EEJ 8 (3) (2018) 331- 341 English Education Journal http://journal.unnes.ac.id/sju/index.php/eej The Realization of Conversational Features in Massively Multiplayer Online Game Chat Entitled ‘Mafia City’ Wanda Giovani, Sri Wuli Fitriati Universitas Negeri Semarang, Indonesia Article Info ________________ Article History: Accepted 24 June 2018 Approved 12 August 2018 Published 15 September 2018 ______________ Keywords: Conversation Analysis, Massively Multiplayer Online Game Chat, Conversational Features ____________________ Abstract ___________________________________________________________________ The aim of this study is to describe and to explain the realization of conversational features in massively multiplayer online game chat entitled Mafia City. This study used a conversation analysis approach with a qualitative design. The focus of this study is the gamers‟ interactions through the game chat. The instrument of the research is I myself as the participant observer. Tables of specifications are the secondary instrument in the observation. The results of the study show that in Mafia City, the use of conversational features such as turn- takings, adjacency pairs, and repairs is possible. Even so, the realization of those features was not exactly similar compared to those in face-to-face conversation. The main problem was the disrupted turn-takings and adjacency pairs. For repairs, the most common type that appeared in the chat was self-initiated self- repair, while the most common pattern that occurred was replacement. Their occurrences were in consequence of the nature of online chat which is visually decontextualized. For jargons, the chatters are creative enough to create acronyms, contractions, and abbreviations freely. For the realization of extra- linguistic features, the chatters used emoticons as the result of the chatters‟ creativity in dealing with the lack of prosodic features. These findings can be beneficial for the development of discourse study, especially in internet-based communication. © 2018 Universitas Negeri Semarang Correspondence Address: Pascasarjana Kampus Kelud Utara III, Semarang, 50237 E-mail: wandotwandot@gmail.com p-ISSN 2087-0108 e-ISSN 2502-4566 http://journal.unnes.ac.id/sju/index.php/eej Wanda Giovani, Sri Wuli Fitriati/EEJ 8 (3) 2018 331 - 341 332 INTRODUCTION Conversation analysis can be defined as the study of talk. It investigates the communication between two people or more who interact in a certain circumstance. According to schegloff and sacks (1973, p.289), conversation analysis is a „naturalistic observational discipline dealing with the details of social action rigorously, empirically and formally‟. It aims to „describe, analyze, and understand talk as a basic and constitutive feature of human social life. There have been a large number of studies on conversation analysis, yet almost all of them are on oral conversations which occur on real life setting, whereas, nowadays people also talk via internet. As the effect of technology, today‟s communication has been changed. Conversation is no longer in form of face-to-face talk, or just via telephones. At present, there is internet that allows anyone to get involved into multi-party conversations which are mostly on online chatting. Online chatting is not in oral form, yet it is not exactly in written form either (in which grammatical rules are applied and well organized). It is like a written spoken interaction through internet. It is also a form of social interaction, because it follows the rules that govern social interaction among the speakers/chatters. Thus it can be analyzed by using conversation analysis. There are some aspects of conversation analysis according to Paltridge (2006), they are; turn taking, adjacency pairs, preference organization, feedback and conversational repair. Three of the aspects mentioned will be the focus of this present study; turn taking, adjacency pairs, and repair. It is because these aspects are the major fundamental organizations of conversation. Turn-taking is a type of organization in a conversation where the participants speak one at a time in alternating turns. Further, turn taking can be defined as a cyclical process which begins with one person speaking, and continues as the speaker gives up control to the next person. The second speaker now has the conversational floor. When the speaker is finished, they give control back to another speaker (in this case, the beginning speaker), thus creating a cycle. The turn taking cycle stops when there is nothing left to say (Woodburn, Arnott, Newell, and Procter 5). In addition, C. E. Ford (2012) states that turn taking refers to a system through which interlocutors manage transfer of speakership, with each next turn in sequence displaying how that speaker has interpreted previous talk and the action so far. Cameron (2001) believes that there is an ordered set of rules for the allocation of the next turn. These rules are: (1) the current speaker selects the next speaker. If this mechanism does not operate, then (2) the next speaker self-selects. If this mechanism does not operate, then (3) the current speaker may continue. On the other hand, Herring (2001) claims that turn-taking rules and assumptions in spoken conversation described above do not hold on online chatting. In addition, Liddicoat (2007) states that the interpretation of turn- taking focus on how individuals know how to take turns. Individuals, if asked, often maintain pauses represent cues they use to know when to take turns. Adjacency pairs refer to two turns, which are usually consecutive and uttered by different speakers. A question followed by an answer is one example of an adjacency pair. "Hello" followed by "Hi" is another adjacency pair. These pairs typically correspond to a specific order& for example, a question always precedes an answer. Indeed, some facets of conversation are intended to initiate a specific range of actions by the other person, called first pair parts (Moss, 2016). That is why, adjacency pairs also known as the concept of nextness in which the second utterance is functionally dependent on the first, as exhibited in conventional greetings, invitations, and requests (Nordquist, 2015). Furthermore, Psathas (1997) summarizes eight major dimensions of the adjacency pair structure, which are; (1) there are at least two turns in length, (2) they have at least two parts, (3) the first part is produced by one speaker, (4) the second part is produced by another speaker, (5) the sequences are in immediate next turns, Wanda Giovani, Sri Wuli Fitriati/EEJ 8 (3) 2018 331 - 341 333 (6) the two parts are relatively ordered so that the first belongs to the class of first pair parts, and the second to the class of second pair parts, (7) the two are discriminately related in that the pair type, the first of which is member, is relevant to the selection among second pair parts, and (8) the two parts are in relation of conditional relevance; the first sets up what may occur as second, and the second depends on what occurred as first. Repairs include any attempts to redress errors and difficulties in the conversation. For example, when individuals cannot retrieve the correct word, they often elongate a previous syllable--as a means to demonstrate they need more time-or deploy non-lexical pertubations such as “um” or “uh”. Similarly, after they complete their turn, but recognize an error, they might interject with phrases that begin with "I mean". On other occasions, the other person might highlight an error, perhaps by emphasizing a word. One person might refer to their desire to lend a book and the other person might repeat this phrase, but emphasize the word lend, to imply that borrow might be the correct term (Moss, 2016). Yule (2006: 211) states that jargon is a special technical vocabulary associated with a particular area of work or interest. In social terms, jargon helps to create and maintain connections among those who see themselves as „insiders‟ in some way and to exclude „outsiders‟. Moreover, Crystal (2003) defines jargon as the technical vocabulary or idiom of special activity or group. The reality is that everyone uses jargon; it is an essential part of the network or occupation and pursuits which make up a society. Halligan (2004) elaborates four kinds of jargon. They consist of acronym, abbreviation, word, and phrase. It is supported by Chaika‟s theory (1982) telling that jargon appears in two ways. Firstly, it is shaped in the form of new words. Secondly, it appears in the form of existed words as common people know. Thus, Jargons include any kind of slang which appears in online chatting; they are acronyms, abbreviations, and contractions that become the signature feature of internet language style. The extra linguistic features of online conversation are signs, smiley and emoticons. These features replace the absence of physical movements as can be found in oral and face-to- face conversation. The problems in different principles of conversation analysis in on line chat is to some extant related to lack of extra linguistic features such as gestures, hand movements, eye contact, pitch and intonation. Absence of these features cause confusion, overlaps, ignorance, ambiguity, and irrelevant responses (Zaferanieh, 2012, p. 7). To compensate the absence of these cues, keyboard characters have been designed. They are called smileys or emoticons. They are emotional graphic-visual ways for expressing the way one feels when the words alone are not enough (Nuckolls, 2005: 66). Furthermore, Puterman (cited in Abdulla, 2005:51) believes that the use of emoticons is a way for keeping a conversation going and she refers to them as „icons used for expressing emotions‟ Thus, this research is conducted to investigate turn taking, adjacency pairs, repair, jargons and extra linguistic feature in online conversation. The setting that is taken is the chat room of an online game named “Mafia City”. This game is a kind of Massively Multiplayer Online Game (MMOG) brought by Anansi Mobile. It is played from smart phones, like android and windows phone by many people from different countries. The main language used in the chats is English. “Mafia City” has at least three different chat rooms for a new player; global chat, flight chat, and newbie chat. Yet, this research focuses on observing the other chat, it is the family chat. The family chat can only be accessed by players who reached level 15 and have a family. The reason of choosing family chat is because it is more alive than the other chat rooms. Players are more talkative in their family chat, because there they talk with their family members. They can freely talk about strategies of the game, share their feelings and thoughts, or even talk gossip about other players from different families. Wanda Giovani, Sri Wuli Fitriati/EEJ 8 (3) 2018 331 - 341 334 METHOD The focus of this study is the gamers‟ interactions through the game chat of Mafia City MMOG. The data were taken from Korrupt Family chat room. The unit of analysis is the features of the conversation itself: turn takings, adjacency pairs, repairs, jargons, and the extra- linguistic features. This research is a qualitative design; the instrument is thus I myself as the participant observer in the game as well as the writer of this study itself. Moreover, for the secondary instruments, there are some tables that were used to note important information of the data taken from the observation about the five conversational aspects which are the focus of this research. The instruments for each aspects can be seen as follows: Instrument of Turn-Takings and Adjacency Pairs Figure 1. Schematic representation of turn- taking in an online chat sample (adapted from Herring, 1999) This schematic representation has been adopted for obtaining the data about turn- takings and adjacency pairs in this research. Even so, the transcription in this research was adjusted into a quiet different form. It is done in order to make the identification become easier. Thus, the transcription of Mafia City game chat thread consists of three parts, which are; the number of turns, the participants, and the messages. For identifying the turn-takings I read the whole chat thread and connect the related chat turns manually (by using curly brackets). It is also done for the adjacency-pairs identification. One curly bracket represents an adjacency-pair, which is the sign of a related turn-taking. The length of the curly bracket signs the number of unrelated turns which appear between two conversing turns. If the curly bracket is long, it means there are a lot of unrelated turns separate one turn to its response. Instrument of Repairs For the instrument of repairs, I use the mechanisms of repairs in conversation adopted from Liddicoat (2007) as the lead theory to determine the patterns and the types of repairs in order to construct the table of specification of repairs in Mafia City game chat. The repairs found are classified according to the patterns of repair such as; replacement, modification, elaboration, etc. The patterns are then classified again into the types of repair which consist of self-initiated self-repair, self-initiated other- repair, other-initiated self-repair, and other- initiated other-repair. Instrument of Jargons Halligan‟s kinds of jargon (2004) is the lead theory for the instrument of jargons in this research. This theory is used to determine the form of jargons in Mafia City game chat which consists of contractions, acronyms, and abbreviations. The meanings of the jargons are also included in the table of specification. They are determined according to the context of the chat lines. The jargons are taken from the chat transcription. After being highlighted and their meanings being understood, the jargons are then transferred into table of specification. Instrument of Emoticons The extra-linguistic features of Mafia City game chat are in form of emoticons. The lead theory for the identification of emoticons in this research uses Ruan‟s classification of emoticons (2011). The data of emoticons are taken from the chat transcription which then highlighted with three different colors based on their kinds. It is done to avoid confusion in the data input process. The emoticons put in the table are Wanda Giovani, Sri Wuli Fitriati/EEJ 8 (3) 2018 331 - 341 335 exactly the same with what have found in the original chat threads. The meanings of the emoticons are also included in the table of specification. After being highlighted and interpreted the emoticons are then classified into the table of specification. The procedures of data analysis are: (1) Fragmenting the conversation according to its topic. In other words, the collected data were simplified onto the important points, such as; turn-takings, adjacency pairs, repairs, jargons, and extra-linguistic features of online conversation. (2) Classifying the data in relation to the focus of the research, or the conversational aspects that previously mentioned. Classification was done by identifying the minor data from each of the five fragmented major data. (3) Interpreting the features of conversation in the chat room according to Brian Paltridge‟s theory of CA (2006). The data reporting covers three stages, such as; discussing, concluding, and giving suggestion. There are two techniques to report the data, they are the formal and the informal technique. In this research, I used the informal technique for my data report. The trustworthiness of this research was gained by doing triangulation. This study used an investigator triangulation. Through investigator triangulation, I asked an expert who is an English Lecturer majoring linguistics in Universitas Negeri Tanjungpura, Pontianak. In addition, in the data analysis, there are numerous adequate theories from various sources about Conversation Analysis and its features that I used as my theoretical framework. RESULTS AND DISCUSSION Turn-taking is about what A says before B and conversely. In spoken conversation, mostly, the turn-taking system is well organized. The conversation runs precisely, where the timing of transition of an utterance to its response has no significance gap. The speakers know how to construct and allocate turns. The turns appear together as adjacency pairs or as the initiation- response-follow up of an exchange. Contradictory, the turn-taking system in Mafia City game chat is in disorganized form. There are many utterances appeared randomly with no coordination to the previous turns. The response of the initial turn often appeared after one or two or even more new turns which are not its pairs. The following excerpts are the examples of chat lines in Mafia City which are in „disrupted turn adjacency‟ forms: Excerpt 1 1. Mama Bear Im not a gold buy n took me ages to collect gold for 1 wheel lol rather collect something to aim for :) 2. Fallen Im waiting double gold to buy 50 :”) 3. Zayed Someone is loaded 4. Khaleesi Anybody home? 5. Khaleesi Im watchin fast and furious 6. Queen Double gold already started though right? Excerpt 2 7. Zayed If McGregor loses u will pay up 8. Lua zZZz 9. Chris ^ :* 10. Chris Mcgregor doesnt lose though 11. Lua ^<33 12. Lua Lmao” you made me angry, bye” 13. Chris Meu amorrrrr *--------* 14. Chris mt dlc :* 15. Lua Meu amoooor gostoooosooo <3 16. Zayed Well hes a fighter not a boxer Wanda Giovani, Sri Wuli Fitriati/EEJ 8 (3) 2018 331 - 341 336 Excerpt 3 17. Chris I will try to call tomorrow lu :* 18. Chris Mcgregor #1 19. Young_ Vezzo And no Zay it wont be interesting lmao its going to ve very lopsided and more than likely not very entertaining which is why im not going to pay to watch it 20. Aslan Reset 51? 21. Zayed His fight with Pacquaio wasn‟t entertaining either 22. Zayed I don‟t follow box. Just read the news XD 23. Young_ Vezzo Exactly and that‟s top lvl boxer 24. Young_ Vezzo Im a big fan of both I used to box 25. Young_ Vezzo I have a lot of the old UFC‟s on dvd 26. Lua Okay amor <3 As can be seen in the excerpt 1, Queen responded to MamaBear and Fallen in an hour later after Zayed‟s turn, even after Khaleesi‟s made a new topic of conversation. Meanwhile in the excerpt 2, the conversation between Zayed and Chris was disrupted by Lua who made gaps between Zayed and Chris‟s lines. Also in excerpt 3, the conversation between Chris and Lua was separated by several lines from other participants who had no relation to Chris‟ and Lua‟s messages. The reason of this happened is that the fact that Mafia City chat is like the other chat where the line of an utterance is typed, appeared and saved on a screen, and it can last for a long period of time. In Mafia City a chat line can last for 15 days. As long as players scroll-up the screen, they can read any lines in various topics of conversation until 15 days ago. They might choose any line they want to response, although it was appeared several minutes of even several hours ago. Unlike in spoken conversation, once an utterance said, it cannot be listened by other speakers who are not in the same place with the first speaker. It is impossible for them to response to the first speaker even after they are finally in the same place with him/her five minutes later. Because the utterance said is already gone. That is why; the turn taking system in spoken conversation is more organized than the Mafia City online chat which is full of overlaps and delays. Similar to the turn-taking principle which is seemed to be violated most of the times, the realization of adjacency pair in Mafia City game chat is also disorganized. There were cases where one statement had several responses which make some respondents being ignored. As can be seen in the following excerpt, when MamaBear asked about the price of security box pieces, it was only Zayed‟s answer that he responded back. Excerpt 4 27. Mama Bear How much security box pieces worth? 28. Fallen Idk 29. Zayed 50-70k a piece 30. Zayed Is a long process to collect all pieces,better buy the whole box with 2k gold 31. Green Day 69k lol but I can give u one for free. U want? 32. Mama Bear Thanks I was told that price b4 just wanted to double check someone asking for 16mil for 160 pieces.. There are also cases when chatters asked questions and their questioned being ignored. This thing happens because of the absence of the eye contact in an online chat like Mafia City. Thus the addressing of the message delivered is unclear. This problem is usually happen to a new member who has no close friend in the chat room. He/she might be confused or shy to deliver their message to specific persons by calling their names. Also the old players are seemed to be more attentive towards the old players who are known as regular chatters. The Wanda Giovani, Sri Wuli Fitriati/EEJ 8 (3) 2018 331 - 341 337 example below shows that Queen which is a new player asked questions and she was ignored. Excerpt 5 33. Khaleesi Im watchin fast and furious 34. Queen Double gold already started though right? 35. Fallen Deve tau ma merda, pra ta no jogo 36. Queen U watching fast and furious 8? 37. Brenda kk fallen. XD 38. Fallen XD Brenda 39. Chris I haven‟t even seen 7 yet XD Similarly, in the excerpt 6 below, it can be seen how chatter requested something and nobody paid attention towards his request. Excerpt 6 40. Splashy Hmmm anyone got 12 fragments 41. Zayed Freaks 40 mine 42. Zayed If McGregor loses u will pay up 43. Lua zZZz 44. Chris ^ :* Not only for question-answer, there was also other kind of adjacency pair like greeting- greeting also failed. This is still caused by the unpopularity of chatter since she/ he is a new member. The following example shows a new member named Diana ! said hello, and nobody response to her. Excerpt 7 45. Zayed Done robbing eggs, gn 46. Binx Ty zay 47. Diana! Hola fam ^-^ 48. Chris How many eggs? 49. Zayed 300. not bad There are two types of repairs that mostly appear in Mafia City game chat, they are self- initiated self-repairs and other-initiated other- repairs. From the data of chat, self-initiated self- repairs appeared 58 times, other-initiated other- repairs appeared 24 times, and followed by other-initiated self-repair 17 times, and the most rarely one to appear was self-initiated other- repair that appeared 7 times. The reason that causes self-initiated self repair becomes the type of repair that mostly appear is that the nature of online game chat which is typified or visually decontextualized. It has no physical contacts as in face-to-face interaction. It takes longer time to read than to hear. It results on delays of feedback by other participants. That is why when chatter thinks that he/she just made a mistake on their utterances; they may initiate to repair the mistake as soon as they aware of it. On the other hand, there are also some other participants who become as active chat readers. They may interrupt the other chatters who make mistakes as soon as they read their lines, moreover if the chatters who become the trouble sources have not realized their mistakes yet. That is why, other-initiated other-repairs becomes the second type of repairs that mostly occur in the data. Meanwhile, the patterns of repairs that occur most are replacement with 26 occurrences, followed by other patterns which are rewording (16 times), specification (15 times), modification (15 times), elaboration (15 times), clarification (5 times), exemplification (3 times), repetition (2 times), and reorganization (1 time). Replacement pattern is often used by the chatters to replace words with another that is considered more appropriate or precise by the repair initiator. The following excerpts are the examples of replacements occurred in Korrupt family chat. Wanda Giovani, Sri Wuli Fitriati/EEJ 8 (3) 2018 331 - 341 338 Excerpt 8 50. Zayed Fate and furious opening today here too 51. Yellow Fate? Is a new movie? 52. Fallen Lmao yellow don‟t be rude 53. Zayed Fast and furious yes 54. Brenda :) Excerpt 9 55. Zayed U will get the medal at 750k stats I think. U didn‟t doubt me 56. Zayed *dodge Excerpt 10 57. Vanda Two hours ago? The chat been so quiet 58. Succubus Siiis *-* 59. Vimay Anybody have barman license?? 60. Vanda Hello cubus *-* 61. Succubus < sist nalu here XD missed u linda In the excerpt 8, replacement occurs as other-initiated self repairs. It can be seen that Brenda tried to repair Zayed‟s utterance by questioning the word “fate”, and then after that Zayed replaced the word “fate” became “fast”. Whereas, in the excerpt 9, replacement occurs as self-initiated self-repair where Zayed replaced the word “doubt” became “dodge”. And in the excerpt 10, the type of replacement is other- initiated other-repair. As can be seen there, Succubus which is Nalu‟s alternative account initiated to replace Vanda‟s utterance “cubus” with “sist nalu” as she revealed herself to Vanda. Nalu and Vanda are close friends in the game, Vanda usually calls Nalu as “Sist Nalu”. However, in the case, Vanda had not known yet that Succubus is Nalu. Jargons are special terms which are only understood by the members of a specific group or community. In Mafia City there are various special terms which become the signature words of the game. These words appear almost everyday in the game chat room where each of them has different purpose. The players often type them in form of contractions, acronyms, or abbreviations. From the data taken, there are 101 kinds of jargons which consist of 36 contractions, 36 acronyms, and 29 abbreviations commonly appear in Mafia City chat. Their occurrences number in total is 774 times taken from 1.230 chat lines. The following examples are the data which show some kinds of jargons that commonly appear in Mafia City game chat. Excerpt 11 62. Lucifer:) Buying super eggs. Mm 63. Lucifer:) Zay you there? In the example above, it can be seen a player named Lucifer:) used acronym “MM” which stands for “Mail Me”. This acronym is very common in Mafia City game chat, not only in Korrupt family chat room, but also in the other chat rooms. It is used as a request, an order, or invitation to make a deal for something (commonly for trading) via private mail. There are also other similar acronyms which stand for subject+predicate, or as a predicate only, such as “IDK” (I don‟t Know), “IDC” (I Don‟t Care), “NVM” (Never Mind), NP (No Problem), etc. Moreover, there are also acronyms used for showing expression of amusement such “LOL” and “LMAO”. Excerpt 12 64. Lucifer:) Don‟t tell others, lol, you can have your share 65. Zayed Got it The acronym “LOL” stands for “Laughing Out Loud”. This kind of acronym is very popular to use among the chatters. In the example, Lucifer:) used it as the response of Zayed‟s joke in the previous line. The other similar acronyms which represent laughing are Wanda Giovani, Sri Wuli Fitriati/EEJ 8 (3) 2018 331 - 341 339 “LMAO” (Laughing My A** Off). This jargon gives stronger expression of laughing than “LOL”. If “LMAO” is used in the chat line, it means that the chatter found something very funny to response. As can be seen in the example below, a player named Fallen gave “LMAO” to response Yellow‟s satire towards Zayed‟s mistake for typing the word “Fast”. Excerpt 13 (the example is taken from excerpt 8) Zayed Fate and furious opening today here too Yellow(33) Fate? Is a new movie? Fallen Lmao yellow don‟t be rude Zayed Fast and furious yes Besides as actions and expressions, the acronyms in Mafia City game chat are also as nouns. They are mostly stand for the things which can be only found in Mafia City MMOG, like “SS” (Sky-Scrapper), “BG” (Body-Guard), “WC” (Wild City), “IB” (Ignos Blade) and so on. These terms might be also found outside Mafia City, yet it is not in the same context and also rarely found in form of acronyms. The main problem of analyzing online conversation is the lack of prosodic features, such as gestures, hand movements, eye contact, pitch and intonation. It can result on overlaps, ambiguities, and confusions among the chatters. As the solution, there are emoticons and smileys which resemble to the prosodic features in spoken conversation. Their functions are as devices and signs to show chatters‟ expressions, or feelings, to make the utterances which are typed on the chat become more understandable. In Mafia City game chat; there is no smiley button which serves a various kinds of expressions that can choose just by one click, as in messengers, or social media. As the consequence, the players create their own emoticons and smileys by combining various letters and punctuation marks on the game keyboard. The common emoticons in Mafia City game chat are those that represent facial expressions, actions, and objects (people, animals, and others). Whereas emoticons that show tones and appearances are seldom found, thus, any emoticon which resembles human face is classified into facial expressions too. The following examples show how the players use emoticons in their utterances. Excerpt 14 66. Chris Im making another champion belt so I can be like Connor McGregor :) In the excerpt 14 above, Chris put an emoticon as one of the facial expressions, which is a happy face. He seems to be a fan of Connor McGregor (an MMA famous athlete) that he showed his excitement to make another champion belt (a kind of armor in the game) so that he can be like McGregor who in the real life owns two champion belts. Moreover, the two following examples are excerpts of the data which show the other form of facial expression emoticons. Excerpt 15 67. Cha Red I think I need to clean my alt again for ss >_< Excerpt 16 68. Brenda kk fallen. XD 69. Fallen XD Brenda 70. Chris I haven‟t even seen 7 yet XD There are two facial expressions in the excerpts above, which are >_< and XD. The first expression is the expression of a grinning face, and the second one is the expression of amused face. These emoticons are very popular to be used in Mafia City game chat. There are also ^-^ which means a happy face, O.O as the representation of a shocked face, : / as the image of a doubtful face, -.-“ as the illustration of a tired face, *-* as the expression of an amazed face, :”) as the sign of an affected face, and : ( for showing a sad face. These emoticons do not only help the chatters to make their utterances Wanda Giovani, Sri Wuli Fitriati/EEJ 8 (3) 2018 331 - 341 340 become more expressive, but also make the chat lines look more attractive as can be seen in the following excerpts below. Excerpt 17 71. Diana! Hola fam ^-^ Excerpt 18 72. Yellow Fck u fallen, im not rude 73. Binx O.O CONCLUSION Based on the analysis it was found that in Mafia City game chat, the use of conversational features such as turn-takings, adjacency pairs, and repairs is possible. In the other words, some features of spoken conversation can be also applied in an online field, as in Mafia City game chat. Even so, the realization of those features was not exactly similar as can be found in spoken or face-to-face conversation. The main problem was the randomness of chat lines which resulted on the disrupted turn-takings and adjacency pairs. Besides, the realization of repairs in Mafia City was also different to those that realize in face-to-face or real life conversation. The replacements, modifications, specifications, elaborations, and rewordings that found in the study did not realize similarly with those that are used in spoken interaction. The participants of the chat often used * for repairing the incorrect words from their own utterances, and they also initiated repair on other chatters‟ error by joking and questioning the words. Also, the absence of autocorrect and edit feature in the game keyboard resulted on high number of typos; some of them were just being read without any repair. Furthermore, the jargons and the extra- linguistic features in Mafia City were even more different, not only with those that realized in face-to-face interaction, but also compared to those that are used in common social media such as Whatsapp, Facebook, Intsagram, etc. There might be found some general chatting jargons like LOL, LMAO, and TY, yet there were also other various jargons such as IB, Alts, DKS etc. which only common to use in Mafia City. Also, there is no rule for creating the jargons. The acronyms and abbreviations found in the field were mostly in lowercase form, and the contractions were mostly lack of apostrophes. Lastly, the absence of chat features in the game keyboard also differentiated the extra-linguistic features used in Mafia City. When other common social media users can use smiley, stickers, gif images and emoticons instantly, the Mafia City chatters can only use emoticons that they must create firstly by combining certain letters and punctuation marks. REFERENCES Abdulla, A. (2005). A linguistic study of internet chat (Unpublished M.A. Thesis: University of Mosul) Abdullah, I. H. H., Laila, M., & Hum, M. (2016). The study of turn-taking In jane eyre movie 2011 (Doctoral dissertation, Universitas Muhammadiyah Surakarta). Retrieved from http://eprints.ums.ac.id/45089/1/PUBL ICATION%20ARTICLE.pdf Cameron. D. (2001). Working with Spoken Discourse. London: Sage Publications. Chaika, E. (1994). Language, the social mirror. Boston: Heinle & Heinle Pub. Crystal, D. (2003). English as a Global Language (Second edition). Cambridge: Cambridge University Press. Ford, C. E. (2013). Conversation analysis and turn taking. The Encyclopedia of Applied Linguistics. Retrieved from https://onlinelibrary.wiley.com/doi/full /10.1002/9781405198431.wbeal0216 Halligan, N. (2004). A short course on writing technical reports. Technical Writing. Retrieved from http://www.technical- writing-course/index.html. Herring, S. (2001). Computer Mediated Discourse. In Schiffrin, D., Tannen, D., http://eprints.ums.ac.id/45089/1/PUBLICATION%20ARTICLE.pdf http://eprints.ums.ac.id/45089/1/PUBLICATION%20ARTICLE.pdf Wanda Giovani, Sri Wuli Fitriati/EEJ 8 (3) 2018 331 - 341 341 Hamilton, H.E. eds. (2001). The Handbook of Discourse Analysis. Oxford: Blackwell Publishing. Herring, S. C. (2008). Language and the Internet. The International Encyclopedia of Communication. Retrieved from https://onlinelibrary.wiley.com/doi/abs /10.1002/9781405186407.wbiecl005 Liddicoat, A. J. (2007). Internationalising Japan: Nihonjinron and the intercultural in Japanese language-in-education policy. Journal of Multicultural Discourses, 2(1), 32- 46. Moss, Simon. (2016). Conversation Analysis. At http://www.sicotests.com/psyarticle.asp? id=214 Nordquist, Richard. (2015). Adjacency Pairs (Conversation Analysis). At https://www.thoughtco.com/adjacency- pair-conversation-analysis-1688970 Nuckolls, K. (2005). IM communicating: a conversational analysis of instant message conversations (Doctoral dissertation, Portland State University). Paltridge, Brian. (2006). Discourse Analysis: An Introduction Bloomsbury Discourse Continuum discourse series. Sydney: A&C Black. Psathas, G. (1995). Conversation analysis: The study of talk-in-interaction (Vol. 35). California: Sage Publications. Schegloff, E. A., & Sacks, H. (1973). Opening up closings. Semiotica, 8(4), 289-327. Retrieved from https://doi.org/10.1515/semi.1973 .8.4.289 Schegloff, E. A., Jefferson, G., & Sacks, H. (1977). The preference for self-correction in the organization of repair in conversation. Language, 361-382. Retrieved from https://www.jstor.org/stable/413107?seq =1#page_scan_tab_contents Yule, G. (2006). The Study of Language. Third Edition. Cambridge: Cambridge University Press. Zaferanieh, Elaheh. (2012). Conversation Analysis of Online Chat. Tehran: Islamic Azad University. Retrieved from https://s3.amazonaws.com/academia.ed u.documents/37034237/conversation_an alysis_of_online_chat.pdf?AWSAccessKe yId=AKIAIWOWYYGZ2Y53UL3A&E xpires=1528194289&Signature=NntbBge xdtP7Nrl1HD5O8dhawD4%3D&respons e-content- disposition=inline%3B%20filename%3D Conversation_Analysis_of_Online_Chat. pdf http://www.sicotests.com/psyarticle.asp?id=214 http://www.sicotests.com/psyarticle.asp?id=214 https://www.thoughtco.com/adjacency-pair-conversation-analysis-1688970 https://www.thoughtco.com/adjacency-pair-conversation-analysis-1688970 https://www.google.co.id/search?hl=id&tbo=p&tbm=bks&q=bibliogroup:%22Bloomsbury+Discourse%22&source=gbs_metadata_r&cad=6 https://www.google.co.id/search?hl=id&tbo=p&tbm=bks&q=bibliogroup:%22Continuum+discourse+series%22&source=gbs_metadata_r&cad=6 https://doi.org/10.1515/semi.1973.8.4.289 https://doi.org/10.1515/semi.1973.8.4.289