BACKCHANNELS IN THE INTERACTIONS OF INDONESIAN L2 SPEAKERS OF ENGLISH IN AUSTRALIAN ACADEMIC CONTEXT

This study investigated backchannels, short verbal responses such as yeah, employed by Indonesian L2 speakers of English in the interactions with L1 speakers of Australian English in Australian academic setting. The naturallyoccurring dyadic interactions were audiotaped and scrutinised in a sequential analysis of conversation analysis. The examination was aimed to scrutinise the pragmatic functions and the placement of backchannels within the sequential organisation of the interactions. The findings showed that they used backchannel responses involving non-lexical items, lexical items, and combinations of lexical and non-lexical items. Backchannel responses existed in different linguistic environments that may be culturally specific such as after you know. Backchannels were used to show attentiveness, agreement, and comprehension of the speaker’s talk. Besides, they also employed backchannels to express politeness to satisfy the supervisors’ positive face in the interactions. In this study, they appear to converge their linguistic devices to that of their supervisors.


INTRODUCTION
In face-to-face interaction, an addressee may provide a variety of short responses to the speaker who speaks for the extended period. Short verbal responses or backchannels are employed to convey the addressee's feedback. Backchannels include vocal expressions such as uh-huh, yeah, and okay as well as head nods, and eye blinks (Bangerter & Clark, 2003;Schegloff, 1982). They show that the addressee of an interaction is not the inactive recipient of the given information; instead they correspondingly 'try to keep speakers informed of their current state of understanding' (Clark & Krych, 2004: 66). Besides, backchannels "are central to conversational success, demonstrating the producer's active participation in not just turn taking, but in the development of the speaker's talk.' (Tolins & Tree, 2014, p. 153).
There are only a few of them that studies backchannels used by the second language (L2, henceforth) speakers. Scholars have examined backchannels involving Mandarin speakers with English as the dominant language and Mandarin speakers with Mandarin domination (Tao & Thompson, 1991). White (1997) examined differences in the use of backchannels, repair, pausing, and private speech between American and Japanese. Stubbe (1998) investigated backchannel behaviour of Maori and Pakeha ethnic groups in dyadic interaction in New Zealand English. Heinz (2003) studied backchannel behaviour of American English, monolingual Germans and bilingual Germans. Cutrone (2005) examined the use of backchannels in dyadic interaction between Japanese and British. Suprapto (2012) investigated aizuchi, short responses in Japanese, used by Indonesian L2 speakers of Japanese. A study from Cutrone (2014) examined the use of backchannels in the interactions of Japanese English foreign language learners of English and Americans. Yazdfazeli and Motallebzadeh (2014) examined Iranian EFL learners' backchannel performances. Barron and Black (2015) investigated listener responses in telecollaborative interactions between German native speaker learners of English and Irish English speakers.
Studies on backchannels used by speakers of English as a second/foreign language have been examined in different settings and social contexts. Backchannels employed by Indonesian L2 speakers of English in intercultural interactions seem to be unexplored by scholars, regardless of functions in interactions. The present study is to scrutinise short verbal responses, backchannels, in the context of the dialogue between Indonesian L2 speakers of English and their supervisors in intercultural interactions in Australian academic context. It examines the placement of backchannels within sequential organisation of the interactions and the functions of backchannels in intercultural interactions between Indonesian L2 speakers of English and L1 speakers of Australian English. It makes an effort to provide an account of backchannels and their placement together with their functions in the interactions of Indonesian L2 speakers of English and L1 speakers of Australian English.  Gardner (2001, p. 2) states that some of the functions of backchannels include: "continues that function to hand the floor back to the immediately prior speaker (e.g. mm hm, uh huh); acknowledgments that claim agreement or understanding of the previous turn (e.g., mm, yeah)". Similarly, White (1989) argues that the use of backchannels in interaction shows that the addressee agrees with what has been said by the speaker and that the addressee gives a signal to the speaker to continue the talk. Maynard (1997, p. 46) categorises the functions of backchannels as continuer, understanding, agreement, support, strong emotional answer, and minor additions.

Backchannels in L2 interactions
In his study of the Japanese backchannel aizuchi, White (1989) showed that the Japanese did not only employ aizuchi to agree with the speaker, but also to show empathy to the speaker. Using frequent backchannels in Japanese interaction is perceived to be polite as they show the addressee's interest in what the speaker is saying. Cutrone (2005) showed that Japanese EFL speakers (JESs) showed politeness by using backchannels even though they did not understand and agree with what the speaker had said. Cutrone (2014) revealed that, in the intercultural interaction between JESs and native English speakers, JESs used a greater number of backchannels because this helped them to feel comfortable in their role as listeners. They sometimes pretended to understand or to agree with the speaker as a way of maintaining the conversation's pleasantness. Svennevig (1999) maintains that the speaker and hearer are being polite in the interaction by showing attentiveness and alignment using short responses or backchannels.
An interlanguage pragmatic study conducted by Tao and Thompson (1991) investigated backchannels in Mandarin conversations in which the subjects were Mandarin speakers with English as the dominant language and Mandarin speakers with Mandarin domination. The results showed that there was a language transfer or inference from the second language to the first one. The Mandarin with English dominant language speakers frequently used backchannels that did not exist in the interactions of Mandarin speakers with Mandarin domination. They used backchannels more frequently both during and at the end of the other party's speaking turn, and their use of backchannels was predominantly as continuers. Mandarin speakers with Mandarin domination used backchannels, especially at the end of the speaker's turns and they infrequently used backchannels in overlap with the speaker's turn. They used them to show understanding, confirmation, and acknowledgment of agreement. Another study on backchannels was conducted by Heinz (2003). She investigated the differences in the behaviour of American English and German speakers in using backchannels and the behaviour of monolingual Germans and bilingual Germans. Her study revealed that backchannel responses and overlapped backchannels were less frequently produced by monolingual Germans than those by monolingual Americans. A pragmatic transfer occurred when the bilingual Germans communicated in German; they used backchannels and overlapped backchannels more than the monolinguals.
Similarly, Suprapto (2012)  between Japanese and Indonesians. The findings showed that the frequency of using aizuchi by Indonesian L2 speakers of Japanese was less than that of L1 speakers of Japanese. The functions of aizuchi involved showing information receipt, continuers, support, agreement and strong emotional response. Aizuchi was also used to ask for information, to add information, and to provide a correction. In using aizuchi, there was less lexical variation, low frequency of use, and irrelevant moments of producing aizuchi that made the Japanese counterpart confused. Suprapto argued that cultural norms influenced the way Indonesian L2 speakers of Japanese used aizuchi.

An overview of Politeness
The use of backchannels is associated with the notion of politeness theory. The notion of politeness theory itself was developed by Levinson (1978, 1987). The theory purports to be a universal model of linguistic politeness. They argued that politeness is accomplished linguistically using diverse strategies across cultures (Félix-Brasdefer, 2008, p. 17). The principal contribution of Brown and Levinson's theory is their effort to associate politeness with the notion of the face in social encounters. Maintaining face means to represent an image that a speaker wishes the hearer to perceive. It is the task of the participants in communicative events to keep and care for each other's face (Bell, Arnold, & Haddock, 2009). Brown and Levinson (1987, p. 61) defined face as "an individual public self-image that every member wants to claim for himself." When a speaker performs an act that potentially challenges her/his face, s/he will soften the imposition by employing politeness strategies (p.59 -60).
Face consists of two specific kinds of wants. Positive face is "the positive consistent selfimage or 'personality' (crucially including the desire that this self-image is appreciated and approved of) claimed by interactants" (Brown & Levinson, 1987, p.61). Negative face is "the basic claim to territories, personal preserves, rights to non-distractionthat is, to freedom of action and freedom from imposition" (Brown & Levinson, 1987, p.61). The notion of face and the social need to position oneself in the interaction are claimed to be universal. In addition, people are commonly cooperative in preserving their face in communication that is based on the mutual vulnerability of face. Thus, every interactant's face is dependent on every other interactant's face in the interaction. So, it is in every interactant's best interest to preserve each other's face (Brown & Levinson, 1987, pp. 61-62).
Brown and Levinson's politeness theory is primarily concerned with politeness strategies to maintain face. Their politeness involves the speaker's intention and the linguistic expression used to convey politeness. The proposed politeness strategies focus on the directness of the expressions that aim to avoid conflict through linguistic interaction as a certain act is considered to inherently threaten the addressee's face. In addition, the speaker should take into account three social variables in performing a face-threatening act (FTAs). Their concept of face, social variables, and the notion of the universality of politeness are under discussion by scholars examining politeness. Regarding backchannel responses, Brown and Levinson (1987, p. 113 and p. 129)  showing attentiveness, and understanding of the talk. Kitamura (2000) states that in interactions the participants play their role, and when they are actively involved in the interaction, they enliven the interaction. By showing engagement or involvement by using short responses in the interaction, the participants comply with the speaker to satisfy their positive and negative face.
Many scholars counter their arguments about the notion of politeness, but there are also many scholars who endorse their politeness theory. Non-Anglophone researchers have found that the concept of face is not appropriate to their cultures ( Bargiela-Chiappini (2003). Similarly, the notion of the face does not represent the Japanese concept of wakimae (discernment) (Ide, 1989;Matsumoto, 1989). Brown and Levinson's model does not match the concept of face in Chinese data (Gu, 1990, pp. 241 -242). Studies of some scholars showed that there is a correlation between social indices and politeness strategies (Benham & Niroomand, 2011;Morgan, 2010;King, 2011).

RESEARCH METHODOLOGY
The present study focused on backchannels in the interactions between Indonesian L2 speakers of English who were postgraduate students and L1 speakers of Australian English who were the supervisors. Their interactions that occurred naturally were audiotaped. Naturally-occurring interactions demonstrate mundane language use. Audiorecording data are purposely employed to preserve the communication as they enable the researcher to have the recordings that present different layers concurrently (Mondada, 2012, p. 306). The strength of naturally-occurring data is in "its validity in reflecting the actual talk-in-interaction" (Rue & Zhang, 2008, p. 35). By using audiotaped data the researcher can scrutinise diverse elements of oral interactions. Besides, an in-depth examination of the strategies employed can be done repeatedly. Thus, oral features in the interactions can be examined in great detail.
Self-selected participants comprising seven postgraduate students and their supervisors were involved in this research. Dyadic interactions between the student and the supervisor were audio-recorded for one session. The length of each session was dependent on the interaction between the students and the supervisors.
The collected data were then transcribed. The data were analysed grounding in a sequential analysis of Conversation Analysis (CA henceforth). According to (Hutchby & Wooffitt, 2008, p. 14), CA principally focuses on revealing the way the speakers understand and provide a response in their interaction. Besides, CA examines in what way "single utterances are intrinsically related to the utterances that precede them and the utterances that come after them" (Peräkylä, Antaki, Vehviläinen, & Leudar, 2008, p. 13).
Turn-by-turn examination procedure as the primary principle of CA was applied in this study. This process displays how the speakers or the interlocutors exhibit their comprehension of the interlocutor's previous utterance through their reply. It is the preceding utterance that creates the context for the subsequent utterance being constructed in the interaction, (Heritage, 2004, p. 109 Each transcript was scrutinised turn-by-turn using Sack's (1987, in Drew, 2013 'nextness' notion. When a phenomenon of backchannels was recognised, it was coded, classified into categories, and ordered sequentially.

RESEARCH FINDINGS
Indonesian L2 speakers of English used backchannel responses in their L2 interactions involving non-lexical items such as oh, ha, hmm yea, umm, yea aah, yeah, yeap, aha aha, and the lexical items included okay, right, all right exactly, okay, and yes. They also used combinations of lexical and non-lexical items such as yea yeah exactly. These backchannel responses were employed to express their attention, agreement, receipt of the information, support of the delivery of the information, and comprehension of the supervisors' talk. Brown and Levinson (1987, p. 113 & p. 129) suggest that backchannel responses express 'brief agreement' and satisfy the interlocutor's positive-face by showing attentiveness, and understanding of the talk. In other words, showing attentiveness, agreement, and comprehension of the speaker's talk, the student participants also show that they satisfied the supervisors' positive face in the interactions.
The following extracts show instances of backchannel responses and their placement or linguistic environments. The functions of backchannels employed are described consecutively.

Hesitation Pauses
Extract Two backchannel responses were used by an Indonesian participant to respond to the supervisor's talk in Extract 107. These two backchannels occurred during hesitation pauses in the supervisor's talk. The first one, Line 149, took place after a preposition, while the second one was after a verbal auxiliary, Line 152. They were used to signify the receipt of information and to encourage the delivery of the information. These uses show that they satisfied the supervisor's positive face. showed his understanding of what the supervisor said before. Expressing understanding of the talk exhibits that the Indonesian participant satisfied the supervisor's positive face.

_________________________________________
Extract 1 -Extract 4 shows that Indonesian L2 speakers of English participants employed backchannels that function to signify the receipt of information and to encourage the delivery of the information, to show understanding of the information given by the supervisor, to show the acceptance of what had been said. These functions correspondingly convey that they fulfilled the supervisors' positive face Brown and Levinson (1987, p. 113 and p. 129).
The examples of backchannels in the previous extracts also explicate that Indonesian L2 speakers of English participants used backchannels in four different environments involving hesitation pauses (Hes), rising intonation (Ris), clausal boundary (Clau), and you know (Ykn). They mostly used backchannels after clause boundaries. The percentage of the occurrences of backchannels in the intercultural interactions is shown in Figure 1.

Figure 1
The percentage of backchannels in ELF interactions Figure 1 shows that Indonesian L2 speakers of English participants produced backchannelling after hesitation pauses, rising intonations, and clause boundaries, and after you know produced by their supervisors. They employed the varied use of lexical and non-lexical items for backchannel responses mostly after clause boundaries for 74%. They correspondingly employed backchannels after you know in turn-final positions. Backchannels, after you know, were used only a relatively small number of times (2%), but were found in the data of IS1, IS3, and IS4.

Backchannelling to express politeness
Backchannelling and turn-taking procedures can express politeness between the speaker and the addressee in the interaction (White, 1997). The use of backchannelling in the interaction of the Indonesian L2 speakers of English participants and their supervisors displayed their uptake of the content of the talk. The backchannel responses also showed explicit understanding, agreement, and collaboration in a moment-by-moment talk (Clark & Krych, 2004;Schegloff, 1982, White, 1989. Backchannelling used in the interactions also conveyed attentiveness to the supervisor's talk or a positive politeness device (Brown & Levinson, 1987) to contribute to the interaction.
The occurrence of backchannelling after hesitation pauses by Indonesian student participants showed that they did not deliver the backchannel responses at the end of a turn-constructional unit, a transition relevance place (TRP, henceforth). TRP is where the turn-change from one speaker to another commonly occurs (Sacks, Schegloff, & Jefferson, 1974). Sacks et al. (1974) did not explicate how TRP is realised linguistically in the interaction, but the constructional unit is defined syntactically in terms of sentences, clauses, phrases, and words. However, the production of backchannelling after hesitation pauses often occurs when a speaker has not performed an utterance thoroughly and is followed by a pause. The Indonesian L2 speakers of English participants used backchannelling to provide a signal to the supervisors to continue with the delivery of the information in progress (Lambert, 2011, p. 13). Thus, the backchannelling employed was supportive and cooperative as it did not gain the floor and kept it for a while with the supervisor. Furthermore, the backchannel responses from the addressee depend on the nature of the dialogical interaction. They show the primary role of the addressee as the recipient and collaborator of the exchange in the interaction who may project further direction of the talk (Gardner, 2001). Svennevig (1999) argued that showing attentiveness and using self-oriented comments to show alignment to the speaker in an interaction are ways of being polite in the interaction. In this case, politeness is the central element related to the use of backchannelling (Heinz, 2003) that displays polite listenership (Lambertz, 2011) as well as the preference of positive politeness strategies (Brown & Levinson, 1987).
In brief, using backchannels the Indonesian L2 speakers of English participants showed attentiveness, cooperation, and involvement with their supervisors' immediate preceding talk. According to Brown and Levinson (1987, p. 113), the backchannel responses expressed 'emphatic agreement' to the supervisor's talk. To this end, they expressed camaraderie and saved the supervisor's positive face (Brown & Levinson, 1987).

Accommodating L2 communication strategies
Indonesian L2  English. This occurred when they had to communicate in English in the Australian context. The culture where the interactions occurred seemed to shape their backchannelling. Heinz (2003Heinz ( , p. 1137) maintains that cultural and language systems determine the differences in backchannel behaviour. The differences were in the form of non-lexical and lexical items, occurrences, and the purposes for which backchannelling was used. Deng's (2008, p. 311) cross-cultural study on backchannelling showed that Australian speakers produced a higher percentage of lexical expressions as backchannel responses. His finding suggests that the Australians expect the use of backchannel responses that are likely to be lexically contentful in their interactions. Furthermore, Deng (2008) suggested that the use of a large number of backchannels as a conversational style showed Australian speakers' preference for using positive politeness strategies and expressing solidarity. Failing to provide backchannel responses may result in the interpretation of a lack of cooperation and involvement.
Similarly, Tao and Thompson (1991) investigated the backchannelling behaviour of L1 speakers of Mandarin Chinese for whom English was their dominant language. Their study revealed that the participants uttered more English backchannel responses than the Mandarin Chinese when speaking Mandarin Chinese. Using a sociolinguistic approach, Hymes (1972) suggests that language norms and rules usually govern how individuals should interact with each other in particular situations and cultural settings. Indonesian L2 speakers of English participants used their available repertoire of backchannel responses in English, such as after you know, that is not employed in Indonesian interactions. The linguistic environment of the clause-final you know is specific to L1 speakers of English. It is an addressee-oriented pragmatic device which is used to ensure the 'taken-for-grantedness' of what is being uttered (Coates, 1989, p. 117 in Stubbe & Holmes, 1995. The use of the clause-final you know asks for the interlocutor's cooperation and acknowledgment of the topic. The productions of you know in the turnfinal position by their supervisors invited Indonesian L2 speakers of English participants to utter backchannel responses. They took up the signal from the use of you know by the supervisors and provided the assurance using backchannel responses. This shows that they seemed to accommodate the way backchannel responses are used in an Australian context (Giles, Coupland, & Coupland, 1991). Giles, Bourhis, and Taylor (1977) maintain that speakers tend to converge or diverge their linguistic codes either for power or for social approval.
The results of backchannelling in this study were in contrast to Suprapto's (2012) study on aizuchi that revealed that Indonesian L2 speakers of Japanese used aizuchi at irrelevant times and with limited lexical variation. The difference might be due to the context of interaction as Tannen (1986( , in Lambertz, 2011 suggested that the context of interaction can greatly influence the frequency and use of backchannels.

CONCLUSION
The Indonesian L2 speakers of English participants in this study used backchannel responses to show the participants' role as listeners who were attentive, supportive, and The participants conveyed their politeness by using backchannelling. They used backchannel responses to show cooperation and attentiveness to the supervisor's talk. The backchannel responses existed in different linguistic environments. They used varied backchannels in their intercultural interactions that may result from converging their language to that of their supervisor and the language community in which they were then operating.
This study may contribute to the body of interlanguage pragmatics, English as a lingua franca, and politeness literature. In the intercultural interactions, there may be unintentional impoliteness that resulted from a lack of L2/FL pragmatic awareness.
Converging their language to that of their supervisor and the language community in which they were then operating may not be perceived to be very blunt and impolite in the L2/FL cultural norms. This study informs us about the nature of academic intercultural interactions and helps us to understand how backchannels play a role in expressing politeness strategies used. Further research on backchannelling may focus on the use of backchannels in online interactions and relate it to intra-and intercultural, and ELF interactions.

Conventions used in this study
/ Pause which is less than one second // Pause which is more than one seconds and less than five seconds /// Pause which is more than five seconds [ Overlapping utterances ] The point where the overlap stops = Latching (x) Indecipherable speech which shows approximate syllable -Word truncation or syllable deletion : Lengthen vowels (( )) Non-verbals e.g. coughs, laughs, and sneezes ↑ Rise in intonation ↓ Fall in intonation → The focus of the analysis ,

Continuing intonation
The author is a lecturer in the Department of English Education at the University of Riau, Indonesia. She holds a PhD from Curtin University. Her main fields of interest include applied linguistics, pragmatics, politeness, and discourse analysis.