Abstract
This is a pragmalinguistic investigation into the speech act of refusing as made by Yemeni learners of English as a foreign language. For this study, 20 Yemeni learners of English were asked to respond in English to six different situations in which they carry out the speech act of refusal. Their English performances were compared to those of Yemeni Arabic native speakers and American English native speakers in order to find out whether the refusal given by the group in question, i.e., Yemeni learners of English, correspond more closely with those of the Yemeni Arabic native speakers or with speakers of the target language, the American English native speakers. The data, collected from a Discourse Completion Test (DCT), were analyzed in terms of semantic formula sequences and were categorized according to the refusal taxonomy by Beebe, Takahashi, and Uliss-Weltz (1990). Results indicate that although a similar range of refusal strategies were available to the two language groups, cross-cultural variation was evident in the frequency and content of semantic formulas used by each language group in relation to the contextual variables, which include the status of interlocutors (higher, equal, or lower status) and eliciting acts i.e., requests, invitations, offers, and suggestions). For instance, Yemeni Arabic native speakers tended to be less direct in their refusals by offering preceding “reasons” or “explanations” (in the first position of the semantic formula order) other than their own desire in refusing. American English native speakers, on the other hand, used different semantic order by preceding “regret” in the first position giving more direct refusals. Due to their high proficiency in English, Yemeni learners of English showed evidence of pragmatic competence of the target language in constructing their refusal styles in three areas: the order in which semantic formulas for refusing were used, the frequency of semantic formula and the content of semantic formulas. However, they at times displayed some of their native speech community norms, falling back on their cultural background when formulating refusals.

Key words: Interlanguage pragmatics, pragmatic competence; pragmatic transfer; speech  act of refusal, Yemeni learners of English.

Introduction
Much of the work in interlanguage pragmatics has been conducted within the framework of speech acts. Speech acts can be thought of as ‘functions’ of language, such as complaining, thanking, apologizing, refusing, requesting, and inviting. Within this view, the minimal unit of communication is the performance of linguistic act. All languages have a means of performing speech acts and presumably speech acts themselves are universals, yet the ‘form’ used in specific speech acts varies from culture to culture. Thus, the study of second language speech acts is concerned with the linguistic possibilities available in languages for speech act realization and the effect of cross-cultural differences on second language performance and on the interpretation by native speakers of second language speech acts (Wolfson, 1989, p.183).

Numerous studies in interlanguage pragmatics have recognized that the learners’ ability to use appropriate speech acts in a given speech act event and to use appropriate linguistic forms to realize this speech act is a main component of pragmatic competence. Fraser (1983) describes pragmatic competence as “the knowledge of how an addressee determines what a speaker is saying and recognizes intended illocutionary force conveyed through subtle attitudes” (p.30). Rintell (1997) also pointed out that “pragmatics is the study of speech acts”, arguing that L2 learner pragmatic ability is reflected in how learners produce utterances in the target language to communicate specific intentions and conversely, how they interpret the intentions which their utterances convey. One of the consistent findings in the empirical studies of speech act behavior is that, although the typology of speech acts appears to be universal, their conceptualization and verbalization can vary to a great extent across cultures and languages. In other words, L2 learners may have access to the same range of speech acts and realization strategies as do native speakers (NSs), but they can differ from in the strategies that they choose. Therefore, it is clear that L2 learners must be aware of L2 sociocultural constraints on speech acts in order to be pragmatically competent.
When second language learners engage in conversations with native speakers, difficulties may arise due to their lack of mastery of the conversational norms involved in the production of speech acts. Such conversational difficulties may in turn cause breakdowns in interethnic communication (Gumperz, 1990). When the native speakers violate speech acts realization patterns typically used by native speakers of a target language, they often suffer the perennial risk of inadvertently violating conversational and politeness norms thereby forfeiting their claims to being treated by their interactants as social equals (Kasper, 1990). Communication difficulties are resulted when conversationalists do not share the same knowledge of the subtle rules governing conversations. Scarcella (1990) ascribes high frequency of such difficulties to the fact that “nonnative speakers, when conversing, often transfer the conversational rules of their first language into the second” (p.338).

The use of rules of speaking from one’s speech act community when interacting or when speaking in a second or a foreign language is known as pragmatic transfer. Uriel Weinreich (1953) says “Those instances of deviation from the norms of either language which occur in the speech act of bilinguals as a result of their familiarity with more than one language, i.e. as a result of language contact, will be referred to as interference phenomena. It is these phenomena of speech, and their impact on the norms of either language exposed to contact, that invite the interest of the linguist” (as cited in Wolfson, 1989, p.141).

What L2 learners must know for successful speech act performance has been presented in a “top-down processing” manner (Kasper, 1984): “Learners first have to recognize the extra-linguistic, cultural constraints that operate in a NS’s choice of a particular speech act appropriate to the context. They also have to know how to realize this speech act at the linguistic level and in accordance with L2 sociocultural norms” (p.3). Cohen (1996 ) terms this “Sociocultural knowledge” as “ speakers ability to determine whether it is acceptable to perform the speech act at all in the given situation and, so far, to select one or more semantic formulas that would be appropriate in the realization of the given speech act” (p.254).

The speech act of refusal
Refusals, as all the other speech acts, occur in all languages. However, not all languages/ cultures refuse in the same way nor do they feel comfortable refusing the same invitation or suggestion. The speech act of refusal occur when a speaker directly or indirectly says ‘no’ to request or invitation. Refusal is a face-threatening act to the listener/ requester/ inviter, because it contradicts his or her expectations, and is often realized through indirect strategies. Thus, it requires a high level of pragmatic competence. Chen (1996) used semantic formula to analyze speech act sets of refusal (refusing requests, invitations, offers and suggestions), and concluded that direct refusal as “NO” was not a common strategy for any of the subjects, regardless of their language background. For example, an expression of regret, common in Americans’ refusals, was generally produced by the Chinese speakers, which might lead to unpleasant feelings between speakers in an American context.

Speakers who may be considered fluent in a second language due to their mastery of the grammar and vocabulary of that language may still lack pragmatic competence; in other words, they may still be unable to produce language that is socially and culturally appropriate. In cross-cultural communication, refusals are known as ‘striking points’ for many non native speakers (Beebe, Takahashi, and Uliz-Weltz 1990). Refusals can be  tricky speech acts to perform linguistically and psychologically since the possibility of offending the interlocutor is inherent in the act itself (Know, 2004). As a face-threatening act, a sensitive pragmatic task and high pragmatic competence concern constructing refusals. As a failure to refuse appropriately can risk the interpersonal relations of the speakers, refusals usually include various strategies to avoid offending one’s interlocutors. However, the choice of these strategies may vary across languages and cultures. For example, in refusing invitations, offers and suggestions, gratitude was regularly expressed by American English speakers, but rarely by Egyptian Arabic speakers (Nelson, Al-batal, and Echols, 1996). When Mandarian Chinese speakers wanted to refuse requests, they expressed positive opinion (e.g., ‘I would like to….’) much less frequently than American English since Chinese informants were concerned that if they ever expressed positive opinions, they would be forced to comply (Liao and Bressnahan, 1996). 

Related Literature
Several major investigations into the speech ac of refusing have been conducted by (Beebe, 1985; Beebe. et al., 1985; Beebe and Takahashi, 1987) cited in Wolfson (1989). The finding of their study (Beebe et al., 1985) demonstrates that Japanese learners of English manifest sociolinguistic transfer in refusals by the sequencing of formulas for refusing the actual frequency in use of formulas, and their specific content. One significant finding was that the status of the addressee is a much stronger conditioning factor in the speech of Japanese speaking both in English and in their native language. An example of the related differences status in the behavior of the Japanese is that, unlike English speaking Americans, they did not apologize or express regret in responses to those of lower position. Additional evidence of status-related differences is manifested in the Japanese responses to invitations from higher-as opposed to lower-status interlocutors.

In contrast, Americans in these situations make a distinction along the lines of social distance by responding in a brief and unelaborated fashion to both higher- and lower-status unequal while offering much longer and more detailed responses to peers. In their analysis of strategies for refusing, they classify refusals into direct and indirect refusals. Direct refusals such as “I refuse” or “no” were found to be used by Americans mainly in response to intimates and status unequal or strangers. Indirect refusals, used by Americans primarily to acquaintances of equal status, included three major strategies which were usually found to be used in sequence at the beginning of a refusal. These were (1) an expression of positive opinion such as “I’d like to,” (2) an expression of regret such as “I’m sorry,” and excuse, reason, or explanation such as “My children will be home that night” or “I have a headache”. Other strategies included a statement expressing a wish to be able to comply with the request, the statement of an alternative, a condition for future or past acceptance (e.g., “If you had asked me earlier….”), a promise of future acceptance (e.g., “I’ll do it next time”), a statement of principle (e.g., “I never do business with friends”) , a statement of philosophy (e.g., “One can’t be too careful”), an attempt to dissuade the interlocutor, a criticism of the request, a request for empathy, a statement letting the interlocutor off the hook (e.g., “Don’t worry about it”), self-defense (e.g., “I’m doing my best”), an unspecific or indefinite reply, a display of lack enthusiasm, and verbal or nonverbal avoidance such as silence or a topic switch, a hedge, or a joke. In another study of refusals as made by Japanese ESL learners at tow levels of proficiency, Takahashi and Beebe (1987) found that low and high proficiency learners differed in the order and frequency of semantic formulas they use.

The lower proficiency learners were also more direct in their refusals than higher-level ESL learners. To investigate the evidence of pragmatic transfer in Japanese ESL learners’ refusals, Beebe, Takahashi, & Uliss-Weltz (1990) compared refusal strategies used by Japanese ESL learners to those used by Americans. They also tested the differences in the order, frequency, and content of semantic formulas used by Japanese and Americans. They found evidence of transfer in all three areas (Beebe et al., 1990). Chen (1996) used semantic formula to analyze speech act sets of refusal( refusing requests, invitations, offers and suggestions) produced by American and Chinese speakers of English. She found that direct refusal was not a common strategy for any of the subjects, regardless of their language background.

There are few empirical studies on speech act behavior involving the Arabic language or even native speakers of Arabic. Umar (2004) studied the request strategies as used by Advanced Arab learners of English as a foreign language as compared to those strategies used by British native speakers of English. He found that the two groups adopted similar strategies when addressing their request to equals or people in higher positions. In this case, the subjects rely heavily on conventionally indirect strategies. However, when requests are addressed to people in lower positions the Arabic sample shows a marked tendency towards using more direct request strategies in performing their request than the British sample.
El-Shazly (1993) studies the request strategies in American English, Egyptian Arabic, and English as spoken by Egyptian second language learners. The results of this study have indicated that there are differences in the requesting strategies used by these groups. The Arab speakers of English demonstrate a high tendency towards using conventional indirectness which depends on the use of interrogatives. Modifiers are also examined among the groups. No differences are found with respect to use of “Upgraders’. “Downgraders”, however, are found to be more frequently used by native Arabic speakers. They display a noticeable tendency to use more than one downgrader in a single utterance. This group is also found to be unique in using religious expressions as downgraders.

Al-Shawali (1997) studies the semantic formulas used by Saudi and American male Undergraduate students in the speech act of refusal. The finding of his study shows that Americans and Saudis use similar refusal formulas except in the use of direct refusal. Saudi and Americans also differ in the use of semantic formulas in the content of their refusals; Saudis are found to use avoidance strategies (e.g., postponement and hedge) or they give unspecified answers.

   This is a sociolinguistic study into the speech act of refusal. This study investigates the strategies of refusal as used by the following subject groups:
-Yemeni learners of English, (YELs)
-Yemeni native speakers of Arabic, (YANSs), and
-American native speakers of English (AENSs)
The intention is to elicit the pragmatic performance of the YELs as compared to those of YANSs and AENSs. By tackling the used semantic formulas with its orders and contents in constructing the refusals styles, this study is intended to address the following questions:

1. When YELs perform the speech act of refusal, are their refusal strategies similar to those used by native speakers of English, AENSs?
2. Does pragmatic transfer occur when YELs make refusals in English?

In this study some of the situations in Beebe, Takahashi, and Uliss-Weltz’s Discourse completion Test (1996) have been reproduced.

Method
Subjects
The participants were 30 graduate students divided to three groups as following:
-Twenty Yemeni native speakers of Arabic (YANSs),
-Twenty Yemeni learners of English (YELs), and
-Twenty Americans native speakers of English (AENSs)
The study included only male subjects because it was conducted abroad, in Delhi, in a context with very few females. The written role-playing questionnaire consists of six situations. The questionnaire on refusal was divided into four categories: refusals to (1) requests, (2) invitations, (3) offers, and (4) suggestions. In each case, the questionnaire was designed so that one refusal will be made to someone of higher status, lower status, or a status equal. The responses of the three groups will be compared to each other to find out to what extent the YELs manipulate their pragmatic competence of the target language to refuse in English.

Procedure
All the subjects were asked to fill out a Discourse Completion Test (DCT) (Appendix A). The DCT is a form of questionnaire depicting some natural situations to which the respondents are expected to respond making refusals. This test was originally designed by Blum-Kulka in 1982 and has been widely used since then in collecting data on speech acts realization both within and across language groups. The questionnaire used in this investigation involves six written situations. They were divided into four groups: two requests, two invitations, one offer and one suggestion. Each type included a status differential: higher, equal, or lower (Appendix B). Each situation could only be answered by a refusal. For the YANSs, the questionnaire was translated into Arabic with the necessary changes in the names of people and places to make them more familiar with the situations.

Data Analysis
The data collected through the Discourse-Completion-Test are analyzed. The analysis was based on an independent examination of each response. The same semantic formulas as employed by Beebe, Takahashi, & Uliss-Weltz (Appendix C) are used. For example, if a respondent refused an invitation to a friend’s house for dinner, saying “I’m sorry, I already have plans. Maybe next time,” this was coded as: [expression of regret] [excuse] [offer of alternative] (Beebe, Takahashi, & Uliss-Weltz 1990, p.57). I then coded the order of semantic formulas used in each refusal. In the above example, [expression of regret] was first, [excuse] second, and [offer of an alternative] third (ibid). The total number of semantic formulas of any kind used for each situation was obtained for each of the three subject groups. Then, I counted the frequency of each formula for each situation and listed them. Finally, the similarities between YANSs and YELs responses and the similarities between YELs and AENSs on the other hand were counted and analyzed.

Discussion of the Results
Table (1) typical order of Semantic Formulas in Refusals of Request 
 Refuser status=Higher

Group	Order of Semantic Formulas
	1	2	3	4
YA	excuse (6) can’t (6) positive opinion (2)	future acceptance (2) excuse (3) alternative (2)
YE	excuse (2) positive opinion (2) regret (4) pause filler (12)	can’t (8) positive opinion (2) future acceptance (2) regret(2) excuse (6)	regret (2) excuse (8) future acceptance (2)	excuse (2)
AE	positive opinion (6) regret (14)	positive opinion (2) can’t (4)	excuse (8)	can’t (4)

  All the three groups used excuses in their refusals of requests. The order in which excuse was used is not the same. It varied according to the social status of the requester as in the analysis in tables (1 & 2).

   According to the data in table (1) the responses of all groups YANSs, YELs, and AENSs slightly differ in the order of the semantic formulas. YANSs used excuses in the first and the second positions of the semantic formulas; the YELs used excuses in all positions, whereas AENSs used excuses only in the third position. In higher status, the YANSs refusals tended to be more direct than the other two groups. Three responses by the YANSs included direct refusal “can’t” in the first position. The other two groups preferred to use the direct form of refusal “can’t” in the second positions by YELs and in fourth position by AENSs. On the other hand, YELs and mostly AENSs used regret “sorry” to start their refusal styles. YANSs tended to be briefer than the other two groups who extended their strategies to three and sometimes four parts.

Table (2) typical order of Semantic Formulas in Refusals of Request  
 Refuser status=lower

Group	Order of semantic formulas
	1	2	3	4
YA	excuse (12) positive opinion (6) regret (2)	Future acceptance (2) can’t (8) regret (2)	excuse (2)
YE	regret (16) excuse (2) empathy (2)	can’t (8) excuse (10) future acceptance (2)	excuse (4)
AE	regret (14) positive opinion (2) excuse (2) empathy (2)	excuse (8) alternative (4)	alternative (4)

In table (2) where the refuser has a lower status, the YELs showed a mixture of pragmatic transfer and pragmatic competence. Pragmatic transfer occurred by the use of the direct refusal “can’t” in the second position by four respondents in each group i.e., YANSs and YELs. On the other hand, none of the AENSs responses included direct refusal in any position of the semantic formula. But in the first position we have something different. From the results in table 2, we find that regret “I’m sorry” was used by most of the YELs and AENSs respondents. This means that the YELs have used the same refusal strategies of the AENSs in refusal. Again with excuse expression, while the YANSs used this expression in the first position, we find that both of YELs and AENSs postponed their excuses to the second position which gives another hint of pragmatic competence of the YELs.

Table (3) typical order of Semantic Formulas in Refusals of Invitations 
Refuser status=Higher

Group	Order of semantic formulas
	1	2	3	4
YA	excuse (14) positive opinion (6)	excuse (6) alternative (6)
YE	regret (14) positive opinion (4) pause filler (2)	excuse (10) can’t (4) positive opinion (2) future acceptance (2)	future acceptance (2) excuse (4) regret (2)
AE	positive opinion (4) regret (6) Gratitude (4) “no” (4) empathy (2)	excuse (10) regret (4)	excuse (4) regret (2)	alternative (2)

   According to the results in table (3) YELs and AENSs tended to be more similar by using the expression of regret “I’m sorry” in the first position, excuse in the second position and extended their excuse expressions to the third position of their refusal styles. The YANSs did not use any form of regret in their refusal at all. They tried to show politeness through excuse in the first and second positions which is indirect refusal. In comparison of the YELs responses among those of YANSs and AENSs, we find that they tried to follow the strategies used by AENSs rather than their native counterparts. The researcher assumes here that the use of “excuse” and not “regret” by the YANSs respondents in refusing and invitation is yielded to the sociocultural norms of the community. Again the YELs give more inclinations of L2 pragmatic competence.

Table (4) typical order of Semantic Formulas in Refusals of Invitations  
 Refuser status=Equal

Group	Order of semantic formulas
	1	2	3	4
YA	excuse (4) regret (10) no (2)	excuse (8) gratitude (4)	excuse (4)
YE	gratitude (4) excuse (4) pause filler (4) regret (4) positive opinion (4)	gratitude (2) regret (4) excuse (8) positive opinion (2)	excuse (8)
AE	regret (12) gratitude (8)	excuse (14)	gratitude (6) alternative (4)

When they are in equal status, YELs tended to use their native speech community norms of refusal. Table (4) shows that YANSs and YELs usually use the same content and order of the semantic formula when refusing invitations by peers. However; YANSs’ responses in this situation were somehow unique. For example, the over use of excuse by some respondents such as “I’m busy, I have to visit my parents” or “Oh, I’m tied up. I have an appointment with my doctor”. Sometimes YANSs were vague with their interlocutors of the same status. For example, “Tomorrow I have something to do” or “Sorry, next Sunday I’ll be busy”. Generally speaking, in equal status all the three groups have more similarities than in the other status. They might share some of the sociocultural norms.

Table (5) typical order of Semantic Formulas in Refusals of Suggestions  
 Refuser status=Equal

Group	Order of semantic formulas
	1	2	3	4
YA	positive opinion (4) future acceptance (2) excuse (4) no (6) regret (4)	excuse (8) regret (4) gratitude (4)	excuse (6) gratitude (2)
YE	no (6) regret (4) excuse (4) negative willingness (4) gratitude (1)	excuse (12) can’t (4)	negative willingness (2)
AE	excuse (12) no (8)	excuse (10) gratitude (4)	Gratitude (2)

Again in equal status, all the three groups YANSs, YELs and AENSs tended to use the same strategies for refusal. They used ‘excuse’ expressions in the first and second positions without differences, neither in the content nor in the order of the semantic formula. “No” the direct refusal expression was also used by all the groups in the first positions and almost by the same number of respondents. Four YELs used their native norms to express ‘regret’ as YANSs did so. Expression of ‘gratitude’ for example, “thank you” appeared in all positions but in different order.

Table (6) typical order of Semantic Formulas in Refusals of Offer  
 Refuser status=Lower

Group	Order of semantic formulas
	1	2	3	4
YA	excuse (6) regret (6) title (6) can’t (2)	excuse (12) can’t (4)	excuse (4) can’t (2)
YE	regret (10) appreciation (4) pause filler (4) excuse (2)	Negative willingness (2) regret (2) excuse (10) statement of alternative (4) appreciation (2)	negative willingness (2) excuse (6)
AE	regret (12) positive opinion (4) excuse (4)	excuse (10) alternative (6)	alternative (4)

In table (6) where the refuser is of lower status rejected an offer by his boss, higher status., the content, order, and frequency of the semantic formula varied from one group to another. The main finding here is that the responses of the YANSs contain the direct refusal “can’t” in different orders. On the other side, some of the YANSs used the title “Sir”, with their interlocutors as a reference to the latter’s superiority as a politeness illusion. Sometimes the YANSs, extended their excuses in two positions as explained in table (4). From the results in table (6) we find that YELs and AENSs tended to use the same styles of refusal by avoiding directness. However; there are similarities among the three groups. Most of the respondents started their refusals using ‘regret’ expression “I’m sorry”, in the first position and then they gave their explanations or reasons in the second and sometimes third position.

Conclusion
Different cultures have different perceptions and interpretations of appropriateness and politeness. This study is a contribution to cross-cultural understanding in that it identifies cross-cultural and linguistic differences between Yemeni Arabic native speakers and American English native speakers in the speech act of refuse. Learners of a second language and in an advanced level of their performance of the target language are highly assumed to share some of the two languages’, i.e., native language and target language, norms of appropriateness and politeness. From this study, it appeared that both of pragmatic transfer and pragmatic competence occurred by the YELs. That was occurred in their refusal strategies according to their social status in the situation, higher, equal, or lower and according to the situation itself, a request, an offer, an invitation or a suggestion. Generally speaking, all the three groups participated in this study mostly used similar strategies of politeness in rejecting offers, invitations, requests, and suggestions except in the higher status of refuser. We find that YANSs used ‘excuse’ in the first position of the semantic formula in rejecting an invitation of lower status. On the other hand, neither YELs nor AENSs used the same expression in the first position which gives inclinations of the L2 pragmatic competence of the YELs. YELs and AENSs tended to use similar contents and orders of the semantic formula. They used ‘regret’ expressions “I’m sorry” or ‘positive opinion’ “It’s nice of you to invite us” in the first position, and ‘excuse’ or ‘regret’ in the second and third positions.

     The less use of direct refusal “no” or “can’t” in the first position by all the three groups refers to the same perception of adopting politeness strategies. They tended to be more direct with peers in rejecting their suggestions. However; YANSs used direct refusal style in the first position when they are in higher status. Hints of pragmatic transfer appeared in the lower status situation of the refuser. There we find the use of direct refusal expressions in the first position followed by statements of excuse in the second and third positions of the refusal semantic formula.

The main finding of the study is that the subject in question, YELs, afforded enough indications of pragmatic competence of the target language. English Grammatical accuracy of the YELs was not examined as the main concern of the present study was the sociolinguistic behavior of the subjects.      

References
Al-Shawali, H. (1997). Refusal strategies in Saudi and American culture. Unpublished   M.A. Thesis. Michigan: Michigan University.

Beebe, L.M., T. Takahshi, and R. Uliss-Weltz. (1985). Pragmatic transfer in ESL refusals.  Paper presented at the Second Research Forum, UCLA. To appear in On the  development of communicative competence in a second language, eds. R. C. Scarcella, E. Andersen, and S. C. Krashen Rowley, MA: Newbury House.

Beebe, L.M., Takahashi, T., & Uliss-Weltz, R. (1990). Pragmatic transfer in ESL refusals.  In R. C. Scarcella, E. S. Anderson, and S. D. Krashen (Eds.), Developing  communicative competence in a second language (pp. 55-94). New York: Newburry  House.

Chen, H.J. (1996). Cross-cultural comparison of English and Chinese metapragmatics in refusal. Indiana University. (ERIC Document Reproduction Service No. ED 408 860).

Cohen, A.D. (1996). Developing the ability to perform speech acts. Studies in Second  Language Acquisition, 18, 253-267.

El-Shazly, A. (1993). Requesting strategies in American English, Egyptian Arabic and English as spoken by Egyptian second language learners. Unpublished M.A. Thesis,  Cairo: American University.

Fraser, B. (1983). The domain of pragmatics. In J. C. Richards & R.W. Schmidt (Eds.).  Language and communication (pp. 29-59). New York: Longman.

Gumperz, J. (1990). The conversational analysis of interethnic communication. In R. C.  Scarcella, E. S. Andersen, and S. D. Krashen (Eds.), Developing communicative   competence in a second language: Series on issues in second language research (pp.  223-238). Boston: Heinle & Heinle. 

Kasper, G. (1984). Pragmatic comprehension in learner-native speaker discourse.  Language Learning, 34, 1-20.

Kasper, G. (1990). Linguistic politeness: Current research issues. Journal of Pragmatics,  14, 193-218.

Know, J. (2004). Expressing refusals in Korean and in American English. Multilingua, 23,  339-364.

Liao C. and Bresnahan, M.I. 1996). A contrastive pragmatic study on American English as    Mandarian refusal strategies. Language Sciences, 17(3), 703-727.

Nelson, Al-Batal, and Echols. (1996). Arabic and English compliment responses:  potential  for pragmatic failure. Applied linguistics, 17, 411-432.

Rintell, E. M. (1997). Getting your speech act together: The pragmatic ability of second  language learners. Working Papers on Bilingualism, 17, 98-106.

Scarcella, R. C. (1990). Communication difficulties in second language production,  development, and instruction. In R. C. Scarcella, E. S. Andersen, & S. D. Krashen  (Eds.), Developing communicative competence in a second language: Series on issues
in second language research. Boston: Heinle & Heinle.

Takahashi, T., and L. Beebe. (1987). Development of pragmatic competence by Japanese   Learners of English. Journal of the Japan Association of Language Learners, 8(2).

Umar, A. (2004). Request strategies as used by advanced Arab learners of English as a  foreign language. Umm Al-Qura University Journal of Education & Social Sciences &  Humanities. 16(1).

Weinreich, U. (1953). Language in contact: Findings and problems. New York:  Linguistics Circle of New York. Reprinted by Mouton, The Hague, 1974.

Wolfson, N. (1989). Perspectives, sociolinguistics and TESOL. Boston: Heinle & Heinle.

Appendix A
Appendix B
Appendix C