaction, it is insufficient to only look at the number of instances of requests or confirmation checks.It is equally important that researchers also look at interaction in a broader sense, for example, how learners interact with one another, how the interaction episodes assist learners in their language learning.It is believed that these will provide insights into the dynamics of the discourse and learning opportunities they may present (Nakahama et al., 2001).
Additionally, understanding the dynamics of learner interactions during different task completion has crucial pedagogical implication.This knowledge can be used as a basis for informed pedagogical practice in the L2 classroom particularly in an EFL context.Language instructors need to seriously consider the lesson objectives when choosing or designing language tasks to be used in L2 classrooms.Tasks should be able to elicit the kind of interaction episodes deemed meaningful and useful to L2 learners.Based on the results of the present study, it was observed that the decision-making task type was the type that encouraged the production of complex patterns of interactions and the generation of more complex ideas as found in episodes of negotiated interaction.The participants could generate the kind of interaction episodes believed to be facilitative of their language learning.This was evident in the interaction episodes when they engaged in the decision-making task completion.Thus, pedagogically, the use of decision-making task type in EFL tertiary classrooms should be encouraged as it is believed that it encourages learner interaction that would support learners' language learning.
Based on Pica's et al. (1993) typology of task characteristics, communication tasks can be placed under five different task types -jigsaw, information-gap, problem-solving, decision-making and opinion-exchange task types.The tasks used in this study were the information-gap, jigsaw and decision-making.As illustrated in the results of the study, different task types encouraged different kinds of learner interactions.Thus, it is believed that by examining learner interactions of all the different task types during task performance and completion, a more comprehensive and complete set of data on the learners' oral discourse could be obtained.Data could be analyzed in order to understand the influence that other task types may have on the learners' oral discourse when engaged in task completion, the language learning opportunities the other task types may create during task completion and the value of other communication task types as a source for possible restructuring of interlanguage which may lead to uptake of language input in an EFL tertiary setting.These findings may assist practitioners in planning tasks that may best suit their learners based on the objectives and aims intended for the learners.Basturkmen (2002) argued that complex patterns of interactions found in episodes of negotiation for meaning are "important in enabling students to develop their own ideas in discussion" (p.233).These patterns generate more complex ideas to emerge and to be negotiated in interaction.When engaged in these complex patterns of interactions, learners are also able to articulate thoughts and clarify thinking more clearly.It is these kinds of interactions that are important particularly for L2 learners at the tertiary level (Watanabe, & Swain, 2007).Therefore, knowledge regarding the kinds of task type that could be used not only to promote language acquisition but also to enhance learning in general, is important.Such knowledge will assist learners in language acquisition using suitable task types for learners' learning to be enhanced.As emphasised by Long (2017), "if the tasks are carefully designed, intellectually challenging and fun, several advantages follow" (p.4).

ACKNOWLEDGEMENT
Findings highlighted in this paper were from a larger study financially funded by the Ministry of Higher Education, Malaysia.We appreciate PPPi, USIM under Skim Pembiayaan Yuran Penerbitan Jurnal.

TABLE 1 .
Communication Task Types for L2 Research and Pedagogy Analysis Based on Interactant (X/Y) Relationship Requirements in Communicating Information (INF) to Achieve Task Goals

TABLE 2 .
Duration of Task Completion

TABLE 3 .
Description of Categories CATEGORIES DESCRIPTION Asking for confirmationAsking for confirmation is a confirmation check that occurs when a participant wants to make sure that what s/he understands what the other means.F: This is over fishing, yes? (ask for confirmation) A: Over fishing?Ha, yes.F: Rmmmm...ok ok.

think _ er_ buntal fish! _ That it has _ er _ what we call… F: But ni (this) _ er _ the yellow and oren (orange)_ it get very popular in Japan _ er _ (expressing disagreement)
An expression of disagreement is an attempt by a participant to express disagreement by providing justifications, explanations or opinions.It is one way of negotiating meaning that the participants employ during their learner interactions.
S: I think, I