Indonesian EFL Journal (IEFLJ) Volume 6, Issue 2, July 2020 p-ISSN 2252-7427, e-ISSN 2541-3635 https://journal.uniku.ac.id/index.php/IEFLJ/index 109 WHAT DO PEOPLE DO TO SUSTAIN A CONVERSATION: ANALYSIS OF THE FEATURES AND THE SPEECH FUNCTION OF A CASUAL CONVERSATION Latifa Ika Sari Department of Port and Shipping, Politeknik Ilmu Pelayaran Semarang, Indonesia Email: latifa.ika@gmail.com Ria Hermina Sari Department of Port and Shipping, Politeknik Ilmu Pelayaran Semarang, Indonesia E-mail: ria.hermina.sari@gmail.com APA Citation: Sari, L. I., & Sari, R. H. (2020). What do people do to sustain a conversation: Analysis of the features and the speech function of a casual conversation. Indonesian EFL Journal, 6(2), 109- 120. doi: 10.25134/ieflj.v6i2.3379. Received: 11-03-2020 Accepted: 24-04-2020 Published: 01-07-2020 Abstract: Conversation Analysis (CA) has a great implication in the second or foreign language teaching and learning. This study was aimed to analyse a casual conversation, to identify what people do to sustain a conversation. The conversation, which lasted for 20 minutes and involved three speakers, was transcribed verbatim. Several features of the spoken text are analysed: spontaneity, interactivity, interpersonal features, coherence features, negotiation of meaning, and speech function. The result of the analysis showed that there are several strategies used by the speakers: time-gaining strategies (filled pauses; frequent use of conjunctions: and, but, so); using chunks and producing one clause or phrase in small ‗runs‘; self-monitoring strategies (repetition, backtracking), and interactional strategies (backchanneling, showing amusement by laughing or chuckling, using certain discourse markers, hedges, vague language, showing empathy by completing and repeating each other‘s utterances). The speakers also negotiate interpersonally and logico- semantically to keep the conversation going on. The equal number of rejoinders that each speaker produces indicates that they are willing to support each other to sustain the conversation. This study implies that when teaching speaking, English teachers should include communication strategies to achieve the goals of communication. Keywords: casual conversation; negotiation of meaning; speech function; strategies. INTRODUCTION Language is considered as an important tool for communication. Communication itself is seen as a transactional process of creating meaning which requires interaction of at least 2 or more individuals (Kpogo & Abrefa, 2017). The communication can be realized using spoken or written text. Spoken text – even though messy – is more than just a collection of random utterances. All utterances are meaningful and have certain purposes. As social creatures, humans spend much of their time to talk and to interact with others. The talking activity and the interaction have meaningful purposes: to find out information, to transfer knowledge, to make appointments, to offer something, to buy and sell goods or services, etc (Eggins & Slade, 1997). Further, Cornish (2006) describes that spoken language or speech typically involves ―face-to- face interaction between two or more participants who share a spatio-temporal environment‖. Spoken language has different characteristics from written language in many aspects. There are actually no clear-cut dividing points, instead the differences are shown in the mode of a continuum. Written language is considered as language as representation, while spoken language is considered as language accompanying action (Verhoeven, 1994). Since spoken language is the language accompanying action, it has special characteristics (Thornburry, 2005). First, it is dynamic in structure which is shown by the presence of interactive staging and open-ended conversations. Second, different from written language, spoken language is often unplanned and unpredictable (Dahal, 2010).Anything can happen during the conversation, and that makes spoken language spontaneous. False starts, hesitations, and repetitions are some of the evidences of spontaneity. Third, spoken language is context-dependent. It means that someone cannot understand well a spoken text without knowing the context. It is mailto:latifa.ika@gmail.com Latifa Ika Sari & Ria Hermina Sari What do people do to sustain a conversation: Analysis of the features and the speech function of a casual conversation 110 shown by the use of deictic expressions which can only be understood by referring to the speaker‘s intention. The last one, while written language employs monological organization, spoken language has turn-taking organization (Halliday, 1989). Speakers take turn in order to sustain the conversation together. During a conversation, individuals have interactions through talk. Describing, analyzing, and understanding talk as a basic and constitutive feature of human social life is called as conversation analysis (Sert & Seedhouse, 2011). In conversation analysis (CA) talk is seen as a vehicle for action. Participants engage in a talk not only simply to transfer information, but they also concern with the actions getting done through talk (e.g., asking, requesting, complaining, noticing, and so on), and the real- life consequences of those actions (Hoey & Kendrick, 2018). Conversation analysis (CA) is a method of ―gathering data involving naturalistic conversational interaction, analyzing it systematically, and reporting on features of its structural organization‖ (Albert, 2017). CA involves ―identification and description of interactional phenomena through naturalistic observation‖ (Kendrick, 2017). CA has a great implication in second or foreign language teaching and learning because it reflects the ‗social organization of natural language-in-use‘. Since social interaction becomes the core of human activities, analysing a conversation can be a great source for learning a second or foreign language (Barraja-Rohan, 2011). Research on conversation analysis has developed greatly in many fields. In recent years, the significance of CA has grown significantly that it is used in other fields than linguistics. Fasulo and Sterponi (2016) used CA to analyse children‘s mental health conditions in their interactional environment. Buchholz and Kächele (2017) found CA was beneficial for psychotherapeutic talk during psychotherapy. CA is able to describe moment-by-moment evolution of talk that is organized as sequences of actions (Peräkylä, 2019) and therefore can be a powerful tool for psychoanalytic practice and psychotherapy research (Buchholz & Kächele, 2013). In education setting, CA can be used to investigate classroom interaction between teachers and students to identify pedagogical teacher problems (Akmaliyah, 2014; Koole, 2013). By using CA, teachers or researchers can address issues in classroom interactions and environment by studying teacher-student interaction and student-student interaction. As Takeda (2013) found that conducting CA will be beneficial for English language teachers to understand their students and interactional issues in the classroom. His finding was supported by Hale et al. (2018) who found that CA can provide teachers with a powerful analytic lens to view language use in their classrooms—both their own language use, and that of their students—in order to make pedagogical changes that can enhance learning. Hidayat (2019) analyzed the use of CA in casual conversation and how it can serve as a potential means in language teaching. His findings showed that turn taking system, adjacency pairs, overlaps, response tokens and repairs were evident in the conversation. Casual conversation is guided by interpersonal objectives in which people aim to expand or maintain interpersonal relations by having a casual talk with other people. Casual conversation covers many ranges of topics which may include daily activities or experiences. Unlike transactional conversation, casual conversation is more unpredictable. Therefore, it can be more difficult for language learners to handle. Chan (2019) investigated the language and communication needs of business professionals in Hongkong and found that the informants face difficulties when handling non- technical genres such as replying emails and socializing because they are harder to predict and therefore harder to prepare. It is different from technical genres which tend to make use of a finite set of technical terms and technical concepts. When having casual conversation, it becomes a challenge for individuals to sustain the conversation to reach the goals of communication. Besides technical knowledge, people also need skills to maintain good interpersonal relations and deal with interpersonal tensions in handling workplace genres. However, Vo et al. (2016) found that the classroom materials which supported relational talk at university were not available. This paper aims to analyse a casual conversation to explore what people do to sustain a conversation. There are several features of the casual conversation that will be analysed including spontaneity, interactivity, interpersonal features, coherence features, negotiation of meaning, and speech function. There are several pedagogical implications which make this study Indonesian EFL Journal (IEFLJ) Volume 6, Issue 2, July 2020 p-ISSN 2252-7427, e-ISSN 2541-3635 https://journal.uniku.ac.id/index.php/IEFLJ/index 111 is important to be conducted. Firstly, understanding the features of casual conversation will enable English language teachers to teach speaking better. In speaking, beside knowledge about vocabulary, grammar, or pronunciation, students also need knowledge and skills to sustain conversation. Secondly, by learning the features of casual conversation, speakers can be more aware with their utterances and thus have more control towards their utterances. For example, how to show disagreement, how to establish solidarity during a conversation, or how to make the conversation becomes more lively and friendly. Thirdly, by understanding the feature of casual conversation, speakers can also learn several communication strategies to reach their communication goals. METHOD Participants There were three speakers involved in the conversation (Henceforth S1, S2, and S3). They are English lecturers working in one of Polytechnic in Semarang. S1 is the Head of the Language Unit. She is 38 years old. S2 is the staff of the Language Unit. She is 33 years old. S3 is a non-permanent lecturer, and she is 27 years old. Data collection The conversation was 20 minutes long and was recorded in a taxi when the speakers had their journey to the Airport. They had attended a workshop in Jakarta and wanted to go back to Semarang. The conversation was conducted as naturally as possible. At some points of the recording, there were sounds of ambulance and horn of cars that interfered with the conversation. Data analysis There are several stages conducted by the researcher in analysing data in this study. First, the researcher transcribed the conversation verbatim. Verbatim means that the transcription is made word for word, exactly as said by the participants (Trippas et al., 2017). The transcription of the conversation follows the framework from Eggins and Slade (1997). The symbols used in the transcriptions are summarized as follow: 1. Turn numbers are shown in Arabic numerals: 1, 2, 3. 2. Clause numbers are shown in lower case Roman numerals: i, ii, iii. 3. Move numbers are shown in lower case letters: a, b, c. 4. NV indicates non-verbal moves. 5. Other symbols: a. == indicates overlap of utterance b. … indicates short hesitation within a turn (less than 3 seconds) c. [pause – 4 secs] indicates pause length d. Dash – then talk indicates false start / restart e. [words in square bracket] indicates non- verbal information. After finishing the transcription, the researcher read the conversation several times and identified the features of the spoken text which include spontaneity, interactivity, interpersonal features, coherence features, negotiation of meaning, and speech function. Tables and diagrams are used to display the data to allow the readers to understand the findings easier. RESULTS AND DISCUSSION Turns, moves and clauses Each speaker produces different number of turns, moves and clauses. S1 produces 117 turns, 143 moves and 198 clauses. S2 produces 81 turns, 91 moves and 83 clauses. While S3 produces 94 turns, 103 moves and 157 clauses. The turns, moves and clauses are summarized in Table 1. Table 1. The number of turns, moves and clauses of each speakers Speaker Turns Moves Clauses S1 117 (40%) 143 (42%) 198 (45%) S2 81 (28%) 91 (27%) 83 (19%) S3 94 (32%) 103 (31%) 157 (36%) Total 292 337 438 Evidence of spontaneity Spontaneity is one of the characteristics of spoken language. It is indicated by filled pauses, repetitions, incomplete utterances, false start and backtracking, frequent use of conjunctions ‗and, but, so‘, and chunks (Thornburry, 2005). In spoken language, speakers produce one clause or one phrase at a time as small ‗runs‘. Different from written text, the clauses or phrases in the spoken text are not embedded in large unit. Table 2 describes the examples of evidence of spontaneity found in the conversation. Latifa Ika Sari & Ria Hermina Sari What do people do to sustain a conversation: Analysis of the features and the speech function of a casual conversation 112 From the shreds of evidence presented in Table 2, we can see that the conversation has the feature of spontaneity. Spoken language is messy because people do not have much time to think and re-think their utterances. Especially in a casual conversation, there is no topic preparation at all. People just say what is on their minds and give responses in a very short time. People can hardly anticipate what will happen in the conversation. Everything just happens spontaneously. The pauses and the repetitions are actually part of strategic competence that are used by people when they communicate (Celce- Murcia, 2007). The pauses give the speakers time to think about what they are going to say. The repetitions also function the same way. They belong to stalling or time-gaining strategies. The strategies enable the speakers to engage in the conversation and give responses appropriately. False start and backtracking prove that the speakers use self-monitoring strategies. In Turn 71, Speaker 3 says ―I see -- I saw that a lot of people also focus on the screen and yaa they will take a lot of time for doing so‖. At the beginning, she said ―I see‖, but after she realized that she should use past tense instead, she corrected her utterance. If the speakers are given time and opportunity to correct their utterances before saying them, they will surely do that. However, it is not possible since conversation happens in a spontaneous way. Therefore, people use chunks to help them talk fluently. Chunks are ―multi- word units that behave as if they were single words and typically consist of formulaic routines that are stored and retrieved in their entirety‖ (Thornbury, 2005, p. 65). ―One kind of the text, a lot of things, afraid of something‖, are some of chunks found in the conversation. The use of chunks will ease the speakers‘ burden to memorize grammar rules and can enhance the speakers‘ fluency which is part of formulaic competence (Celce-Murcia, 2007). In a conversation, people also tend to use one clause or one phrase at a time. Unlike in writing that uses complex sentences, in spoken text, people use clauses or phrases to make them easier to convey their meaning. In turn 139, Speaker 3 says ―Her family… something like that inside of ambulance, the sounds, while you are worrying about your family…she just…you know…close her ears…‖. What she wanted to say was ―she has a friend who is afraid of the ambulance sirens and always close her ears every time she hears the sound because she has a bad experience of her family that was ever carried by using ambulance‖. Cutting off long utterances into short ones is also part of strategic competence to make people understand more about the topic that is being discussed. Table 2. Evidence of spontaneity No. Evidence of Spontaneity Turn Speaker Utterance 1. Filled Pauses 1/b S1 (ii) I‘m I‘m ... I‘m actually making eee what it‘s called… aaa a CONFESSION (iii) that I do impulsive buying on Saturday…(iv) buy a larger speaker 2. Repetition 46 S1 (i) actually I... I didn‘t re regret of buying that thing (ii) but I... I regret that my phone is brokens (iii) That‘s why I cannot afford eee to buy new laptop 3. False start and backtracking 71/a S3 (i) I see -- I saw that a lot of people also focus on the screen (ii) and yaa they will take a lot of time for doing (iii) so , so, so, it just .... 99 S3 (i) One of the mmm class -- one of the my class -- my student notice me (ii) when I was so worry… 4. Incomplete utterance 111 S1 (i) Maybe these students are .....what it‘s called… eee having.... 5. Frequent use of conjunctions: and, but, so 178 S1 (i) And the problem is if we are tend not to update our…what it‘s called …knowledge… well (ii) we will be teaching the same thing for years, and last might be eee… a bit… what it‘s called… 180/b S1 (ii) So, we might not deliver it correctly according to the...what it‘s called… the updated knowledge that the students should have. (iii) So, we tend to ignore eee several important things, ...... (iv) eee yeah, we will be the bad teacher, I think. (v) If we never learn more and more. (vi) Indonesian EFL Journal (IEFLJ) Volume 6, Issue 2, July 2020 p-ISSN 2252-7427, e-ISSN 2541-3635 https://journal.uniku.ac.id/index.php/IEFLJ/index 113 So, learning is important, even for the teachers. 6. Chunks 169 S3 (i) I mean like no one kind of the text, one text. (ii) So we have to prepare a lot of thing like the talks, about the questions, (iii) and that‘s only one, (iv) and… then we jump to another text, (v) so we are...== 7. One clause or one phrase at a time produces small ‗runs‘ 139 S3 (i) Her family… something like that inside of ambulance, the sounds, while (ii) you are worrying about your family…(iii) she just…you know…close her ears… Evidence of interactivity Conversation is interactive. It means that it involves speakers‘ interaction. To make the conversation run smoothly, all speakers should interact cooperatively. Interactivity can be seen from the following aspects: turn-taking, interruptions or overlapping utterances, signals of amusement (grunts, laughs and chuckles), backchanneling, and the use of discourse markers such as well, yeah, but, you know, etc (Thornburry, 2005). Turn-taking and interruptions are clearly seen in the transcript of the conversation. The examples can be seen in Table 3. Table 3. Examples of turn-taking and interruptions Turn Speaker Utterance 1/a S1 (i) And you know what? 1/b S1 (ii) I‘m I‘m ... I‘m actually making eee what it‘s called… aaa a CONFESSION (iii) that I do impulsive buying on Saturday…(iv) buy a larger speakers 1/c S1 (v) that we can use for… the test.. [laugh] 2/a S2 (i) Larger speakers? 2/b S2 (ii) How about the small one? 3 S1 (i) Because eee the small speakers… eee you know, (ii) it get less power than the bigger one… and (iii) this bigger speakers -- it has eee a more… what it‘s called… a more clear ... 4 S3 (i) ==Sound? 5/a S1 (i) Sound 5/b S1 (ii) ya so i choose the larger one, (ii) so that we can eee use it for the…the TOEIC, Marlins Test..and things like that, so…== 6 S3 (i) == You said that your computer can not connect to the speakers? 7 S1 (i) ==Yes…Yeah 8 S3 (i) Another speakers ... (ii) Suck the red one.. 9 S1 [laugh] 10 S2 (i) Actually, i wanted to to buy the same like yours 11 S1 (i) Mmm, mhm 12 S2 (i) The orange one The transcript shows that S1, S2 and S3 interact by taking turns to speak. Turn taking in conversation is an important feature of human interaction. When having a conversation in everyday situations, interlocutors efficiently align their turns-of-talk (Barthel et al., 2017). The distribution of opportunities for talk is very essentials to maintain the conversation (De Jaegher et al., 2016). However, there are times that a speaker interrupts another speaker. For example, in Turn 5 and Turn 6 when S1 says ―Sound, ya so I choose the larger one, so that we can eee use it for the…the TOEIC, Marlins Test..and things like that, so…‖. Before she manages to finish her sentence, S3 interrupts by saying ―You said that your computer can not connect to the speakers?‖. The interruptions and overlapping turns are shown by the sign ==. Laughs as the signal of amusement are found many times in the conversation. From the total 337 moves, there are 22 moves that involve laughing. In a casual conversation in which people have a close interpersonal relationship, the conversation will run more interactively. The speakers feel more freedom to express their feeling by laughing, chuckling, etc. This is also an indication that the power and status (tenor) of each participant are considered as ―balanced‖ during the conversation. Backchanneling is also another interactive device in which the listeners do in order to Latifa Ika Sari & Ria Hermina Sari What do people do to sustain a conversation: Analysis of the features and the speech function of a casual conversation 114 register that they are following the speaker‘s drift (Thornburry, 2005). An expression such as ―Mmm, mhm‖ is used as a sign that the listener pays attention to the speaker. The examples of backchanneling are presented in Table 4. In the conversation, S1 shows her attention towards S2 by uttering ―mmm, mhm‖ when S2 says that she wants to buy the same sound speaker. Backchanneling is important in conversation because it gives a speaker motivation to continue his/her utterance. It is like another way of saying ―Go on with your utterance, I am listening‖. In this conversation, there are 29 moves which use backchanneling. The use of discourse markers such as well, yeah, but, you know, etc. are also important to make the conversation more interactive. Table 4. Examples of Backchanneling Turn Speaker Utterance 10 S2 (i) Actually, i wanted to to buy the same like yours 11 S1 (i) Mmm, mhm 12 S2 (i) The orange one 13 S1 (i) Mmm, mhm Table 5. Example of discourse markers Turn Speaker Utterance 43 S2 (i) ==But... but i think if it is useful for your class, (ii) then... it‘s worth the price== 44 S1 (i) ==Yaaa…== 45 S2 (i) ==isn‘t it? 46 S1 (i) actually I... I didn‘t re regret of buying that thing (ii) but I... I regret that my phone is brokens (iii) That‘s why I cannot afford eee to buy new laptop 47 S3 (i) Because of your phone, (ii) not because of the speakers== 48/a S1 (i) ==[laugh] yeah, 48/b S1 (ii) because of my phone, (iii) I hate that‘s 49 S3 (i) Yeah... Another evidence of interactivity is the use of discourse markers that can be seen in Table 5. Discourse marker ―but‖ shows that S2 responds to S1‘s story that she has just bought a new sound speaker. Since S1 seems to feel guilty because of spending Rp. 2,3 million just to buy a new sound speaker, S2 then signals her contrast opinion by saying ―but‖. In Turn 49, S1 says ―yeah‖ as an agreement response to what has just been said by S3. Discourse markers in a conversation have several functions. They can be the signposts of shifts and turns in an on-going conversation. They also have a connecting function which connect utterances within speakers and across them (Thornburry, 2005). Using discourse markers is said as ―an interactive process that requires speakers to draw upon several different types of communicative knowledge – cognitive, expressive, social, and textual‖ (Schiffrin, 2001). Those evidences show that the conversation has the feature of interactivity which belongs to the interactional competence (Celce-Murcia, 2007). People use strategies such as turn-taking, backchanneling, showing their amusement by laughing or chuckling, and using certain discourse markers to keep the conversation ―alive‖ and interactive. Interpersonal features ―Conversation is not simply the exchange of information but has a strong interpersonal function‖ (Thornbury, 2005, p. 66). People give support to each other when having a conversation and maintain group solidarity. Even if they disagree over something, they will use certain expressions to blunt the force of a disagreement. People use hedges, vague language, showing empathy by completing and repeating each other‘s utterances, and use certain markers such as ―you know‖, ―don‘t you‖, etc. The examples of people completing and repeating each other‘s utterance are presented in Table 6. Table 6 shows several examples of completing and repeating each other‘s utterances as evidence of interpersonal features. In turn 34, S3 completes S1‘s utterance when she finds difficulty in saying her words. S1 then replies by repeating the word ―conversation‖ that has been mentioned by S3. In Turn 60, all speakers repeat the word ―happiness‖ which has been mentioned Indonesian EFL Journal (IEFLJ) Volume 6, Issue 2, July 2020 p-ISSN 2252-7427, e-ISSN 2541-3635 https://journal.uniku.ac.id/index.php/IEFLJ/index 115 by S1. Completing and repeating each other‘s utterances demonstrate empathy between speakers and can function to establish and maintain group solidarity between them. Hedges, laughs, and the use of question tags or special markers are also part of interpersonal features. Table 7 shows the examples of hedges, laughs, the use of question tags or special markers, and exaggerative language in the conversation. In a conversation, people use hedges and laughs to blunt their language so that they will not sound too opinionated. ―Negative responses which reject or decline the initiating action also tend to be delayed in time and may occur with turn-initial particles like ―uh‖ or ―well‖, and include accounts for the negative response (e.g., uh no, I‘m a bit tired actually)‖ (Bögels et al., 2015). Table 6. Example of completing and repeating other’s utterance No. Interpersonal features Turn Speaker Utterance 1. completing other‘s utterance 31/b S1 (vi) You know... 31/c S1 (vii) in a class when you‘re have the eee small discussion and (vii) then I want to play like music to distract their attention or to … what it‘s called… 32 S2 (i) ==Mmm, Mhm 33 S1 (i) ==Cut their…== 34 S3 (i) ==conversation? 35/a S1 (i) Conversation, yes, 2. repeating other‘s utterance 59 S2 (i) It just -- (ii) i think it just postpone your… eee, (iii) what is your... 60 S1 (i) Happiness 61 S1, S2, S3 (i) Happiness [laugh] S3 in turn 126/b says ―I don‘t know‖ in order to blunt her disagreement towards S1‘s utterance. In Turn 119, Speaker 1 tells S3 not to do something again. She then laughs after saying her utterance. S3 also laughs after hearing S1 and says ―Yeah. That‘s funny..‖. Questions tags are also often used in a conversation. Special markers such as ―you know‖, ―don‘t you…‖ are also widely used. S2 in turn 24 says ―Don‘t you think?‖ as a clarification and agreement seeking. She asks S1‘s agreement that the sound speaker that she wants to buy is portable and useful. In a conversation, speakers sometimes also exaggerate their language to make the conversation lively. In turn 271, S1 says that the reading session kills you. This is a kind of exaggerating utterance because the reading session cannot literally kill a human. However, S1 says that to ―light up‖ the conversation to make the topic more interesting. That is also one of strategy to make the conversation keep going. Table 7. Examples of hedges, laughs, the use of question tags or special markers, and the use of exaggerative language No. Interpersonal features Turn Speaker Utterance 1. Hedges and laughs 119 S1 (i) Don‘t do that again! [laugh] 120 S3 (i) [laugh] Yeah. That‘s funny.. 121 S1, S2, S3 [laugh] 122/a S1 (i) Yes, always funny, 122/b S1 (ii) the students are funny…(iii) and at the same time, they are annoying. 123 S3 [laugh] 124 S2 (i) Yeah, funny and annoying 125/a S1 [laugh] (i) Annoying... 125/b S1 (ii) but you cannot angry... with them [laugh again] 126/a S3 (i) That‘s the thing, yes... 126/b S3 (ii) I don’t know… 127 S1 [laugh] 2. Question tag or 21 S3 (i) It is portable? Latifa Ika Sari & Ria Hermina Sari What do people do to sustain a conversation: Analysis of the features and the speech function of a casual conversation 116 special markers 22/a S2 (i) Yeah, portable 22/b S2 (ii) and ...... very mmm… useful 23 S1 (i) Mmm mmm 24 S2 (i) ==Don’t you think? 25/a S1 (i) Yes yes, it‘s very practical. 3. Exaggeration 265/a S2 (i) Yeah.. 265/b S2 (ii) the level of difficulty…== 266 S1 (i) ==Yes 267 S2 (i) I think it‘s higher 268 S1 (i) Yes... higher 269 S2 (i) Than... the previous one 270 S3 (i) Hmm hmm... correctly 271 S1 (i) Yes... the reading... session is... kills you [laugh] 272 S2 [laugh] Coherence features Coherence refers to how the text makes sense. Different from the written language which depends solely on the writer, coherence in spoken language depends on the cooperation of the speakers who involve in the conversation. Speakers co-operate so that what they say is relevant to what has been said before and to the overall purpose of the talk (Thornburry, 2005). Coherence will sustain a conversation and prevent communication breakdown. The example of coherence features found in the conversation can be seen in Table 8. As we can see in the excerpt of the conversation, in Turn 158 S1 asks about S2 and S3‘s opinion about the workshop. S2 and S3 give relevant responses by answering ―great‖ and ―it makes me think a lot‖. Their responses are relevant and help the conversation keep going on. S1 then replies by saying ―a lots and lots of thinking‖. S3 then replies again by mentioning about preparation. Their responses are relevant to the previous turn. The conversation unfolds because the speakers cooperate to keep it going on to prevent the communication breakdown. Table 8. Example of coherence features Turn Speaker Utterance 158 S1 (i) So, about the workshop.... (ii) What do you think about the workshop? 159 S2, S3 (i) Mmmmmmmm 160 S2 (i) Great 161 S3 (i) It makes me think a lot 162 S2 (i) Hmm hmm 163 S1, S2, S3 [Laugh] 164 S1 (i) A lots and lots of thinking, 165 S2 (i) Yeah 166 S1 (i) and trying to figure out whether it’s applicable in our institution or not, (ii) will the students be able to keep up with that or not. (iii) So many questions about== 167 S3 (i) ==The thing is that one more thing as…(ii) the biggest thought is about preparation 168 S1 (i) Yes 169 S3 (i) I mean like no one kind of the text, one text. (ii) So we have to prepare a lot of thing like the talks, about the questions, (iii) and that‘s only one, (iv) and… then we jump to another text, (v) so we are...== 170 S2 (i) ==We need a preparation… 171 S3 (i) But I believe, by time, eemm going, by experience, (ii) we as a teacher will be easy to do so. (iii) We don‘t need to prepare a lot (iv) but at first, of course it is..== Negotiations of meaning ―Negotiation is concerned with interaction as an exchange between speakers: how speakers adopt and assign roles to each other in dialogue and how moves are organized in relation to one another‖ (Martin & Rose, 2007, p.221). There are Indonesian EFL Journal (IEFLJ) Volume 6, Issue 2, July 2020 p-ISSN 2252-7427, e-ISSN 2541-3635 https://journal.uniku.ac.id/index.php/IEFLJ/index 117 two kinds of negotiation: Interpersonal negotiation which negotiates feelings and attitude and Logico-semantic negotiation which negotiates message or news in the conversation. In this conversation, there are 160 moves which use interpersonal negotiation, 126 moves which use Logico-semantic negotiation, and 51 moves which have no negotiation of meaning (only consist of backchanneling and laughing). The negotiation of meaning is summarized in Table 9. Table 9. Negotiation of meaning Negotiation of meaning Occurrence Interpersonal negotiation 160 (48%) Logico-semantic negotiation 126 (37%) No negotiation (backchanneling and laughing) 51 (15%) From Table 9, it can be seen that the speakers negotiate meaning both interpersonally and logico-semantically. It means that they negotiate message or news as well as negotiate feelings or attitude. However, the interpersonal negotiation happens more often than logico-semantic negotiation which indicates that this is a casual conversation in which the speakers have a close interpersonal relationship. The logico-semantic negotiation that happens in this conversation is highly influenced by the context and setting of the conversation as well as the educational background of the speakers. All speakers are English lecturers and they had just attended a workshop about English Teaching and Learning (ELT). That makes the topic of their conversation is mostly about English teaching and learning mixed with personal matters. Speech function analysis As Halliday points out (in Eggins & Slade, 1997, p. 180) when people use language to interact, ―one of the things they are doing is establishing a relationship, between the person speaking now and the person who will probably speak next‖. Halliday (in Eggins & Slade, 1997, p. 180) further elaborates that dialogue is ―a process of exchange‖ which involves two variables: a commodity to be exchanged (either information or goods and services) and roles associated with exchange relations (either giving or demanding). Eggins and Slade (1997) then develop Halliday‘s basic speech function into more detailed speech function classes which are used in this paper to analyse the conversation. The summary of the speech function is described in the Table 10. From Table 10, it can be seen that S1 is the most dominant speaker. She produces 117 turns, 143 moves, and 198 clauses. While the most passive speaker is S2 that produces only 81 turns, 91 moves, and 83 clauses. S1 also has the initiative to open the conversation by attending other speakers, giving questions and asking opinions while S2 and S3 just initiate new topics occasionally during the conversation. It is interesting to note that S3 produces continuing moves more often than other speakers. She elaborates her utterance to make the audiences understand better. Even though her utterances are interfered by other speakers, she tries to get back and gives elaboration to her previous utterances. S2 who plays a quite passive role throughout the conversation is actually not that passive. She gives responses quite often (50 moves) that mostly consist of registering and agreeing. It seems that she just wants to follow the conversation without making any debate or confrontation. Even though she makes 2 confronting moves that she denies acknowledgement of certain information, both moves are aimed to provide a different point of view about something. Table 10. The summary of speech function 1 Speech function S1 S2 S3 The speech function No. of turns 117 81 94 No. of moves 143 91 103 No. of clauses 198 83 157 Non Verbal (Laugh) 19 13 8 Open Attending 3 - - Command 1 - 1 Question: Open: Opinion 2 - - Question: Closed: Fact 1 1 - Latifa Ika Sari & Ria Hermina Sari What do people do to sustain a conversation: Analysis of the features and the speech function of a casual conversation 118 Question: Open: Fact 1 1 1 Statement: Fact 2 2 4 Statement: Opinion 3 2 - TOTAL 13 6 6 Continue Monitor 2 2 4 Prolong: Elaborate 9 3 13 Prolong: Enhance 3 1 6 Prolong: Extend 4 - - Append: Elaborate 10 3 15 Append: Enhance - - 3 Append: Extend 4 1 3 TOTAL 32 10 44 React: Responding Support: Register 29 17 6 Support: Reply: Acknowledge 10 2 11 Support: Reply: Agree 22 18 12 Support: Reply: Answer 2 1 5 Support: Reply: Affirm - 2 - Support: Develop: Elaborate 4 6 3 Support: Develop: Enhance 4 - 3 Support: Develop: Extend 7 2 3 Confront: Disavow - 2 1 TOTAL 78 50 44 React: Rejoinder Support: Track: Clarify 1 2 3 Support: Track: Confirm 4 4 1 Support: Track: Probe 2 3 1 Support: Response: Resolve 5 2 4 Confront: Challenge: Rebound - - 2 Confront: Challenge: Counter 1 1 - TOTAL 13 12 11 Thus, it can be concluded that mostly S2 provides supporting moves when responding to other speakers. The number of responding moves also indicates that S2 is strikingly dependent on the other speakers. She talks most only in reaction to the contribution of other speakers. Rejoinders have a function to sustain the interaction. They express a willingness of the speakers to maintain contact and sustain conversation (Eggins & Slade, 1997). In this conversation, the number of rejoinders produced by each speaker are about equal. S1 produces 13 moves, S2 produces 12 moves and S3 produces 11 moves. Therefore, it indicates that all speakers contribute to the maintaining and sustaining of the talk. Almost all rejoinders produced by speakers is supporting rejoinders. It shows that all speakers want to support each other to sustain the conversation and to create a harmonious talk between them. CONCLUSION Spoken language has different characteristics from written language. It has features of spontaneity, interactivity and interpersonality. Filled pauses, repetitions, false start and backtracking and incomplete utterances produced by the speakers involved in this conversation indicate that they talk spontaneously. The pauses and repetition are used by the speakers as time- gaining strategies. They give the speakers time to think about what they are going to say. False start and backtracking are used as self-monitoring strategies. The speakers correct their utterances when they realize that they make errors in speaking. The speakers also show the use of chunks in their conversation. Chunks is a part of formulaic competence that can ease the speakers‘ burden to memorize grammar rules and can enhance the speakers‘ fluency. The speakers also cut off long utterances into short ones as one of the strategies to make people understand more about the topic that is being discussed. This conversation is also interactive. This is indicated by turn-taking, interruptions or overlapping utterances, signals of amusement (laughs and chuckles), backchanneling, and the use of discourse markers such as well, yeah, but, you know, etc. The speakers use those strategies to keep the conversation ―alive‖ and interactive. Beside interactivity, interpersonal features also appear in many parts of the conversation. The speakers complete and repeat each other‘s utterances and use exaggeration language to show solidarity. They also use certain expressions to Indonesian EFL Journal (IEFLJ) Volume 6, Issue 2, July 2020 p-ISSN 2252-7427, e-ISSN 2541-3635 https://journal.uniku.ac.id/index.php/IEFLJ/index 119 blunt the force of a disagreement. Coherence in the conversation is established by the cooperation of the speakers to give relevant responses towards each other‘s moves. By giving relevant responses, the speakers can sustain the conversation. Analysis of negotiation of meanings shows that the speakers mostly negotiate interpersonally. They negotiate feelings and attitude more frequently than negotiate message or news. This can happen because the speakers have a close interpersonal relationship and they talk in a casual conversation setting. Since this is a casual conversation, the opening, the closing, and the story sequence has no fixed structures. The topic changes several times and there are several openings and closings. The analysis of speech function shows that S1 is the most dominant speaker while the most passive speaker is S2. However, S2 gives responses quite often that mostly consist of registering and agreeing. It indicates that S2 is strikingly dependent on the other speakers in which she talks most to support other speakers. S3 produces continuing moves more often than other speakers. She elaborates her utterance to make other speakers understand better. The number of rejoinders that each speaker produced indicates that they are willing to support each other to sustain the conversation. These findings have a great pedagogical implication in language teaching and learning. English teacher should include communication strategies when teaching speaking. Teacher should not only teach vocabulary, sentence structure, or pronunciation but also strategies to sustain conversation to achieve the goals of communication. Time-gaining strategies, self- monitoring strategies, and interactional strategies may be utilized in communication to help individuals to sustain the conversation and to create a harmonious talk. ACKNOWLEDGEMENT I would like to thank to the fellow lecturers who have been willing to be involved in this study as participants. REFERENCES Akmaliyah, N. (2014). Classroom related talks: Conversation analysis of Asian EFL learners. Englisia Journal, 2(1), 1–19. https://doi.org/10.22373/ej.v2i1.323 Albert, S. (2017). Research methods: Conversation analysis. In M. Schober, D. . Rapp, & A. . Britt (Eds.), The Handbook of Discourse Processes 2nd edition (2nd Editio, pp. 99–108). Taylor and Francis. https://doi.org/10.4324/9781315687384 Barraja-Rohan, A. M. (2011). Using conversation analysis in the second language classroom to teach interactional competence. Language Teaching Research, 15(4):479-507 https://doi.org/10.1177/1362168811412878 Barthel, M., Meyer, A. S., & Levinson, S. C. (2017). Next speakers plan their turn early and speak after turn-final ―go-signals.‖ Frontiers in Psychology, 8(393), 1–10. https://doi.org/10.3389/fpsyg.2017.00393 Bögels, S., Kendrick, K. H., & Levinson, S. C. (2015). Never say no . . . How the brain interprets the pregnant pause in conversation. PLoS ONE, 10(12), 1–15. https://doi.org/10.1371/journal.pone.0145474 Buchholz, M. B., & Kächele, H. (2013). Conversation analysis – A powerful tool for psychoanalytic practice and psychotherapy research. Language and Psychoanalysis, 2(2), 4–30. https://doi.org/10.7565/landp.2013.004 Buchholz, M. B., & Kächele, H. (2017). From turn-by- turn to larger chunks of talk: An exploratory study in psychotherapeutic micro-processes using conversation analysis. Research in Psychotherapy: Psychopathology, Process and Outcome, 20(257), 161–178. https://doi.org/10.4081/ripppo.2017.257 Celce-Murcia, M. (2007). Rethinking the role of communicative competence in language teaching. In E. A. Soler & M. P. S. Jorda (Eds.), Intercultural Language Use and Language Learning (pp. 41–57). Springer Science. https://doi.org/10.1007/978-1-4020-5639-0_3 Chan, C. S. C. (2019). Long-term workplace communication needs of business professionals: Stories from Hong Kong senior executives and their implications for ESP and higher education. English for Specific Purposes, 56, 68–83. https://doi.org/10.1016/j.esp.2019.07.003 Cornish, F. (2006). Understanding spoken discourse (Vol. 13, pp. 227–230). Elsevier Ltd. retrieved on 2 June, 2020 from https://hal-univ- tlse2.archives-ouvertes.fr/hal- 00952132/document Dahal, B. (2010). Analysis of spoken discourse pattern in Nepali ELT classes. Journal of NELTA, 15(1– 2), 22–27. https://doi.org/10.3126/nelta.v15i1- 2.4606 De Jaegher, H., Peräkylä, A., & Stevanovic, M. (2016). The co-creation of meaningful action: Bridging enaction and interactional sociology. Philosophical Transactions of the Royal Society B: Biological Sciences, 371, 1–10. https://doi.org/10.1098/rstb.2015.0378 Eggins, S., & Slade, D. (1997). Analysing casual conversation. Creative Print and Design. Retrieved on 2 June, 2020 from https://books.google.com/books?printsec=frontc over&vid=ISBN0304337293&vid=ISBN030433 https://hal-univ-tlse2.archives-ouvertes.fr/hal-00952132/document https://hal-univ-tlse2.archives-ouvertes.fr/hal-00952132/document https://hal-univ-tlse2.archives-ouvertes.fr/hal-00952132/document Latifa Ika Sari & Ria Hermina Sari What do people do to sustain a conversation: Analysis of the features and the speech function of a casual conversation 120 7285&vid=LCCN96043841#v=onepage&q&f=f alse Fasulo, A., & Sterponi, L. (2016). Understanding children‘s mental health conditions in their interactional environment: Conversation analysis and autism. TPM - Testing, Psychometrics, Methodology in Applied Psychology, 23(4), 453– 470. https://doi.org/10.4473/TPM23.4.3 Hale, C. C., Nanni, A., & Hooper, D. (2018). Conversation analysis in language teacher education: An approach for reflection through action research. Hacettepe University Journal of Education, 33(Special Issue), 54–71. https://doi.org/10.16986/HUJE.2018038796 Halliday, M. A. K. (1989). Spoken and written language. Oxford University Press. Hidayat, D. N. (2019). Conversation analysis and its implications to language teaching. TARBIYA: Journal of Education in Muslim Society, 6(2), 197–209. https://doi.org/10.15408/tjems.v6i2.15138 Hoey, E. M., & Kendrick, K. H. (2018). Conversation analysis. In A. M. . de Groot & P. Hagoort (Eds.), Research Methods in Psycholinguistics: A Practical Guide (pp. 120–134). Wiley Blackwell. https://doi.org/10.4324/9780203809068-16 Kendrick, K. H. (2017). Using conversation analysis in the lab. Research on Language and Social Interaction, 50(1), 1–11. https://doi.org/10.1080/08351813.2017.1267911 Koole, T. (2013). Conversation Analysis and Education. In C. A. Chapelle (Ed.), The Encyclopedia of Applied Linguistics. Blackwell Publishing Ltd. https://doi.org/10.1002/9781405198431.wbeal01 99 Kpogo, F., & Abrefa, K. B. (2017). The Structure of Face-To-Face Casual Conversation Among the Akans. Ghana Journal of Linguistics, 6(2), 56– 84. https://doi.org/10.4314/gjl.v6i2.4 Martin, J. R., & Rose, D. (2007). Working with discourse (2nd Editio). Continuum. Peräkylä, A. (2019). Conversation analysis and psychotherapy: Identifying transformative sequences. Research on Language and Social Interaction, 52(3), 257–280. https://doi.org/10.1080/08351813.2019.1631044 Schiffrin, D. (2001). Discourse markers: Language, meaning, and context. In The Handbook of Discourse Analysis (pp. 54–75). Blackwell Publishers Ltd. https://doi.org/10.4324/9780203813324-7 Sert, O., & Seedhouse, P. (2011). Introduction: Conversation analysis in applied linguistics. Novitas-ROYAL: Research on Youth and Language, 5(1), 1–14. Takeda, R. (2013). Language Learner Identities: Utilizing Conversation Analysis in the Classroom. Accents Asia, 6(1), 32–42. Thornburry, S. (2005). Beyond the sentence: Introducing discourse analysis. Macmillan Education. Trippas, J. R., Spina, D., Cavedon, L., & Sanderson, M. (2017). A Conversational search transcription protocol and analysis. ACM SIGIR Workshop on Conversational Approaches for Information Retrieval. Retrieved on 2 June, 2020 from https://www.johannetrippas.com/papers/Trippas et al-CAIR2017-protocol.pdf Verhoeven, L. (1994). Functional literacy: Theoretical issues and educational implications. John Benjamin B.V. Vo, T. H. Le, Wyatt, M., & McCullagh, M. (2016). Exploring the gap between Vietnamese workplace communication in English and English language teaching at a university. Asian ESP Journal, 12(1), 8–38.