Interlanguage pragmatics in Russian: the speech act of request in email

As face-threatening speech acts, requests are of particular interest to second language acquisition scholars. They affect the interlocutors' public self-images, and thus require a careful consideration of the social distance between the interlocutors, their status, and the level of the imposition, factors that are weighed differently in different cultures. Studies have revealed that while use of direct and conventionally indirect strategies to perform requests seems to be a universal property of language, languages differ with respect to the choice of linguistic means employed in these two types of strategies. Even though nonnative speakers' perceptions of politeness in requests correlate with those of native, differences in performance exist. Findings from former studies suggest that second language learners' sociolinguistic competence is not native like. To date, however, most studies on performance of requests have focused on English as a second or foreign language. The present study broadens current research by examining requests written by native and nonnative speakers of Russian. The second focus of this study is electronic communication. As a relatively new means of communication, the sociolinguistic dynamics of email is not adequately understood. However, its use for daily communication is increasing in all domains of life, including communication between university students and professors. Former research suggests that inappropriately formulated emails can affect how professors perceive students. As with requests in general, however, most studies to date on email have been iv conducted on English data. Using the Discourse Completion Task (DCT) as the elicitation method, the present study examined electronic messages written by native and nonnative speakers of Russian. The messages were rated by three native speakers on three scales: clarity, social appropriateness, and politeness. One-way ANOVA revealed significant differences between the groups on all three scales. In addition, head acts, alerters, supportive moves, and internal modifications were analyzed using the Cross- Cultural Speech Act Realization Project (CCSARP) coding manual. The strategy analysis revealed interesting similarities and differences between the groups. The study concluded that while Russian learners have approximated native sociolinguistic competence on some levels, significant gaps between native and nonnative formulation of electronic requests remain.

INTERLANGUAGE PRAGMATICS IN RUSSIAN: THE SPEECH ACT OF REQUEST IN EMAIL by Anna M. Krulatz A dissertation submitted to the faculty of The University of Utah in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Linguistics The University of Utah December 2012 Copyright © Anna M. Krulatz 2012 All Rights Reserved The University of Utah Graduate School STATEMENT OF DISSERTATION APPROVAL The dissertation of Anna M. Krulatz has been approved by the following supervisory committee members: Jane Hacking , Chair 10/22/2012 Date Approved MaryAnn Christison , Member 10/22/2012 Date Approved Thomas Huckin , Member 10/22/2012 Date Approved Rachel Hayes-Harb , Member 10/22/2012 Date Approved Adrian Palmer , Member 10/22/2012 Date Approved Johanna Watzinger-Tharp , Member 10/22/2012 Date Approved and by Edward Rubin , Chair of the Department of Linguistics and by Charles A. Wight, Dean of The Graduate School. ABSTRACT As face-threatening speech acts, requests are of particular interest to second language acquisition scholars. They affect the interlocutors' public self-images, and thus require a careful consideration of the social distance between the interlocutors, their status, and the level of the imposition, factors that are weighed differently in different cultures. Studies have revealed that while use of direct and conventionally indirect strategies to perform requests seems to be a universal property of language, languages differ with respect to the choice of linguistic means employed in these two types of strategies. Even though nonnative speakers' perceptions of politeness in requests correlate with those of native, differences in performance exist. Findings from former studies suggest that second language learners' sociolinguistic competence is not native like. To date, however, most studies on performance of requests have focused on English as a second or foreign language. The present study broadens current research by examining requests written by native and nonnative speakers of Russian. The second focus of this study is electronic communication. As a relatively new means of communication, the sociolinguistic dynamics of email is not adequately understood. However, its use for daily communication is increasing in all domains of life, including communication between university students and professors. Former research suggests that inappropriately formulated emails can affect how professors perceive students. As with requests in general, however, most studies to date on email have been iv conducted on English data. Using the Discourse Completion Task (DCT) as the elicitation method, the present study examined electronic messages written by native and nonnative speakers of Russian. The messages were rated by three native speakers on three scales: clarity, social appropriateness, and politeness. One-way ANOVA revealed significant differences between the groups on all three scales. In addition, head acts, alerters, supportive moves, and internal modifications were analyzed using the Cross- Cultural Speech Act Realization Project (CCSARP) coding manual. The strategy analysis revealed interesting similarities and differences between the groups. The study concluded that while Russian learners have approximated native sociolinguistic competence on some levels, significant gaps between native and nonnative formulation of electronic requests remain. TABLE OF CONTENTS ABSTRACT ....................................................................................................................... iii LIST OF FIGURES ........................................................................................................... ix LIST OF TABLES ............................................................................................................. xi ACKNOWLEDGEMENTS .............................................................................................. xii Chapter 1 INTRODUCTION ........................................................................................................ 1 Chapter overview ............................................................................................................ 1 Pragmatics ....................................................................................................................... 2 Conversational implicature ............................................................................................. 3 Grice's cooperative principle and the four maxims ........................................................ 3 Critique of Grice's theory ........................................................................................... 8 Relevance theory ....................................................................................................... 11 Searle: direct and indirect speech acts .......................................................................... 14 Speech acts and politeness ............................................................................................ 16 Brown and Levinson's model ................................................................................... 16 Maxims of politeness: Leech and Lakoff .................................................................. 21 Criticisms of the existing models of politeness ........................................................ 24 Politeness and the speech act of request ................................................................... 26 American English vs. Russian .................................................................................. 30 Interlanguage pragmatics .............................................................................................. 36 Communicative competence ..................................................................................... 36 Sociopragmatic and pragmalinguistic failure ........................................................... 44 Pragmatic transfer ..................................................................................................... 46 vi Email ............................................................................................................................. 46 Email conventions ..................................................................................................... 47 Netiquette .................................................................................................................. 47 2 REVIEW OF LITERATURE ..................................................................................... 53 Chapter overview .......................................................................................................... 53 Interlanguage pragmatics: speech acts .......................................................................... 54 Requests .................................................................................................................... 54 Studies on perception of requests ............................................................................. 57 Production studies ..................................................................................................... 61 Requests in English and Russian .............................................................................. 68 Email and second language learners ............................................................................. 70 Electronic requests .................................................................................................... 70 Judgments of nonnative requests by native speakers ................................................ 73 Motivation for the study ................................................................................................ 75 Research questions .................................................................................................... 78 3 STUDY DESIGN........................................................................................................ 79 Chapter overview .......................................................................................................... 79 Participants .................................................................................................................... 79 Materials and procedures .............................................................................................. 81 English control data .................................................................................................. 86 Follow-up interviews ................................................................................................ 86 Data analysis ................................................................................................................. 87 Research question one ............................................................................................... 87 Research questions two and three ............................................................................. 88 4 RESULTS ................................................................................................................... 90 Chapter overview .......................................................................................................... 90 Descriptive statistics ..................................................................................................... 91 Analysis of variance ...................................................................................................... 94 Analysis of strategies .................................................................................................... 97 vii Head act .................................................................................................................... 99 External modifications ............................................................................................ 104 Alerters .................................................................................................................... 105 Supportive moves.................................................................................................... 113 Follow-up interviews .................................................................................................. 117 Conclusion .................................................................................................................. 118 5 DISCUSSION ........................................................................................................... 120 Chapter overview ........................................................................................................ 120 Summary of findings................................................................................................... 120 Sociolinguistic competence ........................................................................................ 122 Pragmalinguistic failure .............................................................................................. 124 Pragmatic transfer ....................................................................................................... 126 Politeness .................................................................................................................... 129 Requests proper vs. email writing conventions .......................................................... 130 Cultural values ............................................................................................................ 132 Extended stay in the target culture .............................................................................. 133 Limitations of the study .............................................................................................. 135 Directions for future research ..................................................................................... 138 Pedagogical implications ............................................................................................ 140 Conclusion .................................................................................................................. 142 Appendices A: QUESTIONNAIRE FOR AMERICAN LEARNERS OF RUSSIAN ...................... 144 B: QUESTIONNAIRE FOR NATIVE SPEAKERS OF RUSSIAN .............................. 146 C: ENGLISH QUESTIONNAIRE FOR RUSSIANs ..................................................... 148 D: DISCOURSE COMPLETION TASK ....................................................................... 150 E: QUESTIONNAIRE FOR CONTROL GROUP......................................................... 119 F: SAMPLE INTERVIEW QUESTIONS ...................................................................... 119 viii G: INSTRUCTIONS FOR RATERS .............................................................................. 154 H: ENGLISH CONTROL DATA ELICITATION INSTRUMENT.............................. 119 REFERENCES ............................................................................................................... 158 LIST OF FIGURES Figure Page 1: Brown and Levinson's (1987) strategies for performing FTAs ……………………... 19 2: Distribution of NS scores …………………………………………………………… 92 3: Distribution of NNS scores ………………………………………………………….. 93 4: Comparison of average NS and NNS scores ………………………………………... 95 5: Ratings of native vs. nonnative messages …………………………………………… 96 6: Distribution of levels of directness …………………………………………………. 101 7: Direct strategies used by NS and NNS …………………………………………….. 103 8: Conventionally indirect strategies used by NS and NNS ……………………..…… 103 9: The most common downgraders used by NS and NNS ………………………….... 106 10: The use of upgraders by NS and NNS …………………………………………….. 106 11: Forms of address (names) used by NS ……………………………………….…… 107 12: Forms of address (names) used by NNS ………………………………………….. 107 13: The use of patronymics in NS and NNS messages ………………………….……. 108 14: Salutations used by NNS ……………………………………………………..…… 109 15: Salutations used by NNS …………………………………………………….……. 109 16: The use of ‘s uvajeniem' (‘with respect') ………………………………………… 111 17: The use of ‘s uvajeniem' (‘with respect') standing along and with other closing expressions ………………………………………………………………………… 112 18: Capitalization of ‘vi' (‘you') by NS and NNS …………………………….……… 112 x 19: Preparators ………………………………………………………………………… 114 20: Distribution of references to time …………………………………………………. 118 LIST OF TABLES Table Page 1: External modifications of requests ……………………………………….………….. 55 2: Summary of different levels of directness used to realize request proper (adapted from Blum-Kulka, House and Kasper, 1989, p. ……….………...………………….. 56 3: Range of NS scores ……………………………………………………….…………. 91 4: NS ratings: mean scores and standard deviations …………………………….…...… 93 5: Range of NNS scores ………………………………………………………….…….. 93 6: NNS ratings: mean scores and standard deviations …………………………….……. 95 7: Types of requests produced by NS and NNS ……………………………………..... 100 8: Closings used by NS and NNS ………………………………………………….….. 115 9: Grounders used in scenarios 1-4 ……………………………………………………. 117 10: Types of disarmers used by NS and NNS ………………………………...………. 117 ACKNOWLEDGEMENTS As my academic journey at the University of Utah is coming to an end, I have several people in mind to whom I would like to express my gratitude. First and foremost, I would like to thank Jane Hacking, my committee chair, for sparking my interest in Russian, and for many years of patience, support, guidance, thoughtful insights and contributions to my dissertation study, as well as her approachability, enthusiasm about my work, and kindness. I am also especially grateful to the other four members of my committee, MaryAnn Christison, Rachel Hayes-Harb, Johanna Watzinger-Tharp, Thomas Huckin and Adrian Palmer. This dissertation could not have been completed without their encouragement and valuable comments on my work. I will never forget Dr. Christison's warm welcome on my first day in Utah; Dr. Hayes-Harb's lessons on conducting ethical research; Prof. Huckin's generous willingness and availability to answer any of my questions; Prof. Palmer's honest advice and sense of humor without which I would have probably given up my graduate studies; and Prof. Watzinger-Tharp's thoughtful feedback on my work, and her enthusiasm to join the committee in the final stages of my dissertation work. My deep thanks go to the chair of the Department of Linguistics, Edward Rubin, for his continuous support of me as a graduate student, and to the department staff for creating such positive ambiance and always being willing to help. xiii I am also deeply indebted to Rimma Garn in the Department of Languages and Literatures, who graciously agreed to help me recruit participants in her Russian classes. Similarly, I owe big thanks to my good friend Helen Benediktova, who has helped me collect data from the native speakers of Russian. Without Rimma's and Helen's help, I would not have been able to finish collecting data for my study. I also would like to express my appreciation to the volunteers and who participated in the experiment. Some of them were students and faculty in distant Krasnoyarsk who have never even met me in person. Others were students or alumni of the University of Utah who agreed to help me without any compensation. Meeting with them and talking about their experiences learning Russian was an invaluable lesson for me, and I will never forget the generosity with which they offered their time and help. I am also deeply indebted to the three raters without whose native-speaker judgments and great enthusiasm for my study I would not be able to finish the data collection process. Finally, I would like to thank my family members and friends for their continuous support and good humor that cheered me up along the way. In particular, I would like to thank Alex, my sweetness and my husband, who stood by my side all those years and not once complained that it was taking too long. The work I have invested in my studies does not even measure up to the sacrifices he has made for me. Hanza, my best German Shepherd Dog, deserves a big hug for all these days when he helped me manage my stress, yet I did not have time to take him for a hike. I am deeply grateful to my parents for their constant support and belief in me. This dream sprouted when I was a little girl, and I would not have accomplished it without them. CHAPTER 1 INTRODUCTION Chapter overview The goal of this chapter is to present the theoretical background for the study. The chapter introduces pragmatics as a subfield of linguistics dealing with language use in a social context. It reviews the main contributions to the field of pragmatics by presenting the work of Austin (1962), Grice (1975), Searle (1969, 1975, and 1979), and Wilson and Sperber (1986, 1995, 1998, and 2002). Next, it provides an overview of the main directions in the theory of politeness including Brown and Levinson's model (1978, 1987), Leech's (1983) Politeness Principle, and Lakoff's (1973, 1977) rules of politeness. In the following section, cross-cultural issues in politeness are discussed, focusing specifically on Russian and North American cultural values. The chapter then introduces the field of interlanguage pragmatics which is interested in nonnative speakers' communicative competence, and provides an overview of the main typologies of language competence including Canale and Swain's (1980) model and Bachman and Palmer's (1982) framework. Finally, the chapter discusses email as a means of communication, with a special focus on email conventions and netiquette. 2 Pragmatics Competent language users have a firm grasp of the rules of syntax, morphology and phonology of that language, and know how to apply these rules in a given social context, at the discourse level. Pragmatics is the study of meaning in use. It focuses on how meaning is constructed and interpreted in a given context, and on how speakers often express more than, or something different than, what they actually say. In Green's words (1996), pragmatics views communication as ‘the successful interpretation by an addressee of a speaker's intent in performing a linguistic act' (p. 1). Similarly, both Austin (1962) and Searle (1969, 1975) argue that the goal of pragmatics is to analyze how speakers use language to perform linguistic acts, e.g., requests, apologies, orders, advice, etc. The linguistic act, commonly referred to as the speech act, is the basic unit of analysis in pragmatics. Following Austin's Speech Act Theory (1962), it is conventional in pragmatics to use a three-fold distinction between different levels of speech act meaning: Locution - the literal meaning Illocution - the meaning intended by the speaker Perlocution - the effect the speech act has on the recipient For example, the utterance: The dog is whining (locution) can imply that the speaker wants the recipient to take the dog out (illocution), and as a result the recipient may indeed take the dog for a walk (perlocution). This distinction is an important one because more often than not, utterances are not interpreted literally (e.g., the above utterance could be interpreted as a simple statement of a fact, or as a request to take the dog for a 3 walk), and because a misinterpretation at the illocutionary level can lead to a perlocution unintended by the speaker (in this case, the listener could lock the dog in his kennel and, as a result, the dog could urinate on his bed). Thomas (1995) points out that the intended illocutionary force is typically interpreted correctly by competent adult language users. However, a lack of communicative competence (defined below) on the part of the speaker, and/or a misinterpretation on the part of the recipient can, as will be shown later in this chapter, often lead to miscommunication at both the locutionary and illocutionary level in a second language (L2). Conversational implicature While Austin's distinction between locution, illocution, and perlocution allows us to differentiate between the utterance meaning and the speaker's intended meaning, Grice uses the terms ‘what is said' and ‘what is meant.' He also introduces the term conversational implicature to refer to the inference made by the recipient when attempting to understand the implied meaning (Grice, 1975). Thomas (1995) clarifies that implicature is generated by the speaker while the listener produces an inference (i.e., deduction based on the evidence in the speaker's utterance) (p. 58). By generating an inference, the recipient gets from the locution (what is said) to the illocution (what is meant). Grice's cooperative principle and the four maxims The Cooperative Principle and four conversational maxims allow the listener to interpret conversational implicature (Grice, 1975). The Cooperative Principle states: 4 ‘make your conversational contribution such as is required, at the stage at which it occurs, by the accepted purpose or direction to the talk exchange in which you are engaged' (p. 78). This means both the speaker and the listener are constantly interpreting each other's goals in a conversation. The four maxims (p. 80) are as follows: Quantity: Make your contribution as informative as required (for the current purpose of the exchange). Do not make your contribution more informative than is required. Quality: Do not say what you believe to be false. Do not say that for which you lack adequate evidence. Relation: Be relevant. Manner: Avoid obscurity of expression. Avoid ambiguity. Be brief (avoid unnecessary prolixity). Be orderly. It is not always possible to observe all four maxims at the same time, as they may be in conflict. For example, in a situation when the speaker is not sure how much background information the hearer already has, the maxim of Manner may have to be sacrificed by saying more than was necessary. Grice distinguished five different ways in which speakers can fail to observe a maxim: flouting a maxim, violating a maxim, infringing a maxim, opting out of a maxim, and suspending a maxim. To flout a maxim means to blatantly fail to observe it in order to generate an implicature. That is, when a maxim is flouted, the hearer is expected to look for a meaning which is different from the expressed meaning. Sometimes speakers flout one of the maxims explicitly, for example 5 by saying ‘I know this is irrelevant, but…' (Relation) or ‘This may not be true, but…' (Quality). More often, however, maxims are flouted without the use of metadiscourse that points it out. In the following scenario, the maxim of Quality is flouted because the speaker got annoyed with the interlocutor's question and wants to signal to him that the topic is unwelcome: Example 1 B just told A that his major is linguistics. A: That's really interesting. What languages do you speak? B: Well… let's see… English, Dutch, Spanish, Italian, Romanian, Turkish, Cuitlatec, Tumak and Shawia Berber. Obviously (to a linguist), B says something that he believes to be false. He does so because A asked a question typical of nonlinguists who often assume that all linguists do is learn to speak foreign languages, and because he wants to prevent further questions about his major. The purpose is not to lie, but to generate an implicature. By saying something that is clearly improbable or untrue, the speaker hints that it is time to switch to a different topic. Thus, the maxim of Quality is flouted when speakers do not have sufficient evidence for what they say or what they say is blatantly untrue. Another example of flouting the maxim of Quality is when at the end of a very unsuccessful presentation audience members make sarcastic comments such as ‘That presenter was very knowledgeable, wasn't she?' By doing so, they violate the maxim of Quality. To flout the maxim of Quantity is to blatantly provide more or less information than is required in a given situation, as the example below illustrates: 6 Example 2 A: What are you doing? B: Nothing. Here B gives less information than A wants and generates some ‘extra meaning,' which in this case would be B's desire to be left alone. Similarly, when the speaker provides more information than required, the maxim of Quantity is flouted. The maxim of Relation is flouted when the speaker's response is irrelevant to the conversation, for instance when the topic of the conversation is abruptly changed or when the interlocutor's question is flagrantly ignored. An example of flouting the maxim of Relation is when, in the middle of a fierce verbal fight, one of the interlocutors suddenly observes: ‘What a gorgeous day it is today! D'you think I should call Amy and invite her for a walk?' By flouting the maxim of Relation, the speaker generates an implicature that he is done with the fight now, perhaps even that he intends to ignore the hearer from now on. Finally, speakers flout the maxim of Manner if the idea is not expressed in the clearest, briefest, and most orderly manner possible. For example, ambiguous statements, which are often utilized in advertising, violate the maxim of Manner, as exemplified below (Chen, 1992, p. 5): Example 3 The Discover card: ‘It pays to Discover' Natural gas: ‘Air pollution is a problem hanging over all of us.' While to flout a maxim is to flagrantly nonobserve it, to violate a maxim, on the other hand, means to do so deliberately (i.e., in order to deceive or mislead the 7 interlocutor). For example, people often withhold information, thus violating the maxim of Quantity, as in the following situation: Example 4 B knows that A's mom was just diagnosed with cancer but doesn't want to break the news to her. A: So what did the doctors say? B: They are still running tests. A maxim can also be violated for the reason of politeness, for example when people tell the host that dinner was delicious whereas, in fact, they did not enjoy it. They choose to violate one of the maxims rather than to offend another person. In addition to flouting a maxim and violating a maxim, Grice draws a distinction between opting out of a maxim and infringing a maxim. Opting out of a maxim takes place when the speaker does not show willingness to cooperate in the way required by a maxim, for example because of legal or ethical reasons. Infringing a maxim happens when the speaker fails to observe a maxim not to generate an implicature or to deceive, but because of the lack of linguistic means. Appropriately observing, flouting or opting out of maxims is a sophisticated part of sociolinguistic competence, and it can be a challenge even for native speakers. This is because in order to correctly create and interpret implicatures speakers have to be familiar with the sociolinguistic rules of the language they are using and have sufficient background / contextual knowledge. These tasks are even more challenging for nonnative speakers whose sociolinguistic competence has not attained native-like proficiency. Thus, second language learners often infringe a maxim, which can lead to miscommunication. 8 Another possible interpretation is that nonnative speakers of a language, because of their lack of sociolinguistic competence, fail to correctly flout a maxim and generate an implicature. Infringing of some maxims, for example of the maxim of Quantity, can also stem from cultural differences, as when Russian speakers of English respond with a long story of their day when asked, ‘How are you?' because in Russian culture responding in such a way is culturally appropriate. Whereas some linguists (see for example Keenan 1976, Thomas 1995) would classify such instances as suspending a maxim, that is, a situation when interlocutors do not expect a maxim to be fulfilled, another possible explanation would be to say that although the maxims may be universal, their interpretations differ cross culturally. In other words, whereas in some cultures the maxim of Quantity in the above example is satisfied by responding ‘I'm fine. How about yourself?' in other cultures such a response would amount to nonobservance of the maxim as a more elaborate response is expected. Critique of Grice's theory While on the surface Grice's system of maxims and different ways of maxim non-observance seems quite straightforward, critics have identified several problems with it. It is not the purpose of this dissertation to critique and revise Grice's theory, but a few brief points will be noted, nonetheless. The first issue is that it is not always clear whether the nonobservance of a maxim is intentional or unintentional, thus making it difficult to distinguish between flouting and violation of a maxim (Thomas, 1995, p. 88). Grice's theory does not provide an explanation of how the hearer can decide whether the nonobservance is a flout, a 9 violation, an infringement, or opting out of a maxim. The second problem is that a given utterance can often produce more than one possible implicature. Grice's theory, however, does not provide an explanation of how hearers determine which implicature is the correct one considering the circumstances in which the utterance was produced. Language users encounter ambiguous utterances like the one in Example 5, below, quite frequently (adapted from Thomas, 1995, p. 89): Example 5 A to her husband: A: The cleaning lady is coming tomorrow. You have some 20-dollar bills lying around your desk. It is not clear from A's utterance if she is implying that the cleaning lady is not to be trusted, if she is asking her husband to give her the money to pay the cleaning lady, or if she is simply saying that all surfaces should be cleaned of clutter so that the cleaning lady can easily dust them. Similar problems can arise when one of the interlocutors comes from an Eastern European culture that differs significantly from Western cultures when it comes to hospitality expectations and table manners. Many Americans who have spent time living in Russia may have experienced a problem similar to the one described below: Example 6 A is an American who was invited to B's house for dinner. B is Russian. A just finished the second portion of cake that was placed on his plate by the host. A: This cake was delicious. Thank you. B reaches out for A's plate to serve yet another slice of the cake to him. 10 In this case, A's utterance could have generated two implicatures: (1) ‘Thank you. I am full now.' or (2) ‘Thank you. Could I have another piece?' Because of the host's Russian background, she chose the second interpretation, but Grice's theory does not predict how this choice of interpretation is made. Overlap of maxims is yet another problem with Grice's theory that Thomas points out (p. 91). In particular, she argues that it is often difficult to distinguish between the maxim of Quantity and the maxim of Manner, as well as between the maxim of Quantity and the maxim of Relevance. To support the first claim, Thomas provides the following example (p. 92): Example 7 A: What did you have to eat? B: Something masquerading as chicken chasseur. This situation can be interpreted in two ways: either B is providing more information than required (‘chicken chasseur' would suffice) thus violating the maxim of Quantity, or B's response is obscure and violates the maxim of Manner. In regards to the latter, Thomas notes it is almost impossible to find instances where the maxim of Relevance is not in operation. She asserts, ‘unless you assume that a contribution is in some way relevant to what has gone before, you will not begin to look for an implicature' (p. 92). Clearly, Grice's theory is quite informal (Thomas, p. 93), and it provides a set of procedures for calculating the conversational implicature in only some simple cases. 11 Relevance theory The most serious challenge to Grice's inferential theory has been advanced by Wilson and Sperber. In ‘Truthfulness and Relevance' (2002), as well as in earlier publications (1986, 1995, and 1998), they propose an alternative theory of communication. Instead of assuming that language interaction is governed by the cooperative principle, to which they term ‘the maxim of truthfulness,' they suggest that it is better accounted for in terms of a principle of relevance. They criticize Grice's model, in which the hearer infers the meaning of an utterance through simply ‘assigning referents to referring expressions, and perhaps [calculating] implicatures' (p. 600), as not accurately reflecting the complexity of human communication. Relevance Theory, on the other hand, is supposed to account for situations in which the discrepancy between the literal meaning and the inferred meaning is much greater. Wilson and Sperber (2002) define ‘relevance' as ‘a property of inputs to cognitive processes which makes them worth processing' (p. 600). In human to human language communication, this translates into ‘improvements in knowledge.' They argue that telling the truth is not a necessary condition for communication; on the contrary, the intended meaning can be derived from the literal meaning, whether it be literal, loose, or figurative, because the speaker constructs ‘an interpretation which satisfies the hearer's expectations of relevance' (2002, p. 599).The theory is based on two general claims: Cognitive Principle of Relevance: human cognition tends to be geared to the maximization of relevance. Communicative Principle of Relevance: every act of overt communication conveys a presumption of its own optimal relevance. 12 The first claim predicts that humans focus on the most relevant stimuli at a time and derive interpretations from them. The second claim ensures that the hearer can assume the utterance is relevant through the speaker's sheer willingness to participate in the interaction. In addition to these two rules, the relevance theory also assumes that utterances are relevant enough to be worth processing and the most relevant one compatible with the communicator's abilities and preferences (presumption of optimal relevance); and that the interpretation of an utterance should require minimal effort and it should stop when the hearer's expectations of relevance are satisfied (relevance-guided comprehension heuristic). The main advantage of Relevance Theory as compared to Grice's model is that it is based on one simple assumption that hearers make sense of an utterance by selecting those features of the context that are relevant, rather than plowing through a complex system of maxims. They assume that utterances are connected in a meaningful way and they utilize their background knowledge of the world in making inferences. Similarly, the speaker has to simply focus on selecting the information that should be relatively easy for the hearer to understand. An utterance is understood if its relevance is proven, i.e., if its meaning is accessible to the hearer. Very often, this means filling in the unspoken words, which Sperber and Wilson refer to as ‘explicature,' and which they render a necessary step before implicature (2002, p. 261). To illustrate how Relevance Theory copes with interactions that Grice's maxims could not explain, let us consider again example 6 above, provided here again as example 8: 13 Example 8: A is an American who was invited to B's house for dinner. B is Russian. A just finished the second portion of cake that was placed on his plate by the host. A: This cake was delicious. Thank you. B reaches out for A's plate to serve yet another slice of the cake to him. Relevance Theory allows the following interpretation. Because A is an American, when saying ‘Thank you,' she intends for B to infer ‘Thank you, I have had enough.' However, because B is Russian, the information that he fills in is ‘Thank you, and I would love some more.' In other words, Relevance Theory, by highlighting the importance of the context and of the speaker's and hearer's background knowledge, eliminates the ambiguity resulting from literal interpretations of utterances and allows the participants to infer the implied meaning. To consider another example, let us look at the conversation below: Example 9 A, B and their dog are in front of their house weeding the yard. The front yard is not fenced. They see a cat across the street. A: Here comes lunch. B leashes the dog that starts barking ferociously. B: Oh, how he loves cats! Grice's theory would account for this situation by explaining that A flouted the maxim of relation by saying something that is very irrelevant to what they were doing (they were weeding their front yard and unless they were the relatives of Alf, we would not expect them to enjoy a serving of cat for lunch). Next, B flouted the maxim of quality 14 by saying something that is contrary to truth (why would she leash the dog if she did not think he was going to chase the cat?). This example clearly shows how inefficient Grice's theory is - it takes several steps for the interlocutors to make sense of this conversation. The Relevance Theory, on the other hand, provides a much more effortless way to interpret A's and B's utterances. All they have to do is recall their background knowledge about the dog (he hates cats; in fact, he chases them whenever he gets a chance), and assume that anything that is mentioned is relevant to the context of the conversation. Thus, while in Grice's theory, the speaker and the hearer are involved in a complex evaluation of which maxims were observed, flouted, or infringed, the Relevance Theory simplifies this process by allowing the interlocutors to assume that everything that is said is related, thus limiting the number of different inferences that can be made. Searle: direct and indirect speech acts An overview of early work in pragmatics would not be complete without mentioning the contributions made by Searle (1969, 1975). While Austin proposes the distinction between locution, illocution, and perlocution, and Grice makes a distinction between ‘what is said' and ‘what is meant,' Searle uses the terms ‘propositional content' (the literal meaning of an utterance) and ‘illocutionary force' (what the speaker intends by what is said). In his ‘Indirect Speech Acts' (1975), Searle differentiates between direct and indirect speech acts. Direct speech acts are those in which the propositional content (i.e., ‘what is said') carries one illocutionary force (i.e., ‘what is meant'). For example, if the speaker says to the hearer ‘Please walk the dog' and by that means that the hearer should 15 put on a pair of shoes and weather-appropriate apparel, put a leash on the dog, exit the house, and proceed down the street, the speaker is performing a direct request. If, however, the speaker utters the statement, ‘The dog is whining,' and by that implies that the hearer should put on a pair of shoes and weather- appropriate apparel, put a leash on the dog, exit the house, and proceed down the street, the speaker is performing an indirect request. Searle (1979) defines an indirect speech act as an act performed ‘by means of another' (p. 60), and states that in indirect speech acts the speaker communicates more than is actually said. Thus, in direct speech acts, there is a connection between the literal meaning and the conventional meaning, or between the form and the function of the utterance. In indirect speech acts, the literal meaning and the conventional meaning are different. Whereas indirect speech acts are a normal occurrence in everyday language use, Searle is concerned with explaining how it is possible for the speaker to generate them and for the hearer to interpret them. He proposes that this can be accomplished because both the speaker and the hearer share the same linguistic and nonlinguistic background information which allows them to create an implicature and make correct inferences, respectively. In particular, he suggests that mutual understanding is possible due to the cooperative principles of conversation (Grice, 1975), the factual background information and accepted conventions that interlocutors share, and the power of inference (Searle, 1975: 61). The illocutionary force of some indirect speech acts can be interpreted based on their conventional use. Searle (1975, 1979) provides a long list of examples of structures conventionally used to perform indirect requests in English. He divides them into the 16 following groups: sentences concerning the hearer's ability to perform an act (e.g., ‘Can you walk the dog?'); sentences concerning the speaker's wish or want that the hearer will do an act (e.g., ‘I would like you to walk the dog'); sentences concerning the hearer doing an act (e.g., ‘Will you walk the dog?'); sentences concerning the hearer's desire or willingness to do an act (e.g., ‘Would you mind walking the dog'); sentences concerning reasons for doing an act (e.g., ‘You should walk the dog'); and sentences embedding one of the above elements inside another or embedding an explicit directive illocutionary verb inside one of the above (e.g., ‘Would it be too much if I suggested that you could possibly walk the dog?') (pp. 65-67). While Searle's categories of indirect requests are not going to be used in the study proposed here, they are interesting in that they show a wide the range of structures used to perform conventionally indirect requests. Speech acts and politeness Brown and Levinson's model One of the reasons why there are so many different ways to perform indirect speech acts is politeness (Brown & Levinson, 1978; Lakoff, 1973, 1977). The notion of politeness is inseparable from the daily use of language. In other words, as people participate in linguistic interactions, they use politeness strategies to accomplish their goals. According to Thomas (1995), politeness is context-dependent, i.e., it consists of a linguistic form, the context in which the utterance occurs, as well as the relationship between the interlocutors (p. 157). One of the most influential models of politeness was developed by Brown and Levinson (1978, 1987). This model is based on the notion of face proposed by Goffman 17 (1967). Face refers to one's public self-image and how this self-image is maintained in interactions with others. According to Brown and Levinson, when people interact with each other, they have to show an awareness of the face. Brown and Levinson distinguish between ‘positive face,' i.e., a person's wish to be a part of a group and sharing involvement with others, which is manifested in expressing friendliness or approval. ‘Negative face,' on the other hand, involves detachment and a need for personal freedom, and is accomplished by giving options, apologizing, and recognizing the status of the interlocutor. When humans use language, they typically aim at establishing a good relationship as part of an interaction, which is accomplished by juggling positive face and negative face. In other words, people show respect for each-other's expectations regarding self-image, they take each-other's feelings into consideration, and they avoid Face Threatening Acts (FTAs). FTAs are those illocutionary acts which can damage or threaten a person's face (Brown & Levinson, 1978, 1987). Invitations, orders, requests, apologies and insults are some examples of Face Threatening Acts. FTAs can threaten both the speaker's face (e.g., an apology) and the listener's face (e.g., an offer). However, there are some strategies available to interlocutors that allow them to reduce the likelihood of damaging or threatening another person's face. An appropriate strategy can be chosen based on the assessment of the FTA. The first available strategy is to decide whether to perform the FTA or avoid it altogether (say nothing). If, however, the speaker chooses to perform the FTA, four more strategies can be used to minimize the effect of the FTA on the hearer: 1. Perform the FTA on-record without redressive action (bald-on-record) 2. Perform the FTA on-record using positive politeness 18 3. Perform the FTA on-record using negative politeness 4. Choose an off-record strategy To perform an FTA boldly on record means to perform the act directly (cf. Searle's distinction between direct and indirect speech acts). Bold on record acts are performed without undertaking redressive action, i.e., without an attempt to save the face. Directness is not always equal to rudeness, however. For example, issuing the request in Example 10 below to a neighbor is highly face-threatening, whereas making the offer in Example 11 to a guest is not. This is because directness often indicates lack of social distance. Example 10 To a neighbor: A: I want you to walk my dog when I am on vacation next week. Example 11 To a dinner guest: A: Have some more cake. Another strategy to minimize the threat to the face is performing the FTA on record using positive politeness. This is accomplished by appealing to the hearer's need to be liked and treated as a member of the same group. Examples of appeals to positive face include cases when the speaker demonstrates closeness and solidarity with the hearer, makes appeals to friendship, makes the hearer feel good or emphasizes common goals. 19 Brown and Levinson's model of strategies for performing FTAs is represented in Figure 1. They list fifteen different strategies to perform positive politeness: seeking agreement; avoiding disagreement; joking; offering or promising; being optimistic; intensifying interest to the hearer; using in-group identity markers; presupposing or asserting common ground; attending to the hearer's interests, wants, needs or goods; exaggerating interest, approval or sympathy with the hearer; including both the speaker and the hearer in the activity; giving reasons; and assuming or asserting reciprocity; and giving gifts to the hearer (pp. 101-129). The third option to reduce the impact of the FTA is to use on record negative politeness. By doing so, the social distance between the speaker and the hearer is stressed, and the interlocutors avoid encroaching on each other's territory. Some negative politeness strategies include giving options, using apology for interference or speech act do the FTA don't do the FTA on record off record without redressive with redressive action, baldly action positive politeness negative politeness Figure 1: Brown and Levinson's (1987) strategies for performing FTAs. 20 transgression, stressing the importance of one's values (e.g., time), using conventional politeness markers such as hedges, or impersonalizing strategies such as passive voice (Brown & Levinson, 1987, p. 70). Finally, it is also possible to go off record, or formulate the speech act in an indirect way. Brown and Levinson define an indirect speech act as an act that "is either more general (contains less information in the sense that it rules out fewer possible states of affairs) or actually different from what one means (intends to be understood)" (p. 211). In other words, off-record strategies require violation of one or more of Grice's cooperative maxims. In order to understand an indirect speech act as intended, the hearer has to make an inference (cf. Grice's conversational implicature). If the listener wants to perform an FTA off record, Brown and Levinson claim, the hearer needs to receive some hints and draw on contextual clues in order to correctly interpret the implicature (p. 213). Brown and Levinson distinguish fifteen off-record strategies: giving hints; giving association clues; presupposing; understating; overstating; using tautologies; using contradictions; being ironic; using metaphors; using rhetorical questions; being ambiguous; being vague; over-generalizing; displacing the hearer; and being incomplete (using ellipsis). These strategies produce an illocutionary force that is likely to be correctly understood by the hearer. In the framework developed by Brown and Levinson, the three social factors affecting the choice of appropriate linguistic strategies to save both the listener's and the hearer's face are the distance, the power relationship, and the ranking of imposition. These factors are used by the speaker in computing the seriousness of an FTA to be performed. Once this is done, the speaker can choose the most appropriate strategy to 21 perform the FTA. Brown and Levinson propose that the politeness of a message increases together with the level of threat posed by an FTA. They also claim that their theory is universal, i.e., that everyone has a positive and a negative face; differences between cultures are accounted for in terms of preferences for politeness strategies, thus gravitating either towards positive politeness (e.g., Western cultures) or negative politeness (e.g., Chinese culture). Maxims of politeness: Leech and Lakoff Leech (1983) and Lakoff (1973, 1977) propose models of politeness in terms of conversational maxims. Leech proposes a Politeness Principle and six conversational maxims: tact, generosity, approbation, modesty, agreement, and sympathy. While Grice's Cooperative Principle accounts for the relation between the sense and the illocutionary force of the utterance, Leech argues that the Politeness Principle with its six maxims is needed to account for the use of indirectness and for the relationship between the sense and illocutionary force of nondeclarative utterances. That is, in cases where the Cooperative Principle alone fails to provide a satisfactory explanation, the Politeness Principle can rescue it (1983, p. 80). Leech's Politeness Principle states, ‘Minimize (other things being equal) the expression of impolite beliefs; Maximize (other things being equal) the expression of polite beliefs' (p. 81). The tact maxim states that speakers should minimize the cost and maximize the benefit to others. That is, this maxim allows speakers to minimize the imposition on the hearer and also to allow options. The maxim of generosity says requires the speaker to minimize the expression of benefit and maximize the expression 22 of cost to self. Thomas (1995) suggests it should be rephrased as follows: "Minimize the expression of cost to other; maximize the expression of benefit to other" (p. 162). Languages vary in the extent to which they apply Leech's maxims. The maxim of approbation requires the speaker to minimize dispraise and maximize praise of others as well as persons or things that they value. That is, whenever possible, the speaker should strive to praise others and to withhold criticism. This maxim helps interlocutors avoid disagreement and make each other feel good by showing solidarity. The maxim of modesty, on the other hand, requires that the speaker should maximize the expression of dispraise of self and minimize the expression of praise of self. It is because of this maxim that interlocutors may reject a complement or diminish the value of a favor they did for someone as in the example below: Example 12 A: Thanks again for walking my dog. It was really nice of you. B: No big deal. The fifth maxim, the maxim of agreement, helps interlocutors avoid conflict. It requires them to minimize the expression of disagreement and maximize the expression of agreement. That is, this maxim predicts that interlocutors should be more direct in expressing agreement than in expressing disagreement. The maxim is apparent in the use of hedges when expressing a different point of view, as Example 13 shows. Example 13 A: You know, I really don't think your dog should sleep in your bed. B: I can see your point, but… 23 Finally, the maxim of sympathy states that one should minimize antipathy and maximize sympathy between self and other; this maxim operates in situations when the speaker expresses regret, condolences, or congratulations. Thus, it explains why interlocutors avoid openly speaking about unpleasant topics and use euphemisms to soften the message. As with all other of Leech's maxims, the application of the maxim of sympathy is culture specific. Anybody who either grew up or has spent a considerable amount of time in a Slavic country such as Poland or Russia will (even if only for a brief moment) be taken aback at the sight of an (overly) excited American congratulating a friend on something as banal as passing a driving test or putting together a bookshelf. Similarly, to give another example, the application of the maxim of Modesty varies across cultures. Although in Western cultures compliments are supposed to be accepted, the response in Example 14 is appropriate in some (Slavic) cultures: Example 14: A: I love your dress! B: Oh, it's so old. According to Lakoff, the following three rules of politeness minimize interlocutors' "wasted effort or friction" or confrontation between them (1977, p. 88), and preserve their positive face and negative face: formality, hesitancy, and equality or camaraderie. The first rule states: ‘don't impose,' (p. 88) thus ensuring that speakers behave in accordance with the etiquette required by the context and mark the social distance between each other using strategies such as passive or V rather than T pronouns (e.g., ‘Sie' rather than ‘du' in German, or ‘вы' rather than ‘ты' in Russian), and that they do not ask inappropriate questions. Lakoff's second rule of politeness requires that 24 interlocutors give each other options. This rule is apparent in the use of politeness strategies such as the use of questions or question intonation, the use of tags or ‘please' in imperatives and instructions, use of particles such as ‘well,' ‘er,' ‘ah' and the use of euphemisms (p. 90). The last rule, equality or camaraderie, requires that the speaker should make the hearer feel good in situations when the speaker is of superior or equal status with the hearer (p. 94). Strategies used to implement the rule of equality involve switching from using titles and last name to first name in the form of address; using informal expressions or switching to the hearer's dialect to show solidarity; and talking about more intimate topics. In sum, both Brown and Levinson's face-saving view of politeness and Leech's and Lakoff's conversational-maxim approach expand Grice's account of language communication in that they view communication as a compromise between attaining maximum efficiency and maintaining positive relationships among interlocutors. Another important similarity is the claim that the level of indirectness is positively correlated with the level of politeness. The theories differ in that the face-saving account emphasizes constant attention to face through selection of appropriate linguistic means, while the conversational-maxim approach views politeness as conflict avoidance. Nevertheless, both approaches have important shortcomings that are discussed below. Criticisms of the existing models of politeness None of the models of politeness presented above is without problems. Leech's and Lakoff's models narrowly focus on the recipient of the message. Additionally, as far as Leech's approach is concerned, it seems that the number of possible maxims is 25 unlimited. In other words, upon encountering a formerly unexplained irregularity in communication, the theory could be expanded by adding a new maxim, which makes it inelegant. The main line of criticism of Brown and Levinson's theory is that it is culturally very ethnocentric as it is based on western notions of politeness. That is, it assumes that face is a personal concept. However, it has been pointed out that in some cultures, e.g., Japanese, face appears to be an interpersonal notion based on group membership (Gu, 1990; Mao, 1992). Another criticism is that the notion of negative politeness is not applicable in collectivist cultures, such as Chinese, which value an individual's social status in a group rather than individual freedom (Gu, 1990; Mao, 1992). In other words, it seems that Brown and Levinson's notion of universal politeness is hard to support. In addition, Brown and Levinson's formula for calculating the level of threat to face using the three variables: social distance, power, and the degree of imposition, suggests that human interaction is "an activity of continuous mutual monitoring of potential threats to the faces of the interactants, and of devising strategies for maintaining the interactants' faces - a view that if always true, could rob social interaction of all elements of pleasure" (Nwoye, 1992, p. 311). In fact, Nwoye proposes the notion of ‘group face' to discuss politeness in cultures in which the collective image takes priority over self-image. Another problem with this model of politeness is that it focuses on the speaker, and, moreover, assumes that speakers make rational choices when choosing an appropriate politeness strategy, as suggested by the binary system of strategy choices (Watts, 2003, p. 88). 26 One of the main critics of Brown and Levinson's theory, Watts (2003), summarizes the weaknesses of Brown and Levinson's theory and goes as far as to question the validity of distinction between positive and negative face. He points out that rather than being a theory of politeness, they propose a theory of face work. While the utterances themselves may not be inherently polite or impolite, when employed in a certain context they carry out face work and thus "may be interpreted as polite within the context of discourse activity" (p. 95). Thus, face work does not equal politeness. Questioning the term ‘polite,' Watts instead proposes that utterances can be ‘politic,' or, in other words, appropriate in a given situation. More specifically, he argues that because linguistic forms are not intrinsically polite or impolite, linguistic behavior should be evaluated based on whether it comprises with the expected norms. That linguistic behavior that is perceived as appropriate in the light of the social expectations should be referred to as ‘politic behavior' (2003, p. 19). Overall, Watts (2003) proposes a discursive approach to politeness whose goal is to ‘recognize when a linguistic utterance might be open to interpretation by interactants as (im)polite' (p. 143). As such, his approach appears more comprehensive. Politeness and the speech act of request The speech act of request is one of the most widely studied speech acts (see Chapter 2 for literature review). Recall that Brown and Levinson categorize requests as FTAs, i.e., acts that threaten the face, and thus call for a redressive action. This is because performing a request involves high social stakes for both the speaker and the hearer (Blum-Kulka, House & Kasper, 1989). The various risks associated with performing a 27 request stem from the fact that the speaker wants a resource that the hearer possesses (whether it is related to material gains, monetary gains, time or talent), and include both a possible refusal on the part of the hearer to grant the request and an infringement upon the hearer's freedom of action. At the same time, the range of linguistic means available for performance of requests is incredibly broad (recall Searle's [1975, 1979] list of strategies to perform conventionally indirect strategies in English as an example), and the selection of the appropriate strategy has to be based on a careful evaluation of the context of the locution and the involved stakeholders. Such a task is extremely complex as the interlocutors employ both the knowledge of grammatical structures and the relevant social and contextual features. If performance of such an intricate speech act as a request is a complex task for native speakers, it must pose serious challenges to second language learners, too. First, they have to carefully consider what language forms are available to them. In earlier stages of second language acquisition these may be quite limited; yet later on, even with progressing development of proficiency, the applicable linguistic forms may not be fully acquired. Second, learners have to consider the context of the utterance, and the relationship between themselves and the interlocutor. They have to be able to assess their own social role, the interlocutor's social role, the social distance between them, and the situation in which the exchange occurs, to name just a few factors. As Kasper and Rose point out, whereas learners are conscious of the need to attend to politeness, "even when their interlanguage lexicon and grammar contain materials deployable for internal speech act modification, [they] often do not use this material for such pragmalinguistic functions [because] internal modification through grammaticalized material requires a highly 28 developed control of processing" (2002, p. 26) The fact that, as will be seen in the next section, the social factors affecting communication may be interpreted differently in different cultures, only adds to the complexity of performing a request in a second language. Cross-cultural differences in politeness That all languages and cultures possess linguistic and nonlinguistic means allowing them to adjust their communication to situational and social context is a widely accepted fact. Language speakers vary their style in accordance with the formality of the situation, the social status of the parties involved, and the purpose of the interaction. However, Brown and Levinson's proposal that certain cultures tend to employ one type of politeness over another (1978, 1987) has been extensively questioned. Their claim that some cultures show preference for positive politeness strategies, whereas others prefer negative politeness strategies has been criticized as ethnocentric, i.e., created from the perspective of a Western English-speaking culture. As stated previously, one explicit critique can be found in work that suggests that Eastern cultures such as Chinese and Japanese stress the importance of the collective image, and their way of creating politeness may be better accounted for in terms of ‘group face' (Nwoye, 1992). In their discussion of the issue of universality in performance of speech acts, Blum-Kulka, House and Kasper (1989) suggest that the notion of politeness is "culturally relativized" (p. 24). For example, direct and indirect strategies may carry different politeness values in different cultures. Two of the goals of their Cross-Cultural Speech Act Realization Project (CCSARP) were to shed more light on realization of speech acts 29 (requests and apologies) across different languages and to investigate the similarities and differences of strategies used to performed speech acts be native and nonnative speakers. Although CCSARP will be discussed in more detail in Chapters Two and Three, it should be noted here that the findings from this project support the cultural relativity of request and apology strategies. Their results also suggest that members of a culture share mutual expectations about what linguistic behavior is appropriate in various situations and that conventional indirectness varies cross-culturally (p. 66). House (2006) points out that the behavioral norms and expectations in a given community constitute politeness, i.e., the behaviors that go unnoticed because they are the default. On the other hand, both over-politeness and impoliteness are noticed because they are inappropriate. House suggests that the differences between cultural norms and expectations can cause clashes, misunderstandings, and mutual accusations of impoliteness between speakers of different languages precisely because they are perceived as either overly polite or impolite (p. 260). Using five dimensions of cross-cultural differences: directness/indirectness, orientation towards Self/orientation towards Other, orientation towards Content/orientation towards Addressees, Explicitness/Implicitness, and Ad-hoc formulation/Verbal routines in her comparison of German and English politeness strategies, House concluded that shared cultural conventions underpin speakers' linguistic choices, and that "linguistic differences in realizing discourse may be taken to reflect deeper differences in cultural preference and expectation patterns" (p. 264). House's statement has important implications for the present study. If preferences for certain politeness strategies are rooted in cultural norms and expectations, and if 30 differences between cultures can cause misunderstanding and clashes in intercultural interactions, then understanding the roots of these cultural differences may be helpful in explaining differences in communication styles. Since the research proposed here is going to focus on electronic requests in American English and Russian, the following section considers some cultural factors that can contribute to the choice of politeness strategies in the respective languages. American English vs. Russian Any American who has either had interactions with native Russians, or has spent some time living in Russia will admit that Russian culture is very different from American culture, or that Russian and American cultures differ from each other. Issues that are frequently commented on range from the Russians' profound hospitability and intimate relations with friends to their fatalistic view of life and nature, pessimism, and tendency to complain which has even been termed ‘the discursive art of suffering' (Ries, 1997, p. 83). Conversely, Russians commonly perceive Americans as strong-willed, independent, and optimistic on one hand, and fake, overly excitable, and unreliable on the other (Zatsepina & Rodriguez, 1999). These views result in the fact that each culture perceives the other through the proverbial glasses of its own values, norms, expectations, and prejudices. The importance of culture in cross-cultural pragmatics cannot be understated. Learning a second or a foreign language is inseparable from learning the culture of the peoples who speak it as their native tongue, and learners' attitudes towards the target culture can even affect their levels of attainment (Schumann, 1978). While the present 31 study does not attempt a contrastive analysis of the American and the Russian culture, understanding the respective cultural values can shed some light on the linguistic choices made by the study participants. A great deal has been written on the topic of cultural differences between Russians and Americans. Kartalova (1996) suggests that Americans value self-reliance, choice, informality, security, self-determination, self-control, individual responsibility, success, punctuality friendliness, and respect of personal boundaries. Russians, on the other hand, may value involvement, hospitality, generosity, trust, concern, sincerity, directness, intimacy, loyalty, emotional commitment, spontaneity, flexibility, and inner freedom of feelings and thoughts pertaining to morality. Kohls (1994) provides a detailed list and elaborate definitions of the core American values, his purpose being to facilitate the introduction of new immigrants to the American society. The first of the values on his list is personal control over the environment - the belief that people are able to control nature and that with persistence, anything can be accomplished. Seeing change as positive is the second value on the list. According to Kohls, unlike other cultures which perceive change as ‘a disruptive, destructive force' (p. 2), Americans value change and link it to progress, development, and improvement. The next value on Kohls' list is control of time: Americans follow schedules and are wary of interruptions that prevent them from finishing what they have planned. They stress the importance of using time wisely, and of attaining goals. It is because of these first three values that Americans cherish hard work, responsibility, and accomplishment. 32 Kohls claims egalitarianism to be one of the most important American values. By this he means that Americans believe everybody should have an equal opportunity to succeed, even if success is not measured in the same way for everyone. Individualism and privacy, on the other hand, explain why many Americans perceive their thoughts and acts as highly individualistic, as well as why they view time spent alone as a positive condition. Kohls defines privacy as ‘the ultimate result of individualism' (p. 3). The next value discussed by Kohls is what he refers to as self-help concept. The self-help concept means individuals can take credit only for their own accomplishments, but also that they take pride in the fruits of their labor. Related to it is the value of competition and free enterprise, because it motivates individuals to producing their very best. Kohls points out that this value may cause a cultural clash in individuals who come from cultures that value cooperation rather than competition (p. 4). In addition to the self-help concept and the values of competition and free enterprise, Kohls mentions the value of future orientation and action/work orientation. The former one is expressed in the belief that hard work will bring improvements in the future and greater happiness. The latter leads Americans to plan very busy, full schedules, including recreation schedules. Kohls claims that this particular value can be blamed for creation of ‘workaholics' but also notes that it also allows Americans to take pride in their work, even if it is physical, unskilled work that may not gain one respect in many other cultures (p. 5). The next two values on the list are informality and directness, and openness and honesty. Kohls notes that because Americans tend to be very informal in comparison with other cultures, they may be perceived as disrespectful. Informality is expressed in both 33 the forms of address and linguistic choices as well as in behaviors and dress styles. In respect to the value of directness, openness and honesty, Kohls actually makes a reference to saving face. He comments that Americans tend to be more direct than people from other cultures, but the choice of direct strategies is not based in the desire to make the interlocutor lose face. Rather, they choose directness because they associate it with honesty and sincerity (p. 6). The final two values discussed by Kohls are practicality and efficiency and materialism/acquisition. Due to the first one, Americans tend to be realistic, practical, and efficient. Practicality is, for them, one of the most important factors affecting important decisions. Kohls credits this value for American contributions to innovations and inventions, as well as for the fact that Americans shun from being sentimental or irrational( p. 7). The value of materialism/acquisition, on the other hand, explains why Americans value material objects and may view them as a reward for their hard work. They also typically value ownership, and often replace older possessions with new ones, even if the old ones are still functional (p. 9). As far as Russian cultural values are concerned, Bashkirova (2001) suggests the following hierarchy of significance based on the data from the 1995-1999 World Values Survey: family, work, friends and acquaintances, free time, religion, and politics. Family is one of the most traditional values in Russia, and most Russians see it "as a necessary condition for happiness" (p. 7). Having a family implies being married, having offspring, and sharing a household and responsibilities. This goes hand in hand with having a reliable, well-paying job. Social involvement is third in the ranking: friends are an important aspect of Russians' daily interactions, and refusing a friend a favor, no matter 34 how imposing, is an uncommon practice. With regards to the last three values in the hierarchy, Bashkirova reports a fair amount of disagreement among the respondents. Thus, it is doubtful that free time, religion and politics can be considered universal cultural values in Russia. In her comparison of American and Russian cultures, Kartalova (1996) contrasts American independence with Russian involvement. She considers a different set of cultural values from that of Bashkirova, namely hospitality and choice, external and internal personal boundaries, friendliness and intimacy, and courtesy and respect. Americans and Russians differ in regards to all of these. First, whereas Americans value freedom and independence of choice, this value can cause a cultural clash when they encounter Russians expressing hospitality and concern towards guests by violating their freedom to choose how much and what they want to consume (as perceived by American guests). Secondly, conflicts may arise when it comes to personal space and privacy boundaries. This is because Americans do not find it appropriate to discuss with acquaintances issues related to money, dating, and religion, but Russians do. Whereas Americans are very particular about their personal possessions, Russians are more open to allow others to share theirs. For example, Russians are less particular than Americans about their private space (p. 79) and for Russians, lending and borrowing money is not as fraught as it is for Americans (p. 80). Moreover, the two cultures differ in terms of the degree of informality and friendliness. In comparison with open, involved, overbearing Russians, Americans appear reserved, closed, and superficial in their relationships with others. Finally, clashes may arise from different perspectives of courtesy and respect, especially when it comes to the views on appropriate male and female behaviors. For 35 example, Russian women may expect men to open a door for them or help them put on a coat, whereas such behavior may be perceived as a violation of personal independence by American women. Bashkirova (1996) claims that the differences in cultural values described in the previous paragraph influence Russian and American views on successful oral communication. For example, she suggests that American ‘superficial' greetings are rooted in the fact that for Americans, independence is more valuable than involvement. On the other hand Russians, who expect more intimacy, may view warm greetings as an invitation to a more involved relationship. Another example provided by Bashkirova concerns complaining which is viewed as inappropriate by Americans (see Kohls' [1994] values of personal control over the environment, self-help concept, and future orientation), whereas for Russians it may imply concern and involvement with others. Americans are often taken aback by pessimistic Russians who seem to think the world is going to end tomorrow. The discussion of cultural values as presented by Kohl, Kartalova, and Bashkirova is not meant to present a comprehensive overview of Russian vs. American culture, nor can their views be trusted as completely objective. Kohl clearly privileges certain American values to present a positive image to immigrants, and Kartalova and Bashkirova seem to focus on traditional ‘core' values, neglecting others that may be perceived as negative by Russians (such as being rich, which is traditionally frowned upon, yet more recently taken as a token of success). Overall, however, it seems that Russians and Americans live in very different cultural worlds. To what extent these values might impact linguistic choices when 36 formulating requests, never mind communicate, is beyond the scope of this study. Nevertheless, some of the differences in cultural values presented here are brought up in the discussion because, as will be seen below, the study findings indicate that responses produced by American subjects are rated as less socially appropriate than the ones produced by Russians. Interlanguage pragmatics Communicative competence The previous section has shown that cultural values can influence not only nonverbal but also verbal behavior. As such, they are an important component of the knowledge nonnative speakers must obtain to successfully communicate in their L2. Interlanguage pragmatics is "the branch of second language research which studies how nonnative speakers […] understand and carry out linguistic action in a target language, and how they acquire [second language (L2)] pragmatic knowledge" (Kasper, 1992: 203). It has also been described as "the study of nonnative speakers" use and acquisition of linguistic action patterns in a second language' (Kasper & Blum-Kulka, 1993: 3). It is defined more narrowly by Kasper and Dahl (1991) as "the performance and acquisition of speech acts by L2 learners" (p. 216). Interlanguage pragmatics is interested in illocutionary force and politeness of speech acts performed by nonnative speakers, in addition to the development of communicative competence. The first studies in interlanguage pragmatics emerged in the late 1970s and early 1980s following the publication of Hymes's ‘On Communicative Competence' (1972) and Canale and Swain's ‘Theoretical Bases of Communicative Approaches to Second 37 Language Teaching and Testing' (1980). Campbell and Wales (1970) and Hymes (1972) were among the first linguists who rejected Chomsky's (1965) definition of competence, which assumes an ideal speaker and hearer and focuses on grammatical correctness of decontextualized sentences. Chomsky defines the goals of linguistics in the following way: Linguistic theory is concerned primarily with an ideal speaker-listener, in a completely homogenous speech community, who knows its language perfectly and is unaffected by such grammatically irrelevant conditions as memory limitations, distractions, shifts of attention and interest, and errors (random or characteristic) in applying his knowledge of the language in actual performance. (p. 3) Thus, Chomsky proposes that linguistics should be preoccupied with providing an account of the language users' unconscious yet perfect knowledge of grammar structure. Chomsky considers performance, or how language is used in context, on the other hand, to be full of flaws, imperfections and deviations from grammar rules (p. 4). Because performance is the only observable evidence of competence, linguists have to rely on it to reconstruct the underlying rules of grammar. However, Chomsky's theory is primarily concerned with ‘discovering a mental reality underlying actual behavior' (p. 4) and thus performance is of secondary importance. While acknowledging that performance might be an imperfect rendition of some underlying system, Campbell and Wales (1970) and Hymes (1972) argued that Chomsky's theory is problematic because it defines competence in isolation from sociocultural factors (1972, p. 271). Hymes pointed out that even though Chomsky's concept of performance is used to account for language use in context, it is not central to the theory, and the theory consequently generates an image of a speaker as "an abstract, isolated individual, almost an unmotivated cognitive mechanism, not, except incidentally, 38 a person in a social world" (p. 272). Instead, Hymes proposed that a part of speakers' knowledge of language, which he refers to as communicative competence (p. 281), is the knowledge of what is contextually appropriate. He noted that communication does not take place in homogenous communities and asserted, "social life has affected not merely outward performance, but inner competence itself" (p. 274). The significance of this definition of competence lies in the fact that in addition to including the knowledge of how to produce grammatically correct sentences, it allows the speaker to know ‘when to speak, when not, and […] what to talk about with whom, when, where, and in what manner' (1972, p. 277). In other words, competence is the knowledge of the rules of language use which include not only the knowledge of grammar in Chomsky's sense, but also the ability to select language that is context appropriate. Competence is linked to the speaker's attitudes, values, linguistic knowledge, and social experience (Hymes, 1972, p.288). The actual meaning of utterances is context-dependent, with the context broken down into the following components: setting, participants, ends, act sequence, key, instrumentalities, norms of interaction and interpretation and genre (1974, p. 62). Several frameworks describing language competence have been proposed since (see, for example, Saville-Troike's [1982] ethnographic view; Gumperz's [1982] ethnographic perspective; and Celce-Murcia, Dornyei and Thurrell's [1995] second language acquisition model). For the purpose of this study, the following three frameworks are the most relevant: the widely accepted Swain and Canale's (1980) and Canale's (1983) model, Celce-Murcia, Dornyei and Thurrell's (1995) model (also discussed in Celce-Murcia and Olshtein [2000]), and Bachman and Palmer's (2010) recently revised approach. 39 Swain and Canale (1980) understand communicative competence as the underlying knowledge required for communication, and "the relationship and interaction between grammatical competence, or knowledge of the rules of grammar, and sociolinguistic competence, or knowledge of the rules of language use," and distinguish it from the communicative performance which they define as the ‘production and comprehension of utterances' through the application of the competencies (p. 6). Swain and Canale distinguish three components of communicative competence: grammatical, sociolinguistic and strategic. Grammatical competence includes the knowledge of the lexicon and the rules of phonology, morphology, syntax and sentence-level semantics (p. 29). Grammatical competence allows language users to both decode and encode the literal meaning of utterances. Sociolinguistic competence is comprised of the knowledge of sociocultural rules, which allows speakers to produce utterances that are appropriate in a given context, and the knowledge of discourse rules, which Swain and Canale broadly define as cohesion and coherence and structuring utterances in terms of topic and comment (p. 30). Sociolinguistic competence helps language users consider factors such as the number and status of interlocutors, as well as the purpose and norms of communication in context and to choose form (verbal and nonverbal) and the meaning appropriate for a given situation. Finally, strategic competence is the ability to overcome any difficulties that arise in communication using both linguistic and nonlinguistic means to compensate for deficiencies in one's interlanguage; for example, the ability to paraphrase or to address strangers when their social status is unknown (pp. 30-31). 40 Swain and Canale's grammatical competence is thus equivalent to Chomsky's competence, but the other two components, sociolinguistic and strategic, are not present in Chomsky's framework. It is important to stress that Swain and Canale distinguish between ‘communicative competence' and ‘communicative performance.' Communicative competence, in their view, refers to both knowledge of language and the ability to use this knowledge when communicating. Communicative performance, on the other hand, is the realization of communicative competence under limiting psychological constraints such as tiredness, nervousness, and background distractions. What Swain and Canale stress is that in order to be competent language users, second language learners need to know what native speakers are likely to say in a given situation (p.16). Canale (1983) maintained the distinction between communicative competence and performance, renaming the latter as ‘actual communication.' In this framework, communicative competence is manifested in ‘actual communication' as "the realization of such knowledge and skill under limiting psychological and environmental conditions such as memory and perceptual constraints, fatigue, nervousness, distractions, and interfering background noises" (p. 5). In addition to restating the distinction between grammatical, discourse, sociolinguistic and strategic competence, Canale (1983) adds discourse competence, which he defines as the knowledge "of how to combine grammatical forms and meanings to achieve a unified spoken or written text in different genres" (p. 9). In other words, discourse competence is the speaker's ability to use cohesion and coherence to create unity in different types of oral and written texts. This entails using cohesion devices (conjunctions, pronouns, synonyms, key term repetitions, etc.) and maintaining 41 relationships among meanings within a text. Notably, Canale (1983) also further stressed the importance of sociolinguistic competence in language users' ability to interpret ‘social meaning' of utterances, such as function and attitude, when these cannot be interpreted based solely on the literal meaning. Thus, using Grice's terminology, sociolinguistic competence plays an essential role in both generating and interpreting conversational implicature. The second model of communicative competence to be discussed here it the one proposed by Celce-Murcia, Dornyei and Thurrell (1995). Their model was also developed from an L2 perspective and includes five components: discourse competence, linguistic competence, actional competence, sociocultural competence and strategic competence. Discourse competence is centrally located and interacts with linguistic competence, socio-cultural competence and actional competence as discourse is created. Discourse competence is responsible for ‘the selection, sequencing and arrangement of words, structures, sentences and utterances to achieve a unified spoken or written text' (p. 13). It is supported by cohesion (e.g., anaphora, conjunctions), deixis (e.g., personal pronouns, temporal expressions such as ‘now,' ‘then'), coherence (e.g., theme-rheme organization, sequence of tenses), genre/generic structure (e.g., narrative, lab report, interview), and conversational structure (e.g., performance of closings and openings, conversational collaboration, adjacency pairs). The second element, linguistic competence, involves the knowledge of syntax (including constituent structure, word order, special constructions, coordination, and subordination), morphology (parts of speech, inflections, and derivational processes), 42 phonology (segmentals and suprasegmentals), orthography (letters or other writing symbols, phoneme-grapheme correspondences, spelling rules, punctuation and mechanics conventions), and the lexicon including formulaic constructions (words, fixed phrases, formulaic chunks, collocations, and idioms). The next component, actional competence, allows language users to express and comprehend communicative intent, i.e., the illocutionary force of speech acts, including indirect speech acts. It also includes the knowledge of conventionalized forms, sentence stems, formulaic expressions and language strategies. Actional knowledge is comprised of two major components: knowledge of language functions and knowledge of speech act sets. The first component is broken down into interpersonal exchange (e.g., making and breaking engagements, complimenting), information (e.g., reporting and remembering), opinions (e.g., agreeing and disagreeing), feelings (e.g., expressing happiness or annoyance), suasion (e.g., giving orders, asking for permission), problems (e.g., complaining, regretting), and future scenarios (e.g., promising, predicting). Celce-Murcia, Dornyei and Thurrell hold that actional competence is distinct from sociolinguistic competence, which they define as "the speaker's knowledge of how to express messages appropriately within the overall social and cultural context of communication, in accordance with the pragmatic factors related to variation in language use" (p. 23). The components of sociocultural competence are as follows: social contextual factors (i.e., participant and situational variables such as age, gender, time and place), stylistic appropriateness factors (politeness strategies, level of formality, registers), cultural factors (e.g., background knowledge about the target community, awareness of regional dialect differences), and nonverbal communicative factors (e.g., 43 body language, use of space, silence). Celce-Murcia, Doenyei and Thurrell stress that language "is not simply a communication coding system but also an integral part of the individual's identity and the most important channel of social organization" (p. 23). Sociocultural knowledge enables speakers to use language appropriately considering the social and cultural context in which communication takes place. Finally, strategic competence in the Celce-Murcia, Dornyei and Thurrell model is conceptualized as "knowledge of communication strategies and how to use them" (p. 26). More specifically, strategic competence in this model consists of avoidance or reduction strategies (e.g., avoiding of certain topics), achievement or compensatory strategies (i.e., manipulation of language to attain a communicative goal), stalling or time gaining strategies such as the use of fillers and repetitions, self-monitoring strategies (e.g., rephrasing), and interactional strategies (e.g., appeals for help, comprehension checks). Finally, Bachman and Palmer (2010) propose a modified framework of what it entails to know a language from a second language acquisition perspective. This framework builds upon the former versions they proposed (Bachman & Palmer [1982]; Bachman [1990]; Bachman & Palmer [1996]), and it situates the ability to perform speech acts as one of the subcomponents of language use. They define language use as ‘the creation or interpretation of intended meanings in discourse by an individual' (non-reciprocal language use) and as "the dynamic and interactive negotiation of intended meanings between two or more individuals in a particular situation" (reciprocal language use) (p. 34). Language knowledge and strategic knowledge are two subcomponents of language use which also engages personal attributes, topical knowledge, affective schemata and cognitive strategies (p. 36). 44 In Bachman and Palmer's model, language knowledge is further subdivided into organizational knowledge which includes grammatical knowledge and textual knowledge, and pragmatic knowledge composed of functional knowledge and sociolinguistic knowledge. Grammatical knowledge is comprised of the knowledge vocabulary and the rules of syntax, phonology and graphology. The next component, pragmatic knowledge, is the speaker's ability to formulate and comprehend messages. It includes functional knowledge and sociolinguistic knowledge. Sociolinguistic competence, which allows language users to relate their utterances to the language use settings, includes the ability to use and understand different registers, non-literal language, and cultural allusions (p. 45). Language users need a competence that allows them to produce and interpret language in a given context. Several models of such a competence have been proposed. While Swain and Canale's (1980) is probably the most commonly referred to, Bachman and Palmer (2010) propose the broadest perspective as they situate the ability to utter socially appropriate discourse within language use. The findings contribute to our understanding of one of the components of language use, namely sociolinguistic knowledge. Sociopragmatic and pragmalinguistic failure Lack of contextual appropriateness can lead to communicative problems more severe than errors in grammar would cause: sociopragmatic failure and pragmalinguistic failure. Leech (1983) differentiates between two areas of study in pragmatics: sociopragmatics and pragmalinguistics. He defines sociopragmatics as the study of the 45 ways in which pragmatic performance varies depending on the social context, whereas pragmalinguistics is preoccupied with how language forms are used in the performance of particular speech acts. These concepts have been applied in the field of interlanguage pragmatics. In ‘Cross-Cultural Pragmatic Failure' (1983), Thomas makes a distinction between two ways second language learners fail to produce appropriate illocutionary acts : sociopragmatic (roughly equivalent to what Canale (1983) refers to as ‘appropriateness of meaning') and pragmalinguistic failure (similar to Canale's ‘appropriateness of form'). By sociopragmatic failure, Thomas understands a failure to perform the speech act required in a given situation (e.g., a failure to apologize for what one did). She claims that sociopragmatic failure results from different beliefs about rights and obligations of the interlocutors. Pragmalinguistic failure, on the other hand, takes place when the speaker performs the required act but uses inadequate linguistic means. It results from mistaken beliefs about the pragmatic force of an utterance. An example of pragmalinguistic failure is when a Polish or a Russian speaker of English accepts a compliment by downgrading him- or herself, rather than by accepting the compliment. Lack of sociolinguistic competence can lead to pragmalinguistic failure, which in turn can cause miscommunication and even affect native speaker perceptions of nonnative speakers. One of the goals of the present study is to determine whether linguistic choices made by nonnative speakers of Russian affect the perceived levels of politeness and appropriateness of the electronic requests they write. Whereas Chapter 3 explicates the study design and procedures, the reader should keep in mind that the concept of pragmalinguistic failure may help account for some potential differences between native and nonnative means of constructing requests. 46 Pragmatic transfer It has been proposed that pragmatic knowledge from the first language exerts an influence on the use and acquisition of pragmatic knowledge in the second language (Beebe et al., 1990; Kasper, 1992; Odlin, 1989; Wolfson, 1989). This phenomenon, referred to as pragmatic transfer, affects language use at both the sociopragmatic and pragmalinguistic level. In other words, a learner's pragmatic knowledge of the L1 affects the choice whether or not to perform a speech act, and what linguistic means are selected to carry it out. In parallel to L2 transfer in other areas, positive pragmatic transfer (when pragmatic conventions are shared by the languages) and negative pragmatic transfer (when pragmatic conventions differ between the languages) are distinguished. Because negative pragmatic transfer leads to nonnative like linguistic behavior, "in the real world, pragmatic transfer matters more, or at least more obviously, than transfer of relative clause structure than word order" (Kasper, 1992, p. 205). Email Because the proposed study will examine the use of politeness strategies by native and nonnative speakers in electronic requests, this section focuses on electronic communication via email. It describes email conventions and provides a brief overview of netiquette, or the rules of polite behavior on line. Communication via email on American campuses has been increasing in recent years. Although it has not yet surpassed the use of face-to-face communication (office hours), it is already used more frequently than phone for student to faculty and faculty to student communication (Sheer & Fung, 2007). This poses a particular challenge for 47 nonnative speakers of English who may lack adequate linguistic and pragmatic skills and might be unfamiliar with the rules of netiquette in English. It can be expected that a similar phenomenon is taking place world-wide, and that learners of other languages very likely experience similar challenges. Email conventions As a relatively new channel of communication, email does not have well-specified linguistic conventions. Rather, as a medium that appears to draw on both spoken and written communication, it is not precisely situated on the formal/informal continuum. Research on computer-mediated language (CMD) shows that language used on-line is "less correct, complex and coherent than standard written language" (Herring, 2001). It often contains non-standard features, abbreviations, typos and mixed case, and as such, it presents particular competency challenges for its users. However, there have been some attempts to provide rules of netiquette, or internet etiquette, in both printed sources and on internet web sites. Netiquette Whereas recommendations on polite online behavior exist, the fact is that not all email users observe these guidelines. Rather, it seems that email is a medium of communication that is situated somewhere along the continuum between spoken and written communication. Academic studies of computer-mediated communication suggest that email is a new variety of discourse which combines features of spoken and written language (Georgakopoulou, 2000, p. 94). One the one hand, email is characterized by an 48 informal style, limited planning and editing, fast feedback, and transience of message. On the other hand, it contains features such as lack of visual and pragmatic clues, written format, and the physical absence of the interlocutors, which characterize written modes of communication. Because its conventions are not clearly specified, and because it borders on oral and written communication styles, email is a mode of communication that can pose particular challenges, especially if it has to be composed in a foreign language. Some attempts have been made to establish guidelines for polite email behavior. Commonly recognized as a classic and trusted source when it comes to good manners, Emily Post's Etiquette, 18th Edition, devotes a complete chapter to computer communication. The general guideline stating that "polite electronic communication requires treating others as you would have them treat you, even when interacting in the virtual world" is followed by more specific rules of politeness (p. 240). The rules are as follows: Human contact still matters Watch what you say - and how you say it Be careful when clicking ‘Send' Address with care Send delay What's your subject? Keep it short and sweet No yelling, please Watch those emoticons Check it over 49 In a nutshell, this handbook of etiquette suggests, first of all, that email should not replace personal interaction, but when used, should be carefully crafted. Because email is impersonal in nature, users often forget the good manners that they would normally follow in face-to-face interactions. Therefore, it is important to carefully check the contents and wording of a message before it is sent. It is also vital to respect the privacy of recipients and always consider sending individual rather than group messages as well as to delay sending the message (the guide does not specify any particular reasons for the latter). The next point to keep in mind is to formulate an informative yet succinct subject line and to write a message that is brief and concise. Finally the handbook advises against the use of all capital letters and a careful use of emoticons, which should be reserved for messages to close friends and family members (pp. 238-242). Scheuermann and Taylor (1997) compiled an overview of published netiquette suggestions, and divided them into global netiquette recommendations, the 10 commandments of etiquette, and specific netiquette suggestions, totaling in 20 rules. The first set of what the authors refer to as ‘commandments' are as follows (p. 270): I. Thou shalt not use a computer to harm other people. II. Thou shalt not interfere with other people's computer work. III. Thou shalt not snoop around in other people's files. IV. Thou shalt not use a computer to steal. V. Thou shalt not use a computer to bear false witness. VI. Thou shalt not use or copy software for which you have not paid. VII. Thou shalt not use other people's computer resources without authorization. VIII. Thou shalt not appropriate other people's intellectual output. 50 IX. Thou shalt think about the social consequences of the program you write. X. Thou shalt use a computer in ways that show consideration and respect. Whereas the rules above specify the norms of nonlinguistics online behavior, the ten commandments of etiquette and the specific netiquette suggestions focus more specifically on how language should be used in an online environment. The ten commandments of netiquette (p. 270) recommend that one be polite, brief, proud of the written message, precise in writing subject headings, considerate of audience, and careful with humor and sarcasm. More generally, they also suggest summarizing the content of the message one is responding to, not repeating what has been said (which seems contradictory), giving back to the community (which is vague), and citing appropriate references. Having reviewed several published articles on the topic, Scheuermann and Taylor also list the most frequent linguistic recommendations for Internet users (pp. 270- 271): Think first. Write in upper and lower case. Avoid abbreviations. Be concise. Avoid smileys. Don't flame. Don't take offense easily. Don't evangelize. Know the audience. 51 More specifically, Internet users are advised to delay writing a messages when angry; to use mixed case as it is a sign of respect (all upper case is like shouting, whereas all lower case is difficult to read); to fully write out words as they are easier to read than abbreviations, to be brief in order to show respect for the recipient's time; and to express feelings and emotions with appropriate word choice rather than emoticons. The rules also prompt internet users to think of email as the same as face-to-face communication, in order to avoid rudeness, not to make hasty interpretations about others' messages, to be humble and use reason in argumentation, and finally to become familiar with the audience before posting the message. The authors caution that the rules are not meant to be definitive, and should rather be used as a general guideline. Overall, they stress that individual internet users should strive for politeness and courtesy as the lack of thereof leads to a disadvantage in virtual communication. In conclusion, because email is a relatively new medium of communication, its nature is still ambiguous. Some attempts to establish email conventions and rules of polite online behavior have been summarized above. Nevertheless, given its spoken/written hybridity, selecting appropriate linguistic means for email communication can be a challenge for native and nonnative speakers alike. CHAPTER 2 REVIEW OF LITERATURE Chapter overview The purpose of this chapter is to provide a review of literature and motivation for research questions in the present study. The chapter begins with a section that presents studies on both perception and production of interlanguage requests, with an emphasis on the studies dealing with how nonnative speakers perceive the level of politeness in utterances. Second, the major studies that investigated the production of requests by native speakers and compared the strategies used to those typically chosen by native speakers are reviewed. Third, the chapter reviews research on requests in English and Russian. Next, studies that examined the use of email by second language learners, including studies that focus on electronic requests, are presented. The last part of the chapter presents those few interlanguage pragmatics studies that have investigated the perceptions of nonnative speakers based on the speech acts they produced. In its conclusion, the chapter identifies the gap in the research that the present study aims to fill and states the research questions. 54 Interlanguage pragmatics: speech acts Requests A great body of research on interlanguage pragmatics has been devoted to requests, perhaps because it is an FTA with a high level of imposition on the hearer, and one that poses a threat to the speaker's face. Native and nonnative speakers find themselves performing requests frequently, both in informal interactions and in academic contexts. Requests pose a challenge to second language learners because of their cross-linguistic variation: they can be more or less direct, and are performed using various strategies. In the Cross-Cultural Speech Act Realization Project (CCSARP) (1989), Blum- Kulka, House and Kasper identify a series of components that requests can be comprised of and list eight different strategy types of the performance of the ‘request proper,' i.e., the head act. Thus, in addition to the head act request itself, a request can contain alerters, supportive moves, and internal modifications. Alerters precede the head act and function as attention-getters. Titles, last and first names, nicknames, endearments or offensive terms, personal pronouns, expressions such as ‘Excuse me,' and combinations of these means can serve as alerters. Supportive moves are used to mitigate or aggravate requests. They can occur either before or after the head act, and include getting pre-commitment, giving reasons and explanations of the request, promising a reward, minimizing the imposition, threatening and moralizing. Internal modifications occur within the head act, and include both downgraders (i.e., syntactical, lexical and phrasal devices that soften the impact of the request) and upgraders (i.e., devices that increase the impact of the request). Examples of downgraders include the use of interrogative, subjunctive or conditional 55 (syntactic), and politeness markers such as ‘please' or ‘Do you think…,' hedges (e.g., avoiding precise propositional specification), downtoners (e.g., ‘perhaps' or ‘possibly'), and appealers (e.g., ‘okay?') (lexical and phrasal). Further examples of external modifications are provided in Table 1. The head act itself can also be realized using a variety of strategies. Bloom-Kulka et al. (1989) list three levels of directness and nine strategy types: direct strategies (mood derivable, performatives, hedged performatives, obligation statements, and want statements); conventionally indirect strategies (suggestory formulae and query preparatory); and nonconventionally indirect strategies (strong hints and mild hints). Direct requests are the easiest to identify because the hearer is not required to make inferences in order to interpret them. The interpretation of conventionally indirect requests, on the other hand, depends on conventions of language, conversational Table 1: External modifications of requests Modifications Examples Lexical downgraders (optional additions to soften the force of the request by modifying it through individual words or phrases) Please, possibly, just Syntactic downgraders (mitigating the force of the request by changing the syntax) ‘could' / ‘couldn't instead of ‘can' ‘would' instead of ‘be' (e.g., Would it be possible vs. Is it possible) Upgraders (elements used to increase the impact of the request) ‘right now' Mitigating supportive moves (giving a reason, promising a reward) ‘I missed the test because I had a toothache' ‘I didn't realize that we needed to take a test on subordination and I missed the deadline' Aggravating supportive moves (threatening, demanding, facts) ‘You have to give me an opportunity to improve my grade' ‘My grade for subordination is 50%' 56 principles, pragmalinguistic conventions, and contextualized conventions (pp. 38-39). In Blum-Kulka's words, conventionally indirect requests are ‘based on general, often tacit consent in regard to both patterns of behavior and the meaning assigned to those patterns (p. 38). Whereas conventionally indirect requests can be interpreted based on the conventions established in the speech communities, nonconventionally indirect requests have to be interpreted based on contextual clues. Blum-Kulka et al. point out that as one moves along the scale from direct to nonconventionally indirect strategies, the process of identifying an utterance as a request becomes longer (p. 18). The summary of request strategies and strategy examples are provided in Table 2. Table 2: Summary of different levels of directness used to realize request proper (adapted from Blum-Kulka, House and Kasper, 1989, p. 18) Level of directness Strategy Example Direct Mood derivable (imperatives) Walk the dog! Performatives (the illocutionary force is explicitly named) I am telling you to walk the dog. Hedged performatives I would like you to walk the dog. Obligation statements You'll have to walk that dog. Want statements I want you to walk that dog. Conventionally indirect Suggestory formulae How about walking the dog? Query preparatory (references to ability and willingness) Would you mind walking the dog? Nonconventionally indirect Strong hints (partial reference to object or element that requires the act) The dog is full of energy. Mild hints (no reference to object or element that requires the act) It's gorgeous outside! 57 In his review of the studies of requests, Ellis (2003) describes two important sociolinguistic aspects of requests. The first characteristic is that requests require a significant amount of consideration for face due to their high level of imposition. To choose appropriate linguistic means to perform a request, the speaker needs to consider the relationship with the listener, the context, and the degree of imposition. The second aspect of requests is that different languages show different preferences for strategies conventionally used to perform requests. These choices affect both the head act as well as external and internal modifications (p. 168). Studies on perception of requests Pioneering research on requests was conducted by Clark and Lucy (1975) and Gibbs (1979). These researchers were concerned with the mental processing involved in the interpretation of direct and indirect requests. Clark and Lucy postulated that comprehension of indirect requests is a longer and more complex task than interpretation of direct requests, as the listener has to deduct the intended meaning from literal meaning. Their findings confirm this prediction. Gibbs, however, argues that the results of the Clark and Lucy study are valid only if subjects are tested on the processing of individual sentences. If, however, context is provided, the time needed to comprehend direct and indirect requests should be comparable. In other words, with the help of contextual clues, one does not have to compute the literal meaning of an utterance before understanding its indirect meaning. Gibb's findings were confirmed by Erwin-Tripp et al. (1987) in experiments with child learners. In this study, both the interpretative model (indirect speech acts are first 58 interpreted literally) and the contextual model (the situation in which an indirect act is used as a basis for its interpretation) were tested. The study concluded that while younger children needed more explicitness, in general, demands for action (requests) can be interpreted contextually, without explicit language. Walters (1979), Carrell and Konneker (1981), Fraser, Rintell and Walters (1980), Tanaka and Kawade (1982), Olshtain and Blum-Kulka (1985), and Van der Wijst (1995) have all researched the perception of requests. Walters (1979) compared how native speakers of American English and nonnative speakers of English perceive politeness in requests in English. The L2 subjects in this cross-sectional study came from 17 language backgrounds, including languages as diverse as Chinese, Armenian and Spanish. The subjects were given pairs of requests and asked to decide which member in each pair was more polite. The results suggest a high correlation between native and nonnative perception of politeness. However, nonnative judgments of extremely polite strategies (e.g., ‘may,' ‘could') and extremely impolite (e.g., imperative) were closer to native female than to native male judgments. Nonnative speakers also exhibited a higher degree of unanimity in their rankings of politeness strategies than did native speakers. These findings suggest that advanced learners of English can perceive the politeness of request strategies almost as well as native speakers do. The results of Fraser, Rintell, and Walters (1980) also suggest that native and nonnative speaker perceptions of deference in requests are similar. In this study, native speakers of English and Spanish-speaking learners of English evaluated requests made in situations that varied in terms of the social status and gender of participants and the levels of imposition of the request. The requests in English were judged by native and nonnative 59 speakers of English, and the requests in Spanish were evaluated by native speakers of Spanish. While in general the judgments of the three groups were similar, the Spanish requests were evaluated as more deferential by the learners than the English ones. The results of this study suggest that, when it comes to perceptions, pragmatic transfer of L1 rules to L2 may not play as important a role as is commonly thought. Interesting findings were also obtained by Carrell and Konneker (1981), who investigated how native and nonnative speakers of English judge levels of politeness in English requests. They created an eight-level hierarchy of requests based on syntactic and semantic features (mood, modals, and tense of modals). They assumed, based on Lakoff (1977) that the level of politeness increases in the following way: imperative mood > declarative mood > interrogative mood. The subjects were asked to rank eight request strategies from least polite to most polite. As in other studies, a high correlation was found between native and nonnative judgments. However, while the native group distinguished five levels of politeness, the nonnative group distinguished eight. For example, declaratives with a past tense modal (‘I'd like X') and interrogatives with no modal (‘Do you have X?'), which were ranked as equal by native speakers, were perceived as different by nonnative speakers. Carrell and Konneker suggest that this fact may be due to an ‘over-sensitivity' to politeness resulting from language instruction, namely from the students' belief that distinct forms carry distinct meaning. These results are confirmed by Tanaka and Kawade (1982), whose subjects (American and Japanese) were also asked to rank a set of request sentences in English. Their results, too, suggest that advanced English learners perceive politeness of requests in a way similar to native speakers, yet with more distinct levels. 60 Van der Wijst's research (1995) stemmed from the observed difficulties in French-Dutch business communication caused by participants' perceived impoliteness which had been ascribed to differences in directness (e.g., Ulijn and Gorter (1989) and Merk (1987)). More specifically, Van der Wijst speculates that Dutch communicating in French may select expressions that are considered impolite by their French interlocutors (p. 484). This study consisted of an experiment in which Dutch and French participants were asked to rank the same request expressed using 19 formulations from least to most polite. Since the study found no significant differences in French and Dutch rankings, Van der Wijst concludes that Dutch and French perceive politeness in similar ways, and suggests the close similarities between the Dutch and the French culture as a potential explanation. Shcherbakova (2010) explored how Russian learners of English (in English as a Foreign Language [EFL] context) perceive appropriateness and politeness of English utterances in various social situations. The study specifically focused on different the effect of different linguistic forms in the head act of requests. The results suggest that Russian EFL learners judge less conventionally indirect requests (e.g., ‘Can you…?') as less appropriate in interactions with superiors (student-professor) than with an equal (student-student). More conventionally indirect requests (e.g., ‘I was wondering if I could') were viewed as appropriate in student-professor interactions. While this study suggests that Russian learners of English may have an ‘intuitive awareness' of what language structures are appropriate in different social contexts (p. 75), it does not demonstrate their ability to correctly use these structures. 61 Production studies The next group of studies focused on production of requests by both native and nonnative speakers. Some studies of production of requests have focused on the differences between native and nonnative preferences for different types of politeness strategies. Most of them have used DCT (discourse completion task) or role plays as the elicitation method, and include Brunak and Scarcella (1979), House et al. (1989), Kasanga (2006), and Hacking (2008). In a pioneering study on pragmatic competence, Brunak and Scarcella (1979) examined the use of negative and positive politeness by Arabic speakers of English in role-play situations. In contrast to native speakers, the nonnative subjects in the study used in-group terms of address and endearment, but almost no slang or ellipses, and no other in-group language such, as ‘ya know,' ‘I mean,' or inclusive ‘we' (positive politeness). They also used fewer or inappropriate hedges, statements of personal desire (‘want,' ‘would like') and direct strategies instead of hints, and almost no exclusive ‘we' (negative politeness). The study concludes that unlike native speakers of English, nonnative speakers' range of politeness features and their ability to differentiate the use based on the social context is limited. Additionally, Brunak and Scarcella pointed out that L2 learners use politeness features without having acquired their co-occurrence and distribution rules. As even the author stressed, however, the results of this study have to be interpreted with caution, as all nonnative subjects were males with the same L1 background. Mills (1993) compared English and Russian requests produced by native speakers of English. The study collected natural speech samples from advanced learners and native speakers of Russian as well as responses on a DCT. The analysis revealed a preference 62 for conventionally indirect requests (interrogatives with negation of the hearer's intention to perform the requested action, e.g., ‘Ty menja ne podvezeš?' ‘You won't give me a ride') in the Russian data. According to Mills, such requests are highly inappropriate in English. While this requestive strategy was also used in nine nonnative speaker requests, this group of subjects also tended to use extensive justification for requests, lexical politeness markers (‘požalujsta,' i.e., ‘please'), and the literal Russian equivalent of conventionally indirect requests in English (‘Could you?' or ‘Can you?'), which is unconventional in Russian. The study concludes that these errors can be largely attributed to pragmatic transfer from English. While the constructions (‘Can you? or ‘Could you') are grammatical in Russian, they are not conventionally used as request. Rather, they refer to one's ability to perform a task. For this reason, they should be considered pragmatic rather than grammatical transfer errors. Within CCSARP, House (1989) investigated the distribution of ‘please' and ‘bitte' in English and German requests. The data used for this study came from native speakers of English, native speakers of German, and German learners of English. While there was no systematic and significant difference in how this marker of politeness was distributed in native English, native German, and nonnative English data, the German native speakers and German English learners used ‘bitte' and ‘please' relatively more frequently than native English speakers in certain situations. Most importantly, the study concluded that ‘please' and ‘bitte' are more likely to occur in what House refers to as ‘standard situations,' i.e., those in which the hearer has a high obligation to comply with the request, the speaker has a strong right to pose a request, and the difficulty of performing the request is low. In the CCSARP data, these requests were typically 63 performed by the most direct requestive strategy, the imperative, or by a Query- Preparatory strategy (an ability question such as ‘Can you' / ‘Could you'). In both cases ‘please'/'bitte' was used as a downgrader. Kasanga (2006) used combined methods (ethnographic observation, DCT data, and native-speaker judgments) to analyze request strategies of native Afrikaans speakers learning English. The results of this study suggest that learners' repertoire of requestive strategies is smaller than that of native speakers and that they show preference for direct requests, which are often perceived as rude by native speakers (p. 141). Although this study did not investigate the effect of interaction with native speakers on the communicative competence, it concluded that this relationship should be examined. Request strategies were also investigated by Lin (2009) who compared native Chinese, native English, and nonnative English (with Chinese as L1) data collected in a discourse completion task. The study suggests that, while conventionally indirect requests and similar modal verbs are used in both languages, the native and nonnative requests can differ in terms of form, function and distribution of these strategies. For example, even though Chinese has the equivalent of ‘Would you mind…?' or ‘Would you like to…?,' these structures are not commonly used in requests (they were used rarely by the nonnative speakers of English in the study). The study concludes that further research is needed of not only head acts, but also of external modifications that support them. Studies that investigated the effect of social distance on a speaker's choice of politeness strategies in requests include Tanaka and Kawade (1982) and Tanaka (1988). Tanaka and Kawade presented their subjects with 10 request situations and asked them to select the request they considered the most appropriate in the given social context. The 64 situations were differentiated in terms of the distance between the speaker and the addressee. Overall, this study found that both native and nonnative speakers (advance