REALIZATION OF THE SPEECH ACT OF REFUSALS AMONG MALAY ESL STUDENTS

Refusing to participate in an action is a negative response that might lead to a dispute and jeopardise the connection between the hearer and the speaker. To avoid the conflict that may arise when a speaker must decline a request

refusal strategies (Beebe et al., 1990) and categorized into the types of sequence orders of the strategies. The findings revealed that the MSE preferred indirect strategies when refusing higher and equal status interlocutors' requests. When using direct strategies, they opted for lesser and the least degree of directness. The sequence orders of the strategies employed to refuse both interlocutors were also similar; immediate refusal and delayed refusal. However, they deferred in terms of their preference between the two types of sequence orders and the specific strategies used in the pre-refusal, head act and postrefusal. These variations show their effort to adapt to the status of their interlocutors; higher and equal status and the given situations. Their use of strategies at the post-refusal for both interlocutors were varied and lengthier compared to the sequence orders revealed in the previous studies. The findings reflect the participants' effort to search for equilibrium between upholding their stance and at the same time remain respectful in the case of their employer and being concerned toward their friend. These findings provide pragmatic input which could be utilized by English language teachers to develop their students' ability to use socially appropriate language for the situation they encounter in the Malaysian context.

Keywords:
Refusal strategies, Semantic refusal Formula, Speech act of request, Open role-play, English as a second language.

INTRODUCTION
Refusal is an act of saying "no" which expresses the non-acceptance of the addressee, declination of or disagreement with "a request, invitation, suggestion or offer" (Wierzbicka, 1987. p. 94). Because a rejection goes against what an interlocutor expects or hopes, it is a disfavoured answer. Brown and Levinson (1978) describe this act as a Face Threatening Act (FTA) as it may cause "shame" to the face of the speaker or the hearer and consequently may jeopardize their interpersonal relationship. Due to this harmful effect, it may carry, many people find it unpleasant and difficult to say no. However, saying 'no' or refusing itself is not a problem. What matters most is how the refusal message is communicated (Rubin, 1983). For instance, if the speaker were too direct in declining his employer's request to work overtime and without giving a valid reason, he or she might not only offend his/her employer but may also jeopardize his/her career. Likewise, if a speaker abruptly rejected a friend's request to help with her homework, then the rejection might ruin their friendship. This shows that communication requires the consideration of contexts which include power, distance, status and situational contexts. As stated by Mey (2001), all speech is situated speech; a speech act is never just an 'act of speech', but should be considered in the total situation of the activity of which it is a part. Dippold (2008) states that to refuse appropriately and at the same time maintain the interpersonal relationship between a speaker and a hearer, interlocutors need to have two components of pragmatic knowledge. The first, pragmalinguistics refers to knowledge of the forms and strategies to convey particular illocutions whereas the second, sociopragmatics relates to knowledge of the use of these forms and strategies in appropriate context. Knowledge of both components will enable learners to determine the situational-appropriate utterance, namely what can be said, where it can be said, and how to say it most effectively (Cohen, 2004). Communicating the "no" message without considering the linguistic knowledge and the sociocultural factors in a situation may result in the speaker performing a face threatening act which may put their interpersonal relationship at stake as well as being regard as impolite (Walaszewska & Piskorska, 2012).
In Malaysian context, Kuang (2009) suggests that declining a request from other parties and in particular people with authority is extremely difficult for most Malaysians. Since authority may imply power, refusal may threaten the speaker's position. The present study focuses on the speech act of refusal as realized by native speakers of Malay (MSE) who were undergraduates of a local university. Specifically, the study aims to ascertain the types of sequence orders of refusal strategies employed by MSE participants when refusing the higher and equal status interlocutor's requests in English.

LITERATURE REVIEW
Communication has become a central activity for us due to the fact that we are sociable creatures who like being around by friends, family, and other people with whom we can share our thoughts and experiences. Nevertheless, communication is not simply a matter of sending and receiving messages. It necessitates appropriate strategies of communicating with others so that participants are at ease with the spoken words and their interpersonal relationships are preserved or improved. In this sense communication requires linguistic expertise as well as a grasp of the situation's social and cultural elements.
The use of appropriate strategies to make others feel comfortable and the emphasis on maintaining relationship in communication is closely related to the concept of politeness. Lakoff (1973) claims that "to be polite is to say socially right things" (p. 53), while Adegbija (1989) links politeness with situations in which a person "speaks or behaves in a socially and culturally acceptable and pleasant to the hearer" (p. 58). In brief, politeness is a set of strategies for doing common communication activities (such as requesting, advising, complimenting, criticising, and reminding) while maintaining one's face i.e. keeping both the speaker's and the listener's faces safe throughout everyday interactions. Likewise, Brown and Levinson (1978) associate politeness with attempts to protect people's feelings, which include the use of "face." The "negative face" is the desire of the person "not to be imposed on," while the "positive face" is the desire of the individual "to be loved and approved of." A face-threatening act usually puts the speaker's or hearer's positive or negative face in jeopardy. To protect the hearers' faces and at the same time also assisting the speakers to achieve their own goals, the speakers tend to change their language usage depending on their interlocutor. Asmah (2000), however, perceives politeness as a trait of Malay culture, which encourages its members to be courteous by treating one another with respect. She also distinguishes the Malay idea of 'face saving,' which is more than merely 'covering one's own or another's face,' as Brown and Levinson view it (1987). Face-saving has broader implications that are not limited to the moment of conversation. Aside from the face of the interlocutor, it also includes the type of education he or she acquired from his or her parents, teachers, and relatives. As a result, a person's good name and dignity are linked to his or her parents' and family's good names and dignity. As a result, Malays are reminded to be cautious while dealing with others, whether in writing or verbally, in order to avoid tarnishing the interlocutor's face. Conflicts may develop if the other person's face is damaged, and family members and friends' faces may be implicated as well.
As can be seen, politeness has different meanings in different cultures, which influences how people communicate. Review of the past studies has shown that indirect strategies are prevalent strategies. This indicates that in general, people desire for politeness but there are differences in the use of strategies due to the influence of culture. Beebe et al. (1990) conducted a study on refusals by Japanese learners of English to examine evidence of pragmatic transfer in the order, frequency, and content of semantic formulas. The study employed a discourse completion test (DCT) which was distributed to 20 Japanese-speaking Japanese, 20 Japanese-speaking English, and 20 American English speakers. They found that the Americans ordered the semantic formulas in the same way for both levels of unequal status (i.e. higher status or lower status) in contrast to Japanese learners who reacted differently to the higher-status compared to the lower-status interlocutors. The findings show that the Japanese learners used expressions of regret (apologies) with higher-status interlocutors more often than with lower-status interlocutors. This appears to be the result of transfer from Japanese since using different semantic formulas in refusals to people of different status is the norm in the Japanese society.
Seven studies involving Malay students' refusal were reviewed. The first five are comparative studies on refusals, all of which used DCT and Beebe et al. (1990)'s refusal semantic formulae. Sattar and Farnia (2014) compared Malays and Iraqis' refusal to invitations in English based on relative power and social distance, while Farnia and Wu (2012) compared Chinese international students' refusal to Malay students' refusal strategies in English when refusing an invitation made by an equal status interlocutor as well as a higher status interlocutor. Al-Shboul et al. (2012) contrasted Jordanian and Malay ESL postgraduate students at a local institution who turned down inquiries, invites, and offers. The three studies found that the groups studied had more similarities than differences in their refusal tactics, such as a preference for indirect techniques and the use of an excuse / reason / justification approach to communicate their refusals indirectly, followed by declarations of remorse. They also discovered that Malay individuals gave more detailed and lengthy replies than their counterparts. Both groups preferred a lower degree of directness, i.e. negative willingness, in terms of the employment of direct approach. The fourth study is by Farhana et al. (2019) who compared Malays and Germans refusing requests made by higher, equal and lower relationship status interlocutor in their own first language. The study revealed that Germans used much more strategies and employed more direct strategies compared to the Malays. Both groups preferred indirectness and the most used strategy from this category was excuse, reason, explanation. However, the Malays tended to use statements of regret particularly apology as part of their refusal strategy more than the Germans. The fifth study by Radhiah (2017) discussed the refusal strategies employed by Malay and Spanish postgraduate students refusing in their mother tongue i.e. Malay and Spanish respectively. Both groups used similar strategies to refuse but vary in regard to frequency and trends. The Spanish employed more direct strategy and tended to start their refusal with a non-performative statement followed by an excuse whereas the Malay tended to apologize initially and then gave an excuse. The content of the reason/excuse given by the Spanish reflected their priority in fulfilling individual desire compared to the Malays who prioritize group importance as evidence in their excuses which specifically related to family matters. The last two studies by Suraya and Nabilah (2016), and by Nur Asyarani (2018) focused on gender differences in refusal strategies among Malay students when refusing in English. In general, both studies found that more indirect strategies were employed compared to the direct strategies indicating their awareness that the use of the indirect strategies lessened the facethreatening effect of the speech act of refusal and that both genders were generally explanative and apologetic when making rejections.
The previous studies show the preference to use indirect strategies much more than direct strategies and the prevalent employment of reason/excuse to justify the refusal. The comparative studies reveal that the Malay students' responses tended to be lengthier, they used less direct strategies but employed more expression of regret compared to the groups compared. All the studies reviewed used DCT to collect their data. However, the present study used open roleplay to elicit data on the distribution order of refusals by MSE in line with findings of Margalef-Boada (1993) which revealed that the refusals data collected via role-plays were richer and more complex, more interactive and more representative of natural data than the data obtained from the DCTs.

METHODOLOGY
This study adopted a qualitative methodology to ascertain the types of sequence orders of refusal strategies employed by 12 Malay native speakers when refusing the higher and equal status interlocutor's request. The small sample size of the participants is in line with Miles, Huberman and Saldana (2020) who maintain that qualitative study normally involves "small samples of people, nested in their context and studied in-depth" (p. 27). The actual study involved 12 Malay native speakers and 12 English native speakers who performed six role-play situations based on three initiating acts i.e. invitation, request and offer and the scope of study comprised comparing the refusal strategies, content of the strategies and distributional order of the strategies of the two groups. However, this article only presents distributional order of the refusal made by 12 Malay native speakers from one initiating act based on two situations i.e. request from a higher status and a lower status interlocutors. Purposive sampling was used to select the participants since it is the best method that can be used in qualitative research to obtain insights or to gain an in-depth understanding by finding participants "who can provide rich and varied insights into the phenomenon under investigation so as to maximize what we can learn" (Dornyei, 2007, p. 126). It is also a common sampling employed in speech act studies (Al Issa, 1998;Al Kahtani, 2008;Al Khatib, 2006;Chen, 1996;Felix-Brasdefer, 2004;Kreishan, 2018;Khadijah, 1993;Margalef-Boada, 1993;Marquez-Reiter, 2000;Saud, 2019;Shareef et al., 2018). The selected participants were Malay first year undergraduate law students whose age ranged from 19-21 years old. The choice of law program students was due to the fact that the program only accepted students with Malaysian University English Test Bands 4 or 5 at the entry point. The reason for selecting students with Band 4 (Competent User) and Band 5 (Good User) was to ensure that the respondents are able to carry out conversation for the role-play without too much difficulty as the focus of the study too was not on students' ability to express themselves but rather on their refusal patterns and strategies.
The main instrument employed was an open role-play. Open roleplays are simulations of social interactions in which participants assume and enact described roles within specified situations. Roleplay was employed since it could provide spoken data that approaches real-life performance. Two open role-plays were enacted by the MSE based on two situations requiring them to refuse the higher and equal status interlocutors. Details such as social distance, power status of the interlocutors and their specific roles in the situation were provided. The participant would have to communicate and negotiate his/her refusal himself/herself spontaneously. Billmyer and Varghese (2000) maintain that equipping the participants with contextualized background information will elicit more elaborated and natural-like data. The first situation required them to decline a request from their employer (higher status interlocutor) to work extra-hour while in the second situation they had to reject their classmate's request to borrow notes (equal status interlocutor). In both situations the interlocutors were distant in terms of social relationship to the participants. Details of the situations are as follows: Situation 1: An employer requesting for an employee to work extra-hour Apart from your regular job, you also work part-time at an established bookstore. The bookstore is open from Sunday to Thursday from 9.00 in the morning to 9.30 in the evening. You work from 7.30 in the evening to 9.30 in the evening from Sunday to Thursday. You get along fine with your boss, but don't socialize together outside work. It is Thursday evening at 9.15 p.m. and your boss has just received a delivery books that had been lost in the mail for three weeks and which need to be on display by Sunday morning. You are finishing an inventory when the boss approaches you and ask you to work extrahour (until 11.30 p.m.) to get the display ready, but you can't stay back to do the job.
Situation 2: A classmate requesting to borrow notes from another classmate You are taking a course on literature this semester. You haven't missed this class even once this term and consider yourself a diligent student. So far you have had a good average in the class, not because it is easy for you, but because you have worked very hard. Among your classmates, you have a reputation for taking good notes. The teacher has just announced that the mid-term exam is next week. One of your classmates, who is taking a class with you for the first time in this term and who has frequently missed the class, but have occasionally done small group work together in the class approaches you for your notes, but you don't want to lend them to her (Adapted and modeled based on previous studies (Norma, 2016;Felix-Brasdefer, 2004;Margalef-Boada, 1993;and Beebe et al., 1990).
The role-plays were recorded and transcribed verbatim. Based on the transcription, refusal strategies and their sequence order were identified and coded according to Beebe et al., (1990) classification of semantic formulas (see Appendix). The semantic formulas obtained were then analysed across full refusal interactions as suggested by Gass and Houck (1999). Thus, the refusal strategies realized by the MSE were analysed as a series of strategies that comprised the entire sequence as suggested by Blum-Kulka, House and Kasper (1989). The sequences are as follows: 1.
Pre-refusal strategies which prepare the addressee for an upcoming refusal; 2.
Head-act or main refusal which expresses the main refusal; 3.
Post-refusal strategies which follow the head act and tend to emphasize, mitigate, or conclude the refusal response.

FINDINGS
The MSE strategies obtained from the role-play interactions were categorized into three categories: direct, indirect, and adjunct to refusal strategies, as well as by the strategies employed throughout the interaction sequences: pre-refusal, head act, and post-refusal. As shown in Table 1 the participants employed 6 pre-refusals, 12 headacts and substantial strategies particularly at the post-refusal stage i.e. 94 strategies to refuse employer's request to work extra-hour and 87 to decline a friend's request to borrow notes to mitigate the impact of their refusal. In total 112 strategies were used in refusing their superior's request compared to 105 strategies employed in refusing their friend's request for notes.
Analysis of the strategies across the three stages shows that in both situations the participants employed two sequence orders i.e.: 1. Immediate refusal sequence order in which the participants expressed the refusal (head-act) immediately after the request was made.

2.
Delayed refusal sequence order in which the participants did not turn down the request immediately but they used prerefusal strategies (pre-refusal stage) to prepare the interlocutor for the upcoming rejection.
In refusing their employer's request to work extra-hour, six participants employed delayed refusal sequence order while five used an immediate sequence order while in declining their friend's request, six of them employed the former and another six employed the latter. Although the participants utilized similar sequence orders for both situations, there are differences in the strategies employed.
At the pre-refusal stage, to decline their employer's request, three participants surprised by the request repeated, "Working extra-hour?", "Tonight work till 11.30?" while another three used positive opinion i.e. "I really would like to help but…" In contrast, to decline their friend's request to borrow notes, one participant repeated "borrow my notes?" while another five used request for information i.e. "Why do you want to borrow from me?" "Yeah, I notice you haven't come to class many times. Why did you skip classes?". The participants used different approaches in the two situations: to their employer they were more cordial by making agreeable statement initially whereas to their friend they were rather hostile as indicated by the questions posted in their request for information.
The choice of head act strategy to confirm their rejection of the request also reveals similar approach. To decline their employer's request, three participants used mitigated refusal i.e. "I don't think it is possible for me …), 3 used apology i.e. "I'm sorry…", and six used reason i.e. "…I have my regular job, I need to work tomorrow" and "I have to take my wife home from work…". They used reason and apology which belong to indirect category and although mitigated refusal is a direct refusal it was the least direct among the direct strategies. This shows they were cautious and took effort to mitigate their refusal. Conversely to decline their friend's request, they were antagonistic as it can be traced by their choice of strategies. Three used negative willingness which is a one level higher in terms of degree of directness: "I just can't lend you…" while another participant highlighted negative consequence of using his notes "You won't be able to decipher my hand writing.". Five used criticism toward the requester: "Oh not so easy my friend. Anyway why have you missed so many classes? "I think it is not fair, we were in class, listened to the lecturer and took notes. You skipped classes and suddenly you want my notes." Two apologized: "I'm sorry but…" while one gave alternative: "Why don't you refer to the text book and do the exercises we did in classes". The participants employed substantial strategies at post-refusal stage.
To decline their employer's request they employed 94 strategies from 16 types of strategies. Among these strategies, reason (21) and negative willingness (18) were employed substantially followed by apology (18), negative willingness (18) and alternative (8). The reasons given were recycled or elaboration of the one mentioned in their head act. These include: "…You know I have a regular job and I really need a good rest…", "I have to leave on time to fetch my wife at her workplace. It's late at night the surrounding is dangerous for her to wait. There were many foreign workers around,"…my only child is at home I'm worried". The negative willingness commonly used were "I can't stay back", and "I really can't". After giving reasons and staying firm with their stance, they apologized: "I'm really sorry". or "I'm so sorry…". Some also provided alternative: "Can you get someone else to help?" or "Why not I come tomorrow to help you?".
In declining their friend's request to borrow notes, 87 strategies from 13 types of strategies were used. Five salient strategies employed at post refusal were criticism (22) and negative willingness (15) as well as principle, apology and alternative all of which were used ten times. Criticism was used dominantly throughout their interaction due to their friend's truant behavior. As a results, the participants criticized her by saying: "…it's your fault not coming to class then you simply come to me for my notes", and "I've worked hard for it and it's not fair for me to give the notes to somebody that hasn't been coming to class". To add insult to injury, four of them added their principle. Among the principles employed are "I'm the type who do not lend notes to people", and "Haven't you heard this saying to succeed is to work hard, so you don't deserve the notes". They also showed their firmness via negative willingness: "I don't want to give you my notes", "I can't give you my notes", and "I won't share my notes with you". The phrases they used to apologise include," "I'm sorry" or just simply "Sorry" while some provided alternatives such as "You should see the lecturer and explain your problem. Perhaps she can give a replacement class," and "Why not ask someone else?" The strategies used in each of the three stages for both situations demonstrate how difficult it is to categorise refusals. In request to work overtime, it entails a variety of speech acts and reemployment of strategies; refusal, request, apology, suggestion, persuasion, promise and so on for request to work overtime. However, in request to borrow notes it involves the use of criticism, principle, negative consequence, suggestion, apology, reason, and others. As shown in Table 2 and Table 3, the length of interaction among the twelve players varies based on the perseverance of the initiator and the recalcitrance of the participants, as their responses via the role-play were open. The majority of the participants put forth effort (thus used more strategies), but a few were either firm or provided clear reasons that the requesters i.e. the employer and the friend (also known as the initiator) were unable to refute (therefore used less strategies) and hence the initiator relented and accepted their rejection.
P5 employed immediate refusal sequence order. He established his refusal instantly after the request was made by using mitigated refusal (head-act). At the post-refusal stages, P5 established his refusal further by elaborating his reason and reaffirmed using mitigated refusal and followed by an apology. When the initiator pressed on by asking P5 to extend his working hours to 10.30 p.m instead of 11.30 p.m requested initially, P5 became assertive and rejected the request via the second highest degree of the direct strategy; flat no then mitigated it by apologizing but reaffirmed his refusal using negative willingness.
Once his stand was clear P5 adopted self-effacing approach by recycling his first reason. This time P5 impressed upon her the promise he made to his son and request for her understanding by explaining how upset his son could be if the promise was not met. When the initiator made a sad comment having to work alone, P5 apologized twice and still reaffirmed his refusal via negative willingness and expressed appreciation by thanking her twice. Head-act: P5 refused the request through mitigated refusal (1). Post-refusal: The refusal was followed by an elaboration on the reason (2), reaffirmation of his refusal via mitigated refusal (3) and an apology (4). The initiator pleaded and P5 responded with a firm flat no (5) and negative willingness (6).
The initiator tried to get P5 to compromise but P5 was adamant and refused with another flat no (8), gave reason (9)

P5
Ah okay, thank you thank you very much(13).
P10 employed a delayed refusal sequence order. She delayed her refusal by employing a pre-refusal strategy; request for information on the reason for not attending class. After hearing her flimsy excuse, P10 established her refusal by criticizing (head-act) her truancy behavior.
At the post refusal stages, P10 confirmed her refusal further by setting condition for past acceptance. When the initiator still did not accept her refusal, P10 turned to a more assertive approach. P10 insulted (criticism) the initiator then reinforced her refusal via mitigated refusal.

DISCUSSION AND CONCLUSIONS
The findings reveal that the MSE participants preferred indirect strategies when refusing the higher and equal interlocutors' requests. This finding corresponds to studies by Sattar et al. (2011), Al-Shboul et al. (2012, Farnia and Wu (2012), Sattar and Farnia (2014), Suraya and Nabilah (2016), Radhiah (2017), Asyarani (2018) and Farhana et al. (2019). However, their finding that their participants used many similar strategies, i.e. employing ample reasons and apologies for both higher and equal status interlocutors, contradicts the findings of the present study that the MSE differ in their types of indirect strategies employed in refusing the two interlocutors. The difference in the indirect strategies can be attributed to the situations given for students to respond and the difference in terms of methods used to collect data; the previous studies employed DCT whereas the present study used open role-plays. The present study found that when refusing the higher interlocutor, the MSE use substantial reasons and apologies. Their apologies were used with intensifier; so sorry, very sorry and really sorry. For instance nine participants i.e. P2, P4, P5, P6, P7, P8, P9, P10 and P11 used apology. From the 18 apologies employed, 12 apologies were used with intensifiers either so or very or really. In terms of directness, if they refused immediately, they employed low degree of directness i.e. either negative willingness or mitigated refusal. While those opted for delayed refusal made an agreeable statement (positive opinion -"That's a good idea but…") before declining the request using reasons or apology or low degree of directness i.e. mitigated refusal. At the post refusal stage, they used ample reasons and enhanced their apologies by using intensifiers but at the same time remained firm by using low degree of directness. Content wise the reasons given by the participants who cited family related matters reflect the cultural values of the Malays i.e. prioritizing group importance. The study by Radhiah (2017) also had the same finding.
On the contrary, when refusing their equal, the MSE used criticism substantially and apology moderately. Their apologies were expressed just the word "sorry" without any intensifier. Their responses towards the equal reveal a blunt approach. Those using an immediate refusal order employed criticism or plain apology ("sorry") as the head-act while those using delayed refusal questioned (via request for info.) the equal interlocutor. At the post refusal stage, their interlocutor was given a long lecture (via criticism) on the importance of coming to class and taking one's own notes. Beebe et al. (1990) stressed that refusals are sensitive to sociolinguistic variables such as power variable and the different approaches employed by the MSE in the two situations have shown this clearly.
In the request to work extra-hour, the power factor is dominant. The interlocutor was their supervisor; hence, they employed more selfeffacing strategies i.e. provided ample reasons, apologize more and attached intensifiers to their apologies to heighten its effect and also requested her understanding and showed her that they care by giving an alternative but at the same time used direct strategies to indicate their stance clearly. Hence, they applied more face saving strategies. In the request to borrow books, their interlocutor was their classmate i.e. their equal, thus she had no authority on them and she was also not close to them. Moreover, the situation dealt with truant behavior. These factors led the MSE to be more explicit in their refusal as their bluntness would not have any impact on them. Their forthrightness also indicates their concern for their friend's misbehavior. Their maneuvering of the strategies is congruent to the statement by Brown and Levinson (1987) that people usually attempt to preserve the hearer's face while also aiding the speakers to achieve their own objectives. Hence speakers often vary their language use based on the interlocutor. Their cautious attitude toward their employer might be related to Eastern ideals such as being polite and careful of not dishonouring one's integrity, and that their own behaviour will reflect their own upbringing (Asmah, 2000).
In terms of accuracy and clarity of communication purposes, the participants' employment of direct methods conforms to Brown and Levinson's on record strategy (1987). Participants may prevent misunderstandings or being perceived as manipulators by being forthright, and their rejection message is clearly delivered. When it comes to declining their employer's request, the power variable has caused participants to utilise the lesser (negative willingness) and least degree of directness to lessen the degree of directness (mitigated refusal). The limited use of the high degree of directness (flat no and performative) to decline the higher status interlocutor, according to Felix-Brasdefer (2004), is due to the fact that it presents a larger threat to the hearer's positive face. Likewise, in the request to borrow notes, the MSE maintained their preference for a low degree of directness, preferring negative willingness and attenuated refusal over flat no and performative at the very least. The MSEs' forthright attitude, on the other hand, is evident in their use of indirect technique i.e. criticism and less use of apologies, which Brown and Levinson (1978) characterised as a typical practise among friends and family members.
The order of the strategies also corresponds to Margalef-Boada's (1993) findings that refusal data collected through role-plays is richer and more complex than refusal data collected through the DCT, making it more interactive and representative of natural data. Analysing the interaction at the discourse level allows us to understand how refusals are sequenced and the types of responses utilized (Gass & Houck, 1999). The produced variance in distribution orders can be utilised for educational purposes. The resources, for example, can be used as practical input to improve ESL students' capacity to recognise illocutionary and perlocutionary consequences of words spoken in a specific context when the setting is explicitly stated.
It is hoped that discussions generated from the analysis of the patterns of refusal may be able to help students to be better communicators as they become more aware of the social dynamics at play when they interact. As a result, they will grow more sympathetic and learn to reduce the face-threatening behaviour when they have to perform such an act (as in the case of refusal) and at the same time are able to maintain or enhance their interpersonal relation with the interactors. The present study's employment of enhanced open role-play and the analysis of the refusal interaction at discourse level have produced rich patterns of refusal and politeness strategies which evolved from the beginning until the resolution of the interaction. These patterns of refusal and politeness strategies help us to understand the process involved in refusing. Hence, it is suggested that similar studies should be conducted with a high number of participants to confirm the patterns that have emerged from this study. Another suggestion is for the researcher to supplement the open roleplay with another approach, such as interviews after the roleplay, in order to gain a better understanding of the metacognitive process and to corroborate the impression provided by the role-play.