Misunderstandings in English as a Lingua Franca: An Analysis of ELF Interactions in South-East Asia (Chapters 1 to 3) David Deterding Berlin: De Gruyter Mouton 2013 ISBN: 973-3-11-028651-9 See : http://www.degruyter.com/view/product/183358 Chapter 1 Introduction This book is about misunderstandings in English as a Lingua Franca (ELF), so it is best if my use of these terms is immediately explained. First, misunderstandings. In this book I will use the term to include all cases where a listener does not understand something that is said to them. Note that this does not necessarily involve a breakdown in communication, as the interaction often proceeds quite smoothly even when a few words are not understood. Nevertheless, in considering what contributes to intelligibility, it is informative to analyse the words that are not understood even in cases where the listener can in fact grasp the overall gist of the utterance and so the conversation appears to continue without a problem. The methodology of identifying such instances of misunderstanding will be discussed in Chapter 3. Next, ELF. Seidlhofer (2011: 7) defines it as “any use of English among speakers of different first languages for whom English is the communicative medium of choice, and often the only option”. Note that this is a fairly broad definition, as it includes native speakers so long as their interactional partners are not native speakers. An alternative is that by House (1999: 74) who defines ELF interactions as “between members of two or more different linguacultures in English, for none of whom English is the mother tongue”. This alternative definition excludes native speakers. Actually, the concepts of mother tongue and native speaker are not easy to define (McKay 2002: 28; Davies 2003). For example, there are some people in Brunei who grow up speaking English at home and it may be their best language although they also regularly speak Malay. So should they be classified as native speakers or not? In describing my data, I will make reference to the Three Circles model of English proposed by Braj Kachru (1985). The conversations analysed in this book do not include speakers from Inner-Circle countries such as Britain, the USA or Australia; but I will not worry about whether the participants are native speakers or not. So the term ELF as it is used in this book can be understood as follows: it is the English used by speakers from postcolonial Outer-Circle countries such as Brunei, Malaysia and Nigeria as well as Expanding-Circle places such as Taiwan, Japan and Laos when they are conversing with speakers from other countries in the Outer or Expanding Circles. 2 1. Introduction I should emphasize that this indicates the scope of the data analysed in this book but it does not attempt to provide a definition of ELF. Other researchers will adopt alternative, perfectly valid definitions of ELF that are appropriate for their work, and in many cases they will include speakers from the Inner Circle to a certain extent. Furthermore, the focus on speakers coming from different countries is not one shared by all researchers into ELF. It just characterizes the data on which my research is based. This book, then, is about the factors that cause misunderstandings to occur in spoken interactions in English between people from different countries in the Outer and Expanding Circles. In addition, it considers how those misunderstandings are dealt with and how misunderstandings in ELF conversations are avoided. 1.1. Scope of the investigation Most people would probably agree that intelligibility is of primary importance in ELF interactions: if you cannot be understood, then your language use is not successful. However, intelligibility is actually a complex, multifaceted concept (Nelson 2011), something I will consider in greater depth in Chapter 2. The research in this book builds on the seminal work of Jenkins (2000) in analysing what enhances and what hinders intelligibility in ELF interactions, though I will extend the investigation beyond her main focus on pronunciation to consider grammatical and lexical issues, and also to discuss how misunderstandings are dealt with and avoided. Much research on misunderstandings in ELF tends to investigate breakdowns in communication: it analyses data to find out what causes a breakdown in communication to occur and how it is repaired. This book is rather different. Most of the participants themselves were subsequently involved in the transcription or else they contributed to the analysis by providing invaluable feedback about what they had said and also what they had not understood. This has allowed me to find numerous instances where a participant does not understand something but where the conversation continues with no apparent problem, and I only know about the misunderstanding as a result of the subsequent feedback from the participants. This provides a rich source of data on intelligibility. Even though the majority of the instances that I will analyse involve no overt breakdown in communication, nevertheless it is important to consider which features of speech have an impact on intelligibility, and what it is about the pronunciation, lexis, 1.1. Scope of the investigation 3 grammar, and general patterns of usage that have caused certain words or phrases not to be understood by the listeners. At this point, it is important to emphasize that misunderstandings are very much the exception rather than the rule in my data, and the conversations that constitute my corpus generally proceed smoothly. I will therefore endeavour not just to consider features of speech that cause misunderstandings to occur but also those that serve to enhance intelligibility. Examples of non-standard features (in terms of Inner-Circle Englishes) that might be suggested as improving intelligibility in an ELF setting are: the occurrence of a full vowel rather than a reduced vowel like [ə] in the first syllable of a word such as computer or consider and in function words such as of and at; the use of a plural suffix on logically countable words such as furnitures and equipments; and prominent topic fronting often accompanied by a resumptive pronoun, as in my sister, she lives in Singapore. 1.2. Overview of the book In Chapter 2, I will consider the background to this study, including the concepts of intelligibility, misunderstanding and repair, and I will also provide a brief overview of Conversation Analysis. Chapter 3 describes the Corpus of Misunderstandings from the Asian Corpus of English (CMACE), including the speakers, the recordings, and the methods of identifying and analysing misunderstandings. In Chapter 4, the role of pronunciation is investigated, and then in Chapter 5, other features that sometimes cause misunderstandings are discussed, particularly grammar and lexis. In Chapter 5, I will additionally consider code-switching. Then in Chapter 6, I analyse how misunderstandings are dealt with, occasionally by the interactants asking for clarification though more usually by them adopting a ‘let it pass’ strategy (Firth 1996) in the hope that things will sort themselves out naturally. Chapter 7 deals with the pedagogical implications of my research and offers a few conclusions. Finally, a full listing of all the tokens of misunderstanding from the CMACE corpus is provided in the Appendix. Chapter 2 ELF, intelligibility and misunderstandings The patterns of usage found in English as a Lingua Franca (ELF) have been treated as a serious topic of investigation particularly since the publication of Seidlhofer (2001), and the concept of a limited set of pronunciation features for ensuring intelligibility in ELF interactions, the Lingua Franca Core (LFC), has been the focus of substantial discussion since the publication of Jenkins (2000). Nevertheless, the status of ELF and especially the proposed LFC continue to be controversial, and many writers have criticized them in derisive terms reminiscent of the “half-baked quackery” used by Quirk (1990) to dismiss the emergence of varieties of World English as models for teaching English around the world. This chapter reviews the status of ELF, in particular how it differs from World Englishes (Kachru 2005: 15). Then there is an overview of the LFC and a discussion of some of the hostility that has been targeted at it. Next I will consider the nature of intelligibility, including the distinction between intelligibility, comprehensibility and interpretability proposed by Smith (1992), and I will offer a brief overview of some research on the relative intelligibility of different varieties of English around the world. Closely linked to intelligibility is the concept of misunderstandings, and their nature will be discussed before I consider repairs and the typical ways that misunderstandings are dealt with and avoided. One way of avoiding misunderstandings is by means of accommodation, the adaptation of one’s listening or speaking to the needs of one’s conversational partners, so some basic concepts in accommodation will be discussed. Finally, I will offer a brief overview of Conversation Analysis (CA), the discipline that often provides the basis for research on misunderstandings and repairs, and I will consider the ways in which research into ELF interactions may differ from analysis of the patterns of native-speaker conversations that are usually investigated in CA. 2.1. ELF and World Englishes The term English as a Lingua Franca (ELF) has been in use at least since the mid-1990s (Jenkins 1996). However, it has only been widely adopted as a formal term to describe English as it occurs in international settings since 2.1. ELF and World Englishes 5 2000, partly as a result of Barbara Seidlhofer’s (2001) paper. One may note that although one of the most prominent figures in this field of study, Jennifer Jenkins, discussed the term ELF in her 2000 book (Jenkins 2000: 11), she chose at the time to use the alternative term English as an International Language (EIL) and only adopted the term ELF more widely in later publications (e.g. Jenkins 2005). However, research on ELF is now firmly established, with its own journal (Journal of English as a Lingua Franca), a dedicated series of conferences (the fifth ELF conference was held in Istanbul in May 2012), and a burgeoning array of monographs (e.g. Jenkins 2007; Smit 2010; Kirkpatrick 2010; Seidlhofer 2011; Cogo and Dewey 2012) as well as edited volumes (e.g. Mauranen and Ranta 2009; Archibald, Cogo, and Jenkins 2011). One problem with the term ELF is that the concept of a lingua franca often has negative connotations (Seidlhofer 2011: 80), referring to an impoverished code that has limited domains of use and is merely adopted as a last-resort means of communication between speakers with no other means of talking to each other. In fact, recent work on ELF shows that it can be immensely sophisticated, characterized by highly resourceful patterns of interaction. It is therefore inappropriate to regard it as an impoverished code, and furthermore it is of considerable interest to investigate the innovative ways that ELF speakers ensure that they can interact successfully. In this respect, I will now briefly discuss one feature of language usage: the occurrence of idioms. While one might expect ELF speakers to avoid the most obscure English idioms as well as colourful but opaque sayings transferred from their respective first languages, one actually finds that they often incorporate quite a lot of idioms into their English and they even sometimes develop fresh ones during their interactions. In fact, it has been suggested that ELF speakers may be particularly adept at this innovative extension of language usage. For example, Seidlhofer and Widdowson (2009) show how participants at an international conference created and then adopted the idiom endangered fields (by deliberate analogy with endangered species) to refer to areas of study that were perceived to be under threat. So, far from being an impoverished code, we generally find that ELF is rather creative, partly because it adopts expressions and patterns of usage from a wide range of different backgrounds, and also because its interactants frequently do not feel constrained by native-speaker normative rules. Indeed, it has been suggested (e.g. Cogo and Dewey 2012) that ELF interactants tend to pay little attention to standard grammar as they focus much more on the most effective ways of getting their message across, and the patterns of English that they adopt can be highly sophisticated. One of 6 2. ELF, misunderstandings and intelligibility the goals of the research in this book is to investigate the extent to which the use of non-standard grammar as well as idioms might result in misunderstandings among ELF interactants from different backgrounds. It is important to distinguish the study of ELF from that of World Englishes, as the two reflect quite different perspectives. The World Englishes paradigm investigates new varieties of English as independent, named, regional varieties, such as Singapore English, Indian English and Nigerian English, and it generally focuses on features of pronunciation, lexis, grammar and discourse that make each variety distinct from the others (Kirkpatrick 2007; Jenkins 2009). One of the key aspects of studies into World Englishes is how they emerge with their own independent identity, with endonormative standards of pronunciation, lexis and grammar that are not linked to the standards found in traditional Englishes from places such as the UK and the USA (Schneider 2007). In contrast, research into ELF typically considers how people from different countries interact. In other words, while studies of World Englishes generally focus on the distinct features of national varieties of English, research on ELF usually considers common patterns that are shared by speakers from disparate backgrounds. Nevertheless, it is fundamentally incorrect to suggest that research on ELF is proposing the emergence of a single variety of English. Indeed, even though it seeks to investigate some of the shared patterns by which people from different backgrounds communicate, it always acknowledges and indeed celebrates the fact that there continues to be wide variation in the ways that English is used around the world. A model that has been highly influential in the study of World Englishes is that of the Three Circles of English (Kachru 1985, 2005), in which the traditional varieties of English such as those of the UK, the USA and Australia are classified as being in the Inner Circle, postcolonial varieties such as those of Singapore, India and Nigeria are in the Outer Circle, and varieties in places with no colonial background and where English is taught as a foreign language such as Japan, Germany and Brazil are in the Expanding Circle. While there are some issues with this model because it is geographically and historically based (Jenkins 2009: 20–21) and it fails to reflect some of the dynamic ways that English is nowadays being used around the world (Cogo and Dewey 2012: 9), it still offers a constructive way of conceptualizing some of the different kinds of English that exist. In this book, I will make reference to the Three Circles, specifying, for example, that all the speakers in the recordings on which my research is based are from the Outer and Expanding Circles. 2.1. ELF and World Englishes 7 Despite its wide acceptance nowadays, the study of ELF still encounters substantial opposition, particularly from those who believe that we should continue to focus on native-speaker norms and patterns of English usage. Such opposition, even entrenched hostility, is especially targeted against the LFC, so I will discuss that in the next section. 2.2. The Lingua Franca Core (LFC) The Lingua Franca Core (LFC) represents a finite set of pronunciation features which, it is claimed, are necessary for achieving international intelligibility in spoken English. It is further suggested that non-core features which occur in native-speaker pronunciation are not needed for maintaining intelligibility, and indeed some features of pronunciation (such as the use of reduced vowels in the weak forms of function words) might actually interfere with international intelligibility. As a result, it is proposed that it is not necessary to teach non-core features, though some learners may decide that they want to approximate to native-speaker patterns of speech, and this choice should of course be respected. The features of the LFC were set out in Jenkins (2000), and were then summarized in various subsequent papers and books (e.g. Jenkins 2005: 201, 2006: 37, 2007: 23–24). The following are considered core features:       all the consonants of native-speaker English except [θ] and [ð] aspiration on initial voiceless plosives initial and medial consonant clusters vowel length distinctions the quality of the NURSE vowel the placement of the intonational nucleus In contrast, the following features are outside the LFC, so there should be flexibility in how they are realized:         the TH sounds final consonant clusters dark-L vowel quality (except for the NURSE vowel) vowel reduction (especially in the weak forms of function words) rhythm word stress the shape of the intonational contour (rising, falling, etc.) 8 2. ELF, misunderstandings and intelligibility While many people would accept that some features of pronunciation are more important than others, and furthermore they would agree that there is no need for learners always to mimic speakers from the UK or USA too closely, the details of what should be included and what might be excluded from the LFC remain contentious. For example, many teachers assert that word stress is vitally important; and furthermore excluding the quality of vowels from the LFC would be alarming to many teachers and learners. In contrast, not all teachers would agree that vowel length distinctions and aspiration on initial voiceless plosives are essential for a speaker to be intelligible in international settings. It is one of the main goals of the current study to provide further data that allows us to extend the evaluation about which features of pronunciation should be included in the LFC and which might be excluded. The LFC proposals were derived on the basis of just 40 tokens of misunderstanding between speakers from Japan and Switzerland who were engaged in comparing different versions of a map with each other (Jenkins 2000: 85). Only 27 of these tokens involved phonology, so there is clearly a need to extend the research, to analyse more instances of misunderstanding from a wide range of speakers in different conversational settings, and a few other studies have done that. Matsumoto (2011) basically concurs with the findings of Jenkins, though the suggestion that the final [t] in present is a core feature (p. 102) is a little surprising given that it is part of a final consonant cluster which is usually considered non-core. Osimk (2011) shows that the voiceless TH sound at the start of things is understood better when it is pronounced as [t] than when it is [θ] for listeners from Spain, Italy, France and Germany, though the performance when it is pronounced as [s] is not so good; and she also confirms that initial voiceless plosives are recognized best with substantial aspiration. However, McCrocklin (2012), challenges the LFC proposals when she asserts that word stress is important for intelligibility, based mainly on evidence from studies of listening by native speakers, though she provides no new data from ELF settings to support this claim. Unfortunately, there have not been many other attempts to replicate Jenkins’ original findings (Jenkins, Cogo, and Dewey 2011: 288). Varying attitudes have been reported among international students towards different features of the LFC. Fowler (2010) reports that, while the majority of international students believe that use of [θ] and [ð] for the TH sounds is important, most of them do not feel that use of weak forms is helpful. There is some logic to this: pronunciation of voiceless TH as [t] instead of [θ] can be regarded as a loss of information, as it results in tin 2.2. The Lingua Franca Core (LFC) 9 and thin not being distinguished; but the avoidance of weak forms retains useful distinguishing information, so for example have and of will always be differentiated (while they may both be said as [əv] by speakers in the Inner Circle). In fact, the attitudes of ELF speakers towards the TH sounds may depend on how they are realized. Shaw, Caudery, and Petersen (2009: 195) found that, over a period of several months, exchange students in universities in Sweden and Denmark stopped using [s] and [z] for the TH sounds, but the use of [t] and [d] remained stable. This finding supports the suggestion of Osimk (2011) that the latter pronunciation is more acceptable in ELF settings. Although it seems that some international students may welcome the LFC proposals, many English language teachers have less positive attitudes. For example, Scheuer (2005) asserts that LFC-based teaching is harmful for students, and Sobkowiak (2005) insists that it is not suitable for adoption in Poland. Jenkins (2005) contends that this opposition arises because of misconceptions about what the LFC represents, and Kirkpatrick (2007) argues that ELF-based teaching should be empowering and liberating for nonnative English teachers. But perhaps we should not be too surprised if teachers are reluctant to adopt quite radical proposals that seem to challenge so much that they believe in. There is, therefore, a need to consider in depth the extent to which LFCbased teaching might interfere with intelligibility, or whether conversely it might actually enhance the ease with which speakers can make themselves understood in international settings. 2.3. Intelligibility Smith (1992) makes a helpful distinction between three different kinds of understanding:    intelligibility: recognition of words and utterances comprehensibility: understanding the meaning of words and utterances interpretability: understanding the meaning behind words and utterances In other words, intelligibility refers to our ability to identify the words in an utterance, comprehensibility is about whether we know what the utterance means, and interpretability is concerned with the pragmatic implication of an utterance. 10 2. ELF, misunderstandings and intelligibility While these three concepts usefully remind us that intelligibility is a multifaceted concept, so it is not just about correctly identifying words, in fact applying the terms when dealing with real data can be problematic. For example, Nelson (2011) discusses the three-way distinction in some detail, and he suggests (p. 63) that the following quote from a novel by Chinua Achebe raises issues for interpretability: I want one of my sons to join these people and be my eyes (meaning ‘I am sending you as my representative’). But is it really true that this would be hard for people from other cultures to understand? Although some people around the world might find it an odd way of saying something, there does not seem to be much difficulty in comprehending what it means. In contrast, Nelson claims (2011: 108) that when his Australian-born sister-in-law noticed some yoghurt was spoiled and said that it was off, this was not an issue for intelligibility, but rather a “lively and interesting” way of expressing herself. But on what basis can we determine that saying some food is off is understandable to people from other backgrounds while asking someone to join these people and be my eyes causes problems for interpretability? This illustrates that identifying the interpretability of utterances is tough, as it is difficult to know the extent to which people really understand the implications of everything that others are saying. Pickering (2006) similarly notes that the concept of interpretability is hard to measure. For this reason, in my research on misunderstandings, I will be focusing mostly on intelligibility at the word and phrase level: I classify something as an instance of misunderstanding if there are some key words that the listener cannot identify or does not understand, even if at the wider level they may be able to follow the gist of the utterance quite successfully. And the two principle questions I investigate are: which phonological, lexical and grammatical factors have an influence on intelligibility? And how are misunderstandings dealt with and avoided? Munro, Derwing, and Morton (2006) also make a useful but different three-way distinction between intelligibility, comprehensibility, and accentedness. Intelligibility involves the recognition of words and sentences, so this is similar to the way the term is used by Smith (1992). However, the other two terms are different: comprehensibility is concerned with the ease with which listeners understand an utterance; and accentedness is the degree to which the pronunciation of an utterance deviates from a norm. The research of Munro et al. is based in Canada, and their classification only really makes sense in an Inner-Circle setting where a norm is reasonably well specified, as it is not clear what accentedness would mean in many Outer-Circle contexts, a point that Munro (2008: 193) acknowledges when 2.3. Intelligibility 11 he notes that the distinction between a foreign and a native accent is blurred in the context of nativized varieties of English. For example, if one asked listeners in Singapore to estimate the accentedness of an utterance, some of them would judge RP British pronunciation to be more accented than Singapore speech, while others would make the opposite judgment. Therefore, while Munro et al. make some insightful observations about the multifaceted reactions to accented speech, for example showing that familiarity with a variety of English does not always enhance the intelligibility of utterances in that variety, I will not adopt their classification here. It is of course important to recognize that intelligibility is not just about whether something is understood or not, and Munro and his colleagues conduct valuable research into the ease with which listeners understand various kinds of speech. However, assessing the comprehensibility of conversational data remains a problem. While it is not too difficult to determine whether listeners understand an utterance, for example by asking them to transcribe the words that they hear, it is rather harder to measure the ease with which interactants in a conversation understand the words, and any such evaluation is inevitably rather subjective. Some interesting work in this respect has been done by Björkman (2009), who uses questionnaires to determine how irritating certain non-standard features of speech are judged to be by ELF listeners in a Swedish university, and she finds (p. 242) that disrupted word order is the feature which is most often reported to be irritating, while tense usage and non-marking of plural nouns are among the features that are judged to be the least irritating. Quite apart from the classification of the different aspects of intelligibility, there seems to be widespread agreement that English spoken by people from the Inner Circle is not necessarily more intelligible than that produced by people in the Outer and Expanding Circles. For example, Smith and Rafiqzad (1979) report that the speech of someone from the USA was found to be less intelligible than that of someone from Malaysia, and Smith and Bisazza (1982) found the same when comparing an American speaker with someone from India. Furthermore, there is plenty of anecdotal evidence that many ELF speakers have problems understanding people from the Inner Circle. House (2003: 567) reports that few misunderstandings emerge in ELF discourse in contrast with the many misunderstandings she found in her native–non-native data. And Shaw, Caudery, and Petersen (2009: 192) quote exchange students at universities in Sweden and Denmark who observed that although Americans and Australians were friendly, they were hard to understand, partly because they did not pronounce all the sounds. In fact, there is widespread agreement that ELF speakers are able to 12 2. ELF, misunderstandings and intelligibility make their speech more intelligible if they do not try to approximate nativespeaker norms, often because they are good at adopting suitable accommodation strategies such as repetition and paraphrasing (Cogo 2009: 257). One might expect that familiarity with a variety of English should improve the degree to which it is intelligible. However, in a review of research on intelligibility in ELF, Pickering (2006) concludes that this is not necessarily true. She also notes that there is a wide range of factors that influence intelligibility, including the attitude of the listener, familiarity with the speaker or the topic, and level of tiredness. Something that might be discussed in connection with intelligibility is the concept of fluency. Most speakers of English hope that their speech is judged to be fluent, and achieving fluency seems to be one of the central goals of language teaching. But what do we actually mean by fluency? Is it connected with rate of speaking? Or with linking words together? Or with avoiding pauses? In fact, Hüttner (2009) observes that the concept of fluency becomes even more problematic in an ELF setting. And one might observe that all three features just mentioned may under some circumstances interfere with intelligibility. In fact, speaking more slowly, avoiding too much linking between words, and using appropriate pausing would seem to be quite advantageous in making oneself easily understood. And, as will become apparent, a fast speaking rate is one of the factors that contributes most often to misunderstandings occurring. One other aspect of fluency suggested by Prodromou (2008) is the easy use of idiomatic fixed chunks. However, although use of these established phrases certainly facilitates the production of speech, it can also lead to misunderstandings if listeners do not know the idioms. I will discuss the occurrence of unfamiliar idioms in Chapter 5. 2.4. Misunderstandings Misunderstandings of course occur in all kinds of communication, nativespeaker as well as ELF. The question arises, therefore, whether misunderstandings are more frequent in ELF settings or not. Conventional wisdom suggests that they are. Yet research indicates that this common-sense assumption is not borne out in reality, as ELF speakers tend to be particularly adept at avoiding misunderstandings, and ELF discourse is actually usually rather successful (Mauranen 2006; Kaur 2010: 205). It is, nevertheless, of considerable interest to investigate what kinds of misunderstandings occur, what causes them, and how they are dealt with. 2.4. Misunderstandings 13 Kaur (2009, 2010) makes a useful distinction between ‘misunderstandings’ and ‘non-understandings’: with misunderstandings, the listener thinks they know what is said but gets it wrong; but with non-understandings, the listener does not know what is said. However, while this is an insightful distinction in theory, in practice it is not always easy to apply. For example, someone often has an idea about what is said but is not too sure, and it is difficult to categorize instances such as these. Similarly, there are tokens in which listeners may be able to make a correct guess about the identity of the words but they are still not certain and therefore ask for clarification. I will include such instances in my data and analyse how they are dealt with. But should they be classified as misunderstandings or non-understandings? Kaur (2010: 194) also notes that there are various degrees of understanding. Clearly, instances in which a listener is able to guess a word but seeks to confirm it are at a different position on the continuum from instances in which the listener is unable to make any guess. Yet objectively placing each instance somewhere along this continuum is not straightforward. In addition to considering a possible distinction between misunderstandings and non-understandings, and also the degrees of understanding, one might consider the level of analysis. Somewhat similar to the distinction made by Smith (1992) between intelligibility, comprehensibility and interpretability that I discussed in the previous section, misunderstandings can be analysed at the word level, the utterance level, or the overall pragmatic level. In fact, as will be seen in Chapter 3, in this book a systematic distinction will be made between word-level misunderstandings (which are here termed ‘tokens’ of misunderstanding) and utterance-level misunderstandings (which I will refer to as ‘instances’ of misunderstanding). In the sections on pronunciation and lexis, I will mostly discuss tokens of misunderstanding; but in the sections on grammar and repairs, I will deal with instances of misunderstanding. While discussion of the classification of misunderstandings is important, we must also consider what constitutes an instance of misunderstanding. Most research on ELF interactions analyses only those instances in which there is some obvious breakdown in the interaction, for example when one participant asks for clarification or else when someone clearly gets the wrong meaning and so answers inappropriately, though Kaur (2009) also looks at cases where a pause may signal a breakdown in communication. However, in addition I will include instances in which there is no apparent breakdown so the conversation seems to progress smoothly, but one of the interactants subsequently reported that there were some things they had not 14 2. ELF, misunderstandings, and intelligibility understood. The methodology for identifying such instances will be presented in Chapter 3. While the focus of this book is on misunderstandings, it is important to keep in mind the fact that ELF interactions are usually rather successful, so in addition to analysing what causes misunderstandings to happen, I will consider features of speech that reduce their occurrence, including for example the use of full vowels rather than reduced vowels such as [ə]. The wider issue of how misunderstandings are avoided can be considered under the heading of repairs, which I will now discuss. 2.5. Repairs Repairs to misunderstandings can firstly be analysed as self-initiated or other-initiated (Kitzinger 2013). In addition, they can be self-repairs or other-repairs (Have 1999: 116; Liddicoat 2011: 210). Typical instances of each of these categories are:     self-initiated self-repairs: a speaker realizes something may not be clear so provides a paraphrase or explanation without being prompted. self-initiated other-repairs: the speaker is unable to think of a word and asks for help, and the other person then offers a suggestion. other-initiated self-repairs: a listener does not understand something so asks for clarification, and the original speaker explains what they meant. This might be regarded as the archetypical kind of repair, but it is actually quite rare in ELF interactions, though Smit (2010) suggests that it may be more common when the interactants have got to know each other well so they feel that they can be quite direct in asking for clarifications. other-initiated other-repairs: the listener does not understand something and then provides the answer. This category includes instances where the listener corrects the original speaker, sometimes as a checking strategy to ensure that they got it right. Instances of correction are rare in ELF interactions (though it is possible that they are more common when a native speaker is involved), but they do sometimes occur, and I will discuss a few such instances. Liddicoat (2011: 248) suggests that there is a strong preference for selfrepair, which is hardly surprising, as asking for help (self-initiated otherrepair) and correcting (other-initiated other-repair) are not common strategies. But, in fact, one common kind of repair that has been suggested in 2.5. Repairs 15 ELF interactions is when the listener does not actually indicate that there is a problem and instead adopts the ‘let it pass’ strategy (Firth 1996) in the hope that things will sort themselves out. It is hard to classify these repairs under the four categories listed above – maybe these repairs could be described as zero-initiated self-repairs. Or maybe they could be classified as natural repairs, as one can say that the problem naturally fixes itself. Firth suggests that the ‘let it pass’ strategy is particularly common in ELF interactions, especially the telephone conversations in an international company that he studied, though it remains uncertain whether the strategy is really more common in ELF settings than among native speakers. Perhaps it is actually the normal way we all behave when faced with something we do not understand. Indeed, Mortensen (2013: 35) concludes that the ‘let it pass’ strategy is “a basic interpretive procedure that human beings follow in order to make sense of the world”. Misunderstandings involving silence are sometimes overlooked in research on misunderstandings in ELF because they generally do not become apparent from the transcript. Furthermore, we might also say that a few misunderstood words often do not have too much impact on the successful continuation of ELF interactions, as it is normal for things to sort themselves out naturally, and also we do not need to understand every single word in order to follow the gist of what is being said. Ehrenreich (2009: 144) quotes someone working in a German multinational company who said that he knew what a Chinese colleague wanted, but he could not tell what the individual words were. And maybe this is a common pattern. Nevertheless, in the current study, I will be looking at individual words that are not understood, as this gives important information about what enhances and what interferes with intelligibility. Furthermore, as I will show, sometimes the ‘let it pass’ strategy does not work, and it subsequently becomes apparent to the listener that they have to sort out the identity or meaning of misunderstood words. Indeed, this is illustrated in the study by Firth (1996: 244) where one of the interactants realized that he had to find out what blowing cheese meant even though he had let it pass the first time it was said. Similar to the strategy of keeping quiet, minimal backchannels such as mm or yeah are common ways of adopting the ‘let it pass’ strategy. The primary role of such backchannels is to confirm that the listener is following what the speaker is saying, but they can also be used to conceal the fact that the listener does not actually understand something in the hope that it will all get sorted out in time. As we will see, such use of backchannels is common in ELF discourse, though once again it is not certain if the patterns 16 2. ELF, misunderstandings, and intelligibility of usage are special to ELF interactions or are broadly the same as in native-speaker conversations. Finally, Kitzinger (2013: 255) notes that repairs “can be used as a way of managing interpersonal conflicts and difficulties”, so they are not just about dealing with problems of understanding. Indeed, under the rubric of repairs, we can consider some strategies which are adopted to ensure that there is no breakdown in communication. For example, House (2003: 568) suggests that a kind of response which she calls ‘re-present’ is common among Asian speakers. This involves echoing part of the previous speaker’s words, partly to encapsulate the information, but also to support working memory and to create a linkage between turns. Whether this kind of echoing of words is more common in Asian discourse is hard to evaluate; but I will offer a few examples of it occurring in my data, and I will suggest that one of the reasons it is sometimes adopted is to conceal the fact that the listener does not understand everything. Many of these kinds of conversational strategies involve a high degree of cooperation between interactants, so they might be described as forms of accommodation. I will now briefly consider that. 2.6. Accommodation There is sometimes a perception that the proposals for ELF-based teaching are based on negative things: students do not need to aspire to nativespeaker norms in pronunciation or grammar, and according to the LFC suggestions, it is not necessary to learn to use some of the inventory of English sounds. However, this overlooks an extremely important component of the proposals: the need for speakers to accommodate to their listeners. The importance of accommodation in ELF discourse has long been emphasized (Jenkins 2000: 167–175) and has subsequently been investigated in further depth (Jenkins 2007: 25). Accommodation usually refers to adapting one’s patterns of speech to make it more similar to that of one’s interlocutor, though of course divergent behaviour sometimes also occurs, such as when someone wants to emphasize that they are different from their conversational partners, for example in order to exert power over them. Much of accommodation is connected with modifying one’s pronunciation, including such things as speaking rate and pausing frequency (Giles, Coupland, and Coupland 1991: 7). However, there are different kinds of accommodation, affecting more than just pronunciation. First, there is listening accommodation, which 2.6. Accommodation 17 means getting used to the patterns of speech of one’s conversational partners; and furthermore, there is accommodation in terms of grammar and word usage, so for example, Shaw, Caudery, and Petersen (2009: 193) report that, over a period of several months, exchange students in Sweden and Denmark sharply reduced the frequency of using the phrase I think, while at the same time well and like became used substantially more often, and they argue that this was a sign of them accommodating their speech patterns to those of the other speakers around them, suggesting a shared trend towards the use of a wide range of discourse management devices. In Chapter 4, I will consider phonetic accommodation, both listening and speaking, in Chapter 5, I will discuss code-switching, a form of accommodation in which conversational interactants seek to use all the linguistic resources that they share, and in Chapter 6, I will analyse pragmatic accommodation that is found in ELF data as speakers adopt a range of discourse strategies to try to ensure that they are not misunderstood. 2.7. Conversation Analysis Conversation Analysis (CA) originated in the work of Harvey Sacks on the description of interactional behaviour in audio recordings from a suicide hotline in Los Angeles in the 1960s, and it has been extended to provide an extensive methodological framework for the analysis of all kinds of conversational interactions (Have 1999: 3). Key aspects of the methodology associated with CA which I will adhere to in this book are:     the descriptive, non-judgmental reporting of conversational interactions, which avoids reference to preconceived notions of ‘correct’ usage the use of natural conversations rather than specially designed experimental materials or read data the attempt to find out not just what people are saying (the form) but why they are saying it (the function) the inclusion within the transcription of lots of detail, including information about overlaps, pauses, changes in the tone of voice, and anything else that may help to explain what is going on In one other respect, the current work might be seen as following the principles of CA. In his original work on telephone conversations, Harvey Sacks found just one exception to the conventional behaviour where the receiver speaks first, but instead of dismissing this instance as deviant, he focused on it, to try to determine why this unexpected behaviour occurred. 18 2. ELF, misunderstandings, and intelligibility We can say, therefore, that one crucial aspect of work on CA is that deviant cases should be taken seriously (Have 1999: 39). And similarly, in the current work, misunderstandings are very much the exception rather than the norm, as most ELF interactions progress remarkably smoothly without too many misunderstandings. So, in this sense, by focusing on misunderstandings, I am conducting a detailed analysis of the abnormal cases. However, the analysis in this book is distinct from much work in Conversation Analysis in two respects. First, CA tends not to refer to accounts by the participants themselves, on the basis that they often do not know what they are doing or why they are doing it (Have 1999: 33). Indeed, ‘pure CA’ treats the conversational recordings and transcripts as pure data which should be interpreted in their own right, unencumbered by distracting information about the backgrounds or the settings (Have 1999: 55). However, in this work I make extensive use of feedback from participants in two crucial ways: I get them to tell me what they said and also what they did not understand. Although it is certainly true that speakers often do not know why they say something, they usually do know what words they are trying to say. And this feedback from the speakers is vital, as I will show in Section 3.1.3, as there are many instances when I was simply unable to decipher the words without the input of the speakers. In addition, and equally importantly, I depend substantially on feedback from the participants to determine what they do not understand. This feedback must be treated with caution, as sometimes participants may subsequently claim they did not understand something when in fact they did understand it, possibly because they had no problems in context but later on were no longer able to decipher what was said when they listened to the recording. Nevertheless, as we will see, this feedback provides insightful information about intelligibility, and I believe that the overwhelming majority of the misunderstandings that are reported are genuine. Second, most work in CA is qualitative, with detailed analysis of transcripts but little attempt at quantification, and while adverbs such as commonly or typically may occur, numerical quantification is often avoided (Liddicoat 2011: 11). In fact, the usual practice is to find one or two key exemplars of a conversational exchange, examine them in considerable detail, and then investigate whether they are characteristic of interactions in general (Have 1999: 144). In contrast, in my investigation, I will try to quantify some aspects of the data, in particular what features of pronunciation are most likely to give rise to misunderstandings and which have little impact on intelligibility, and also the frequency of use of various repair strategies. The work of Smit (2010) is similarly grounded in CA but at the 2.7. Conversation Analysis 19 same time attempts to quantify the relative frequency of occurrence of the patterns that are found, and my research takes that as a model to follow. In many ways, therefore, the current study can be described as a mixedmethods study, combining qualitative and quantitative research, something that Dörnyei (2007) suggests is rather common in linguistic investigations. Let me finally discuss one further way in which analysis of ELF interactions may not fit into CA as it was originally conceived. One fundamental conjecture of CA is that all conversations are highly structured and orderly, and it is the goal of the analyst to work out this structure. Liddicoat (2011: 7) observes that it is assumed “participants share the same procedures for designing and interpreting talk”. However, Firth (1996: 252) suggests that this may not apply to ELF interactions, as the interactants do not necessarily share the same knowledge about how to participate in conversations, and as a result the structure of their interactions may be more flexible than that of native-speaker conversations. One of the goals of my work is to investigate some of the ways that the participants in ELF settings negotiate their interactions in order to achieve mutual understanding, avoid miscommunication, and deal with those misunderstandings that do occur. 2.8. Conclusion Research on ELF interactions is becoming an increasingly important and vibrant field of study, offering substantial insights into the ways that people around the world are communicating in English. ELF interactants are generally focused on maintaining intelligibility, but it seems that they are sometimes not too concerned about adhering to native-speaker norms, so it is vitally important to determine what features of speech enhance intelligibility and which are more likely to cause misunderstandings to occur. Analysis of this is the main focus of this book. It is also instructive to investigate how repairs are achieved. We need to be aware that many of the patterns involved may differ from the ways that repairs are enacted in native-speaker discourse, so some of the complex turn-taking routines reported in Conversational Analysis may not be applicable in ELF interactions. I will now discuss the data that I collected and how it is analysed in this book. Chapter 3 Data and methodology One way of conducting research on misunderstandings is by analysing a corpus of spoken data. In this chapter, the Corpus of Misunderstandings from the Asian Corpus of English (CMACE) is described, and then the methodology for the analysis of the data is discussed. 3.1. CMACE Corpora of naturally-occurring English from around the world are now widely used in the analysis and description of speech. One of the most influential of these is the International Corpus of English (ICE), which provides a carefully balanced sample of spoken and written material in a range of different Englishes from places such as Britain, Singapore, India, Hong Kong, and the Philippines, and some substantial analyses have emerged (e.g. Nelson, Wallis, and Aarts 2002). The ICE corpus is restricted to Englishes from the Inner and Outer Circles, so it excludes speakers of English in the Expanding Circle. However, Seidlhofer (2011) has argued that there is no reason to continue to exclude speakers in the Expanding Circle from corpora of English usage. They now outnumber all other speakers of English, so it is important to conduct research into the ways that they interact. Indeed, the VOICE corpus that Barbara Seidlhofer and her colleagues have assembled in Vienna quite deliberately includes lots of speakers from the Expanding Circle (VOICE 2011). The Asian Corpus of English (ACE) is inspired by the VOICE corpus, and it adopts much of the same methodology as well as most of the transcription conventions. The ACE corpus is headed by Andy Kirkpatrick and is led by a team at the Hong Kong Institute of Education, and it consists of conversational data collected and transcribed by a number of teams, including those in Brunei, Hong Kong, Malaysia, the Philippines and Singapore. The Brunei component of the ACE corpus has been collected at the University of Brunei Darussalam (UBD) by a team led by Salbrina Sharbawi. It consists of 14 recordings, each lasting between 20 minutes and one hour, totalling about ten and a half hours. Five of these recordings involve three or four people having a discussion in a classroom at UBD, while the 3.1. CMACE 21 other nine are conversations between two people, with one interviewing the other, in a lecturer’s office at UBD. All the recordings were made using a Handy H4n recorder, to ensure a high quality of recording for the speech of all the participants, and they are saved in WAV format. Nine of these ACE recordings made in Brunei, totalling about six and a half hours, have been investigated for this book in order to find instances where one speaker fails to understand something that is said by another speaker. These nine were selected because the people involved were able to come back subsequently and help identify where misunderstandings occur and also to clarify what is actually said. One of the recordings involves a discussion between four people, while the other eight are conversations involving two participants, the interviewer and the interviewee. The speakers, further details about the recording sessions, and the methodology adopted in identifying the instances of misunderstanding are described below. In this book, I will refer to these nine recordings as the ACE data; and I will refer to the corpus of misunderstandings derived from the ACE data as CMACE. 3.1.1. Speakers Five female and four male participants are involved in the current study. They are identified by their gender (F or M) followed by a two-letter code indicating their country of origin. All of them speak English fluently and proficiently, but some of them have a distinctive regional accent that can sometimes cause problems in intelligibility. They are listed in Table 3.1. Table 3.1. Participants in the recordings Code FBr FCh FJp FMa FTw MHk MIn MLs MNg Speaker Female from Brunei Female from China Female from Japan Female from Malaysia Female from Taiwan Male from Hong Kong Male from Indonesia Male from Laos Male from Nigeria Age 23 22 22 24 56 56 58 34 28 Background research student exchange student exchange student research student housewife university academic university academic government officer undergraduate student One consideration in making the recordings was whether native speakers might be included. Some writers have observed that the dynamics of 22 3. Data and methodology interactions can change when native speakers are involved. For example, one of the participants in the study by Matsumoto (2011: 109) observed that he hesitates in the presence of native speakers and only feels at ease when he is talking to non-native speakers. We might note that Jenkins (2000) excluded native speakers from the data which provided the original basis for the LFC, as her research mainly focused on interactions between learners of English from Japan and Switzerland. However, native speakers are not completely excluded from other ELF corpora, such as the ELFA corpus being compiled in Helsinki (Mauranen 2006: 129). Indeed, in more recent work, Jenkins has acknowledged that native speakers may be included in ELF data (e.g. Jenkins, Cogo, and Dewey 2011: 283). One issue, as noted in the introduction to this book, is that the concept of ‘native speaker’ is rather hard to define (McKay 2002: 28; Davies 2003). In the current study, two of the participants, FBr and FMa, state that English is their best language, even though both have at least one other language that she uses regularly at home. So, are they native speakers or not? They would certainly be classified as coming from the Outer Circle, as they are from post-colonial countries in South-East Asia. But if their first and best language is English, surely they should be classified as native speakers. They are included in the study, because the conversations they are involved in yield lots of highly informative instances of misunderstandings. The data that constitute my corpus are rather different from some other ELF data in one important respect: all of the speakers were complete strangers, as none of them had ever met any of their conversational partners before the recordings. Much other work on ELF interactions focuses on communities of practice, people who form a group because they work or study together or else have shared interests, and these groups can develop innovative idioms and other shared patterns of language usage over a period of time. For example: Ehrenreich (2009) reports on the English usage in multinational companies based in Germany where some special terminology and modes of interaction have become established among the participants; and Smit (2010) conducts a detailed investigation into the English used by students from a range of countries involved in a two-year hotel management course at a school in Vienna. However, in the current study, all the participants were together for less than one hour, so we would expect to find far fewer examples of shared innovative patterns of language usage emerging in their conversations. Furthermore, as we will see, there are rather more instances of misunderstanding arising because of pronunciation in the data investigated here than reported by Smit (2010). And finally, some of the patterns of dealing with repairs are likely to be substan- 3.1. CMACE 23 tially different. Smit (2010) reports that the participants she studied could be quite direct in asking for clarification, as one might expect for people who know each other well. In contrast, in the data I am analysing here, the participants are likely to be less direct, and it is possible that the ‘let it pass’ strategy is adopted rather more frequently in my data. Of course, both kinds of data provide important insights into various aspects of ELF interaction, but my recordings provide data that is conducive for the study of misunderstandings. 3.1.2. Recordings The nine recordings that are analysed in depth are listed in Table 3.2. The identifying code for each recording consists of a two-letter code for the country of origin of each of the two participants, the first country listed being the country of the interviewee and the second being that of the interviewer. For the most part, therefore, I am investigating how well the second participant is able to understand the first, though in a few cases the misunderstanding goes the other way. For example: in one case in the Jp+Br recording, FJp gets the wrong meaning for find said by FBr; in an instance in the Hk+Ma recording, MHk is unable to understand the phrase fell into place said by FMa; and at one point in the In+Tw recording, MIn misunderstands the word Hash said by FTw and hears it as harsh. Table 3.2. Recordings Code Ch+Br Hk+Ma Hk+Tw In+Ma In+Tw Jp+Br Ls+Br Ng+Br Tw+Ma Participant 1 FCh MHk MHk MIn MIn FJp MLs MNg FTw Participant 2 FBr FMa FTw FMa FTw FBr FBr FBr FMa Total : Duration (min:sec) 51:03 23:13 47:14 36:29 42:03 52:28 45:07 58:17 29:01 6:24:55 The Ng+Br recording is a group discussion involving four participants. However, two of them, a male from Pakistan and a male from Brunei, say very little during the recording, so the analysis here focuses on the interaction between MNg and FBr, in particular what FBr is unable to understand in the speech of MNg. The Ng+Br recording is therefore treated as an inter- 24 3. Data and methodology action between two people, so it is similar in this respect to the other eight conversations. I will refer to extracts from these recordings by means of the identifying code of the recording together with the time in seconds from the start of the recording, so for example Ch+Br:357 refers to an extract which begins 357 seconds from the start of the conversation between FCh and FBr. 3.1.3. Transcription In the early days of the development of large-scale corpora of conversational speech, Sinclair (1985: 251) observed that there was a need to overhaul the systems for the description of English to deal with real conversational data; and Seidlhofer (2009: 38) suggests that the creation of largescale ELF corpora has necessitated a similar paradigm shift in our descriptive systems. However, this has already largely been achieved, particularly arising out of the work of Barbara Seidlhofer and her colleagues in Vienna in transcribing and analysing their VOICE corpus. I will therefore adopt the VOICE transcription conventions. They have been used in transcribing the ACE data, and they will be adopted in presenting extracts from the CMACE data in this book. They are summarized at the front of the book. The conventions used for the VOICE data are fairly consistent with those adopted in other work in Conversation Analysis (Have 1999: 87). One salient difference is the use of angled brackets <1> … </1>, <2> … </2> etc. to show overlaps instead of the traditional use of square brackets ‘[ … ]’. An advantage of the angled brackets convention is that it facilitates showing multiple overlaps that can occur in interactions with lots of participants, as details of the progression of the overlaps can sometimes get confusing with the simpler square bracket notation. In fact, in the current research, nearly all of the interactions involve just two participants; but it was decided to stick with the convention adopted in most work on ELF nowadays, to make the current work consistent with other research on ELF. One other decision was whether to use upper case to show all stressed syllables or whether it should only be used for unexpected stress (Have 1999: 88). Here, it was decided only to show the unexpected occurrence of stress, particularly when this may lead to misunderstandings (though, as will be seen in Chapter 4, misunderstandings arising from stress placement are in fact infrequent). See the Transcription Conventions on page x. One of the unusual aspects of the CMACE data is the extensive involvement of the participants in the analysis. House (1999: 76) observes that it is difficult to derive a watertight definition of a misunderstanding, so 3.1. CMACE 25 she suggests that it should depend on whatever the participants perceive as a misunderstanding. But if we do not obtain their detailed feedback, how can we find out what they perceive to be a misunderstanding? Therefore, I involved the participants in the transcription and analysis to provide a valuable insider (‘emic’) perspective onto the data (Smit 2010: 97). Seidlhofer (2011: 56) has noted that there is a tendency for people working with speech corpora to focus too much on form and pay too little attention to function, and there is a need to get behind the data, to find out what is really going on. I believe that the involvement of the participants in the analysis facilitates obtaining the kind of in-depth insights that she envisages. The methodology adopted here offers two crucial advantages: first, it has enabled me to correct the transcription in many places when what is said is not clear (though, inevitably, it has not been possible to resolve all such instances); and secondly, it has made it possible for me to identify instances of misunderstanding that are not actually signalled in the recordings, which actually represents the overwhelming majority of the tokens. In this way, I can find instances where one of the participants does not understand something even when there is in fact no apparent breakdown in communication. Let me illustrate this with an extract from the Ng+Br recording. Extract 3.1. Ng+Br : 2742 (Token 173) Context: MNg is talking about how it is easy for some people to get a job after graduating from university. 1 2 3 4 5 MNg FBr MNg … they are not looking for anything. that time they will be calling them because they are hot cake (1.3) they need their service. yeah yeah. not like. oh i want to no no no no no no no when they see this one you do in brunei today. if you have anything … In Extract 3.1, MNg says it is easy for some people to get a job because they are hot cake (line 2), meaning that they are so talented that employers automatically come looking for them – so it refers to people rather than things as is more common with the idiom hot cakes in British English. FBr does not understand this, partly because there is no [h] at the start of hot and furthermore she is not familiar with this use of the idiom, but there is no indication of her lack of understanding from listening to the recording. In fact, in line 3 FBr provides the backchannel yeah in response. I only found out that she does not understand it because she subsequently tran- 26 3. Data and methodology scribed it as out kick; and the only way I know that it is actually hot cake is because MNg came and explained it to me. The value of the substantial involvement of the participants in the analysis can be appreciated when comparing the CMACE corpus with other studies of misunderstandings. For example, Mauranen (2006: 132) only found six clear cases of misunderstanding in her five hours of data from Finnish universities; Kaur (2010) based her analysis of problems of understanding on a corpus of fifteen hours of speech, but her report includes a detailed consideration of just twelve extracts; Björkman (2009) analysed 70 hours of lectures and group work in English in a Swedish university and lists only four tokens in which paraphrasing or self-correction occurs because of overt disturbance in the discourse, and she observes (p. 237) that it is very hard, or maybe impossible, to assess comprehension of the lectures; and Matsumoto (2011) analysed three recordings totalling nearly one and a half hours, but only six instances of misunderstanding are discussed. While it is likely that rather more tokens of misunderstanding were found in most of these studies and just a few key instances were selected for detailed consideration in order to illustrate the issues that are under investigation, it also seems probable that in all cases there were rather a lot of instances of misunderstanding that were not identified. In contrast, in the six and a half hours of data analysed here, 147 instances of misunderstandings have been identified and analysed, and the overwhelming majority of these would not have become apparent without the invaluable feedback from the participants. Even with such extensive feedback from the participants, inevitably there remain some tokens which I cannot decipher. For example, consider Extract 3.2. FBr subsequently transcribed the final word in line 6 in brackets, to indicate that she could not determine what word it is. So what in fact is FCh trying to say? Extract 3.2. Ch+Br : 2083 Context: They are discussing movies that they like to watch. 1 2 3 4 5 6 7 8 9 FBr FCh FBr FCh FBr FCh FBr all right. i don’t particularly watch chinese movies? (.) cos i don’t know a lot? but (.) there’s ONE that erm i know it’s quite f- it’s really funny erm (.) kung fu hustle? have you heard of it? kung fu hustle? yeah. kung fu (.) the (chose) kung fu hustle erm: (1.6) maybe i know the (.) chinese name but not the english o:h okay 3.1. CMACE 27 It is loud and clear as [tʃoʊs], but I cannot determine what the word is. When I asked her about it, FCh herself suggested that she might have started to say Chinese, but despite careful listening, she was also unable to work it out. Instances like this are not included in the corpus of misunderstandings because, in the absence of knowing what the word is, I cannot draw any conclusions about what the problem might be. Fortunately, with the help of the participants, there are not too many instances like this. I am very grateful to the generous contribution of all these people in the project. 3.2. Analysis 147 instances of misunderstanding were identified and collated in the CMACE database, which is available on-line. This section describes how the analysis of the misunderstandings was conducted. 3.2.1. Instances and tokens Many instances of misunderstanding involve more than one token. Here, I will explain the difference I make between instances and tokens. Extract 3.3 shows an instance of misunderstanding from the Ng+Br recording. FBr understands most of what MNg is saying, but she does not get some of the details, including in line 2 the word portion (a Nigerian term for an instance of punishment) and in line 3 the words weed (a misunderstanding that may arise out of the length of the vowel) and cutlass (perhaps because in Brunei one would refer to a parang instead). Extract 3.3. Ng+Br : 693 (Tokens 151 to 153) Context: MNg is talking about the punishments given in schools in Nigeria. 1 2 3 4 5 6 FBr MNg FBr MNg what what what kind of punishment. <1> how far </1> <1> yeah the </1> punishment will be (.) giving you portion (.) go there (.) you have to go and weed the something with cutlass (.) or you have to kneel down inside the sun (.) for some hours under the hot sun YEAH that is it Therefore, although this extract represents one instance of misunderstanding, it includes three separate tokens. While the features that contribute to each of these tokens can be analysed separately, at the same time it is im- 28 3. Data and Methodology portant to consider them as a single instance, because sometimes there is substantial linkage between the tokens. For example, consider Extract 3.4. Extract 3.4. Hk+Tw : 999 (Tokens 50 & 51) Context: MHk is talking about practising what you are good at, such as swimming. 1 2 3 4 5 MHk FTw MHk … you can swim but you never swim you know you never enjoy swimming (1.0) you never enjoy you know the freedom you know floating mm on the water you know (.) all that stuff … In line 3, MHk says floating, but FTw does not understand this word, largely because the [l] is missing, and she subsequently left a question mark in her transcript. Then in line 5, she hears under instead of on the, partly because the consonant at the start of the is pronounced as [d]. But if she had understood floating, she would probably also have understood on the water. So these two tokens of misunderstanding, floating and on the, are linked. Of course, a single token may sometimes consist of more than one word, particularly when a fixed phrase is involved. In Extract 3.5, FCh has been talking about films she likes to watch, and even in this context about watching recent films (in a recording made in February 2011), FBr is unable to understand Black Swan until FCh adds that it stars Natalie Portman. Extract 3.5. Ch+Br : 2013 (Token 10) 1 2 3 4 5 FBr FCh FBr FCh FBr ... what about movies. you like to watch movies? yes: i just saw the (.) black swan. i liked it. rex one? yeah the natalie portman OH: RIGHT: Here, Black Swan consists of two words, but it is just one entity, so it is treated as a single token of misunderstanding. Inevitably, it is sometimes not clear whether a phrase should be analysed as consisting of one token or more than one token. For example, consider Extract 3.6. Subsequently, FMa reported that she initially heard want the wall (line 2) as won the war, principally because there is no [l] at the end of wall. Although want and war could be treated as separate tokens within this instance, here I regard them as a single token because they are part of the same short phrase. 3.2. Analysis 29 Extract 3.6. Hk+Ma : 710 (Token 19) Context: MHk is talking about painting murals in America. 1 2 3 MHk you all you know have this kind of background okay (0.8) then (.) i want the wall (.) this is i the wall i (.) everyday i will (.) go to work i will see i go to school i will see this wall … There is one other issue with regard to Extract 3.6. FMa misunderstands wall both times it occurs in line 2 (though she finally figures it out by the time it is said again in line 3). However, I do not treat the second wall as a separate token of misunderstanding as it occurs so soon after the first. In fact, if a word is misunderstood more than once within the same instance, it is classified as a single token of misunderstanding. In contrast, if the same word is misunderstood again somewhat later and it therefore belongs in a different instance of misunderstanding, then it is classified as a separate token. In Extract 3.7, FTw misunderstands the word plastic said by MHk, hearing it as past (largely because the [l] is missing, but also because there is no final [k]). Extract 3.7. Hk+Tw : 1405 (Token 56) Context: MHk is talking about wastage. 1 2 MHk … for example how much you know plastic you use. if you using shopping bag … Then about two and a half minutes later, FTw again misunderstands the word plastic spoken by MHk, as shown in Extract 3.8, hearing it as past once more. As these two tokens of plastic are separated by over two minutes, they are in different instances and so they are treated as separate tokens. Extract 3.8. Hk+Tw : 1564 (Token 63) Context: MHk is talking about how long it takes for garbage to decay. 1 2 MHk ... how long you know for the (.) nature you know to digest this plastic container … In the 147 instances of misunderstanding, there are 183 tokens, all of which are listed in the Appendix. The largest number of tokens in a single instance is four, as shown in Extract 3.9, in which FTw is unable to understand the words Tech, Lubbock, northern and west spoken by MHk, and she 30 3. Data and Methodology subsequently transcribed them as there, lot of, low term and where respectively. Extract 3.9. Hk+Tw : 274 (Tokens 31 to 34) Context: MHk is talking about when he got married. 1 2 3 4 MHk … in texas you know at that time (.) we are you know study in texas tech you know in the graduate school. in lubbock texas. er northern you know er (.) west you know part of the texas. so (.) that february’s (.) that day is snowing too (.) so … The first two of these tokens involve names, though FTw does not realize that they are names, while northern is pronounced as [lɒftən], and west has no final [t]. 3.2.2. Classification After all the tokens of misunderstanding were collated, I attempted to classify the factors that give rise to each one, principally as involving pronunciation, lexis, or grammar. Inevitably, it is not always possible to identify a single factor, as quite often there are multiple causes. Consider for example Extract 3.10. Extract 3.10. Hk+Ma : 1211 (Token 26) Context: MHk is discussing the problems of moving to a new place. 1 2 3 4 MHk FMa MHk … because every time when i relocate you know. either really (.) cold freezing cold mm or flaming hot (0.8) it’s used to be. it’s okay … FMa subsequently reported that she was unable to understand flaming hot, and she guessed that the first word was fuming. The immediate cause of this can be assumed to be phonetic, as MHk simplifies the initial [fl] consonant cluster in flaming by omitting the [l], something he does rather often (as we will see in Section 4.1.4). However, he similarly omits the [r] from freezing in line 2, and FMa has no problem understanding this word. How can this be explained? The answer is that the phrase freezing cold is rather common in English, occurring 197 times in the on-line Corpus of Contemporary American English (COCA 2013), while the phrase flaming hot is much less common, occurring just twice. So we should actually conclude 3.2. Analysis 31 that FMa’s inability to understand flaming hot is a combination of pronunciation and lexis, something which is quite common. Another instance from the same recording is shown in Extract 3.11. Extract 3.11. Hk+Ma : 154 (Token 14) Context: MHk is talking about why he decided to become an artist. 1 2 3 MHk … what subject you chose to (.) study. (0.7) for your <spel> m a </spel>? your you know f- (.) terminal degree? (.) base on (1.0) your interest In this case, FMa is unable to understand terminal, and she subsequently guessed that it could be common (though that does not make much sense in this context). This might be considered a problem with pronunciation: terminal is pronounced as [tɜːmu], with just two syllables. But that does not tell the whole story. The real problem here is that the phrase terminal degree is frequently used in the United States to refer to one’s final degree, so for example it occurs 40 times in the COCA data. However, it is not so widely used outside the United States. Indeed, I had to get MHk to come and explain it to me. The conclusion is this: if terminal degree had been a familiar phrase for FMa, such as freezing cold in Extract 3.10 above, then saying it fast with some phonetic reduction would not matter too much. But given that it is not such a common phrase outside the USA, phonetic reduction is more likely to cause a problem. So, once again, I conclude that this token involves a combination of pronunciation and lexis. One should note that the issues illustrated in Extracts 3.10 and 3.11 are subtly different. In Extract 3.10, MHk is using his own idiosyncratic phrase flaming hot which does not occur frequently elsewhere, while in Extract 3.11, he is using a phrase terminal degree which is common in the USA but not elsewhere. Nevertheless, both of these extracts illustrate that many tokens of misunderstanding involve a range of different factors, and it is not always possible to identify a single cause. Despite these difficulties, I have attempted to classify the tokens of misunderstanding as involving pronunciation, grammar or lexis. In cases where there appear to be multiple causes, I cross-classify the tokens, so Extracts 3.10 and 3.11 are classified as involving both pronunciation and lexis. The overall results of this classification of the 183 tokens of misunderstanding are shown in Table 3.3, in which the percentages add up to more than 100% because of the cross-classification of tokens. 32 3. Data and Methodology Table 3.3. Classification of tokens of misunderstanding Classification Pronunciation Lexis Grammar Code-Switching Miscellaneous Total (%) 158 (86.3%) 41 (22.4%) 25 (13.7%) 4 (2.2%) 2 (1.1%) It can be seen from Table 3.3 that the overwhelming majority of tokens of misunderstanding arise because of pronunciation, though it should be noted that 45 out of these 158 tokens involve something else as well. The figures shown here are broadly in line with those of Jenkins (2000: 85–87), who reports that among her 40 tokens, 27 tokens were classified as involving pronunciation, eight as lexis, just one as grammar, and four others. We might further note that, although the percentage of tokens involving grammar is rather higher than that reported by Jenkins, in fact in most of the 25 tokens, grammar appears to be quite a minor factor, as we will see in Chapter 5. The proportion in the CMACE data arising from pronunciation is much higher than the percentage found by Smit (2010: 202) who reports that just 4.1% of the repairs in her data occurred because of pronunciation. However, the data is very different, as her data was collected from participants who were together for two years while they studied together on a Hotel Management Programme, so it is hardly surprising that they got used to the patterns of speech of their classmates and teachers. Moreover, Smit reports that the occurrence of ‘mishearings’ in her data, some of which are related to pronunciation, decreased substantially over time, so 54.5% of these repairs occurred in the first few weeks when the participants had just started to get to know each other, while only 13.6% occurred during the third and final time that she investigated the same participants, almost two years later. In contrast, she reports that there was no such reduction over time in repairs arising out of issues of fact or because of vocabulary. As pronunciation seems to give rise to the majority of the tokens of misunderstanding in my data involving speakers who have not met each other before, I will consider it first in Chapter 4, before I discuss other factors in Chapter 5. References Algeo, John 2005 The Origins and Development of the English Language, 6th ed. Boston, MA: Wadsworth Cengage. Archibald, Alasdair, Alessia Cogo, and Jennifer Jenkins (eds.) 2011 Latest Trends in ELF Research. Newcastle, UK: Cambridge Scholars Publishing. Baskaran, Loga 2004a Malaysian English: Phonology. In A Handbook of Varieties of English. Volume 1: Phonology, Edgar W. Schneider, Kate Burridge, Bernd Kortmann, Rajend Mesthrie, and Clive Upton (eds.), 1034– 1046. Berlin: Mouton de Gruyter. Baskaran, Loga 2004b Malaysian English: Morphology and syntax. In A Handbook of Varieties of English. Volume 2: Morphology and Syntax, Bernd Kortmann, Kate Burridge, Rajend Mesthrie, Edgar W. Schneider, and Clive Upton (eds.), 1073–1085. Berlin: Mouton de Gruyter. Björkman, Beyza 2009 From code to discourse in spoken ELF. In English as a Lingua Franca: Studies and Findings, Anna Mauranen and Elina Ranta (eds.), 225–251. Newcastle, UK: Cambridge Scholars Press. Brown, Adam 1999 Singapore English in a Nutshell An Alphabetical Description of its Features. Singapore: Federal. CAA 2012 CAP 413: Radiotelephony manual, 20th ed. Norwich, UK: UK Civil Aviation Authority. http://www.caa.co.uk/docs/33/CAP413W.pdf. Accessed 22 March 2013. Carter, Ronald, and Michael McCarthy 2006 Cambridge Grammar of English: A Comprehensive Guide. Cambridge: Cambridge University Press. Celce-Murcia, Marianne, Donna M. Brinton, and Janet M. Goodwin 1996 Teaching Pronunciation: A Reference for Teachers of English to Speakers of Other Languages. Cambridge: Cambridge University Press. Clynes, Adrian, and David Deterding 2011 Standard Malay (Brunei). Journal of the International Phonetic Association 41 (2): 259–268. 192 References Cogo, Alessia 2009 Accommodating difference in ELF conversations. In English as a Lingua Franca: Studies and Findings, Anna Mauranen and Elina Ranta (eds.), 254–273. Newcastle, UK: Cambridge Scholars Press. Cogo, Alessia, and Martin Dewey 2012 Analysing English as a Lingua Franca: A Corpus-driven Investigation. London: Continuum. Couper, Graeme 2011 What makes pronunciation teaching work? Testing for the effect of two variables: Socially constructed metalanguage and critical listening. Language Awareness 20 (3): 159–182. Cruttenden, Alan 2008 Gimson’s pronunciation of English, 7th ed. London: Hodder Education. Crystal, David 2003 English as a Global Language, 2nd ed. Cambridge: Cambridge University Press. Davies, Alan 2003 The Native Speaker: Myth and Reality. Clevedon, UK: Multilingual Matters. Deterding, David 2005 Listening to Estuary English in Singapore. TESOL Quarterly 39 (3): 425–440. Deterding, David 2006 The pronunciation of English by speakers from China. English World-Wide 27 (2): 175–198. Deterding, David 2007 Singapore English. Edinburgh: Edinburgh University Press. Deterding, David 2012 Issues in the acoustic measurement of rhythm. In Pragmatics and Prosody in English Language Teaching, Jesús Romero-Trillo (ed.), 9–24. Dordrecht: Springer. Deterding, David, and Andy Kirkpatrick 2006 Emerging South-East Asian Englishes and intelligibility. World Englishes 25 (3/4): 391–409. Deterding, David, and Salbrina Sharbawi 2013 Brunei English: A New Variety in a Multilingual Society. Dordrecht: Springer. Deterding, David, Jennie Wong, and Andy Kirkpatrick 2008 The pronunciation of Hong Kong English. English World-Wide 29 (2): 148–175. References 193 Dewey, Martin 2009 English as a Lingua franca: Heightened variability and theoretical implications. In English as a Lingua Franca: Studies and Findings, Anna Mauranen and Elina Ranta (eds.), 60–83. Newcastle, UK: Cambridge Scholars Press. Docherty, Gerry 1992 The Timing of Voicing in British English Obstruents. Berlin: Foris. Dörnyei, Zoltán 2007 Research Methods in Applied Linguistics. Oxford: Oxford University Press. Duanmu, San 2007 The Phonology of Standard Chinese, 2nd ed. Oxford: Oxford University Press. Ehrenreich, Susanne 2009 English as a Lingua Franca in multilingual corporations—Exploring business communities of practice. In English as a Lingua Franca: Studies and Findings, Anna Mauranen and Elina Ranta (eds.), 126– 151. Newcastle, UK: Cambridge Scholars Publishing. Field, John 2008 Bricks or mortar: Which parts of the input does a second language listener rely on? TESOL Quarterly 42: 411–432. Firth, Alan 1996 The discursive accomplishment of normality: On ‘lingua franca’ English and conversation analysis. Journal of Pragmatics 26: 237– 259. Fletcher, Janet 2010 The prosody of speech: Timing and rhythm. In The Handbook of Phonetic Sciences 2nd ed. William J. J Hardcastle, John Laver, and Fiona E. Gibbon (eds.), 523–602. Oxford: Wiley-Blackwell. Fowler, Helen 2010 ELF and the pronunciation goals of international students. Speak Out! The Newsletter of the IATEFL Pronunciation Special Interest Group 42: 10–13. Gargesh, Ravinder 2004 Indian English: phonology. In A Handbook of Varieties of English. Volume 1: Phonology, Edgar W. Schneider, Kate Burridge, Bernd Kortmann, Rajend Mesthrie, and Clive Upton (eds.), 992–1002. Berlin: Mouton de Gruyter. 194 References Giles, Howard, Justine Coupland, and Nikolas Coupland 1991 Accommodation theory: Communication, context, and consequence. In Contexts of Accommodation, Howard Giles, Justine Coupland and Nikolas Coupland (eds.), 1–68. Cambridge: Cambridge University Press. Grabe, Esther, Greg Kochanski, and John Coleman 2005 The intonation of native accent varieties in the British Isles: Potential for miscommunication? In English Pronunciation Models: A Changing Scene, Katarzyna Dziubalska-Kołaczyk and Joanna Przedlacka (eds.), 311–337. Bern: Peter Lang. Gut, Ulrike 2004 Nigerian English: phonology. In A Handbook of Varieties of English. Volume 1: Phonology, Edgar W. Schneider, Kate Burridge, Bernd Kortmann, Rajend Mesthrie, and Clive Upton (eds.), 813–830. Berlin: Mouton de Gruyter. Hancock, Mark 2013 At the talk face 9: Accommodation games 2. Speak Out! The Newsletter of the IATEFL Pronunciation Special Interest Group 48: 30– 31. Have, Paul ten 1999 Doing Conversation Analysis. London: Sage Publications. Holliday, Adrian 2005 The Struggle to Teach English as an International Language. Oxford: Oxford University Press. Horvath, Barbara M., and Ronald J. Horvath 2001 A multilocality study of a sound change in progress: The case of /l/ vocalization in New Zealand and Australian English. Language Variation and Language Change 13: 37–57. House, Juliane 1999 Misunderstanding in intercultural communication: Interactions in English as lingua franca and the myth of mutual intelligibility. In Teaching and Learning English as a Global Language, C. Gnutzmann (ed.), 73–89. Tübingen: Stauffenburg. House, Juliane 2003 English as a lingua franca: A threat to multilingualism? Journal of Sociolinguistics 7: 556–578. Hughes, Rebecca 2005 Investigating turn-taking in the NIE corpus of spoken Singapore English. In English in Singapore: Phonetic Research on a Corpus, David Deterding, Adam Brown, and Low Ee Ling (eds.), 115–125. Singapore: McGraw-Hill. References 195 Hülmbauer, Cornelia 2009 “We don’t take the right way. We just take the way we think you’ll understand” – The shifting relationship between correctness and effectiveness in ELF. In English as a Lingua Franca: Studies and Findings, Anna Mauranen and Elina Ranta (eds.), 323–347. Newcastle, UK: Cambridge Scholars Press. Hülmbauer, Cornelia 2013 From within and without: The virtual and the plurilingual in ELF. Journal of English as a Lingua Franca 2 (1): 47–73. Hüttner, Julia 2009 Fluent speakers–fluent interactions: On the creation of (co)-fluency in English as a Lingua Franca. In English as a Lingua Franca: Studies and Findings, Anna Mauranen and Elina Ranta (eds.), 225–297. Newcastle, UK: Cambridge Scholars Press. Jenkins, Jennifer 1996 Native speaker, non-native speaker and English as a Foreign Language: Time for a change. IATEFL Newsletter 131: 10–11. Jenkins, Jennifer 2000 The Phonology of English as an International Language. Oxford: Oxford University Press. Jenkins, Jennifer 2005 Misinterpretation, bias, and resistance to change: The case of the Lingua Franca Core. In English Pronunciation Models: A Changing Scene, Katarzyna Dziubalska-Kołaczyk and Joanna Przedlacka (eds.), 199–210. Bern: Peter Lang. Jenkins, Jennifer 2006 Global intelligibility and local diversity: Possibility of paradox? In English in the World: Global Rules, Global Roles, Rani Rubdy and Mario Saraceni (eds.), 32–39. London: Continuum. Jenkins, Jennifer 2007 English as a Lingua Franca: Attitude and Identity. Oxford: Oxford University Press. Jenkins, Jennifer 2009 World Englishes: A Resource Book for Students, 2nd ed. London: Routledge. Jenkins, Jennifer, Alessia Cogo, and Martin Dewey 2011 Review of developments in research into English as a lingua franca. Language Teaching 44 (3): 281–315. Kachru, Braj B. 1985 Standards, codification and sociolinguistic realism: The English language in the outer circle. In English in the World: Teaching and Learning the Language and Literatures, Randolph Quirk and Henry Widdowson (eds.), 11–30. Cambridge: Cambridge University Press. 196 References Kachru, Braj B. 2005 Asian Englishes: Beyond the Canon. Hong Kong: Hong Kong University Press. Kalocsai, Karolina 2011 The show of interpersonal involvement and the building of rapport in an ELF community of practice. In Latest Trends in ELF Research, Alasdair Archibald, Alessia Cogo and Jennifer Jenkins (eds.), 113– 137. Newcastle, UK: Cambridge Scholars Publishing. Kaur, Jagdish 2009 Pre-empting problems of understanding in English as a Lingua Franca. In English as a Lingua Franca: Studies and Findings, Anna Mauranen and Elina Ranta (eds.), 107–123. Newcastle, UK: Cambridge Scholars Publishing. Kaur, Jagdish 2010 Achieving mutual understanding in world Englishes. World Englishes 29: 192–208. Kirkpatrick, Andy 2007 World Englishes: Implications for International Communication and English Language Teaching. Cambridge: Cambridge University Press. Kirkpatrick, Andy 2010 English as a Lingua Franca in ASEAN: A Multilingual Model. Hong Kong: Hong Kung University Press. Kitzinger, Celia 2013 Repair. In The Routledge Handbook of Conversation Analysis, Jack Sidnell and Tanya Stivers (eds.), 228–256. London and New York: Routledge. Klimpfinger, Theresa 2009 “She’s mixing the two languages together” – Forms and functions of code-switching in English as a Lingua Franca. In English as a Lingua Franca: Studies and Findings, Anna Mauranen and Elina Ranta (eds.), 348–371. Newcastle, UK: Cambridge Scholars Press. Ladefoged, Peter, and Keith Johnson 2011 A Course in Phonetics, 6th ed. USA: Wadsworth Cengage Learning. Laver, John 1994 Principles of Phonetics. Cambridge: Cambridge University Press. Levis, John 2005 Prominence in Singapore and American English: Evidence from reading aloud. In English in Singapore: Phonetic Research on a Corpus, David Deterding, Adam Brown, and Low Ee Ling (eds.), 86–94. Singapore: McGraw-Hill. References 197 Liddicoat, Anthony J. 2011 An Introduction to Conversation Analysis. 2nd ed. London: Continuum. Local, John 2005 On the interactional and phonetic design of collaborative completions. A Figure of Speech: A Festschrift for John Laver, William J. Hardcastle and Janet Mackenzie Beck (eds.), 263–282. Mahwah, NJ: Lawrence Erlbaum Associates. Low Ee Ling, Esther Grabe, and Francis Nolan 2000 Quantitative characterizations of speech rhythm: syllable-timing in Singapore English. Language and Speech 43: 377–401. Matsumoto, Yumi 2011 Successful ELF communications and implications for ELT: Sequential analysis of ELF pronunciation negotiation strategies. The Modern Language Journal 95: 97–104. Mauranen, Anna 2006 Signaling and preventing misunderstanding in English as a lingua franca communication. International Journal of the Sociology of Language 177: 123–150. Mauranen, Anna, and Elina Ranta (eds.) 2009 English as a Lingua Franca: Studies and Findings. Newcastle, UK: Cambridge Scholars Publishing. McCrocklin, Shannon 2012 The role of word stress in English as a lingua franca. In Proceedings of the 3rd Pronunciation in Second Language Learning and Teaching Conference, Sept 2011, John Levis and Kimberly LeVelle (eds.), 249–256. Ames, IA: Iowa State University. McKay, Sandra Lee 2002 Teaching English as an International Language. Oxford: Oxford University Press. McLellan, James 2010. Mixed codes or varieties of English. In The Routledge handbook of World Englishes, Andy Kirkpatrick (ed.), 425–441. London and New York: Routledge. Mesthrie, Rajend, and Rakesh M. Bhatt 2008 World Englishes: The Study of Linguistic Varieties. Cambridge: Cambridge University Press. Mortensen, Janus 2013 Notes on English used as a lingua franca as an object of study. Journal of English as Lingua Franca 2 (1): 25–46. 198 References Mugglestone, Lynda 2003 Talking Proper: The Rise of Accent as Social Symbol. Oxford: Oxford University Press. Munro, Murray J. 2008 Foreign accent and speech intelligibility. In Phonology and Second Language Acquisition, Jette G. Hansen Edwards and Mary L. Zampini (eds.), 193–218. Amsterdam: John Benjamins. Munro, Murray J., Tracey M. Derwing, and Susan L. Morton 2006 The mutual intelligibility of L2 speech. Studies in Second Language Acquisition 28: 11–131. Nelson, Cecil L. 2011 Intelligibility in World Englishes: Theory and Application. London & New York: Routledge. Nelson, Gerald, Sean Wallis, and Bas Aarts 2002 Exploring Natural Language: Working with the British Component of the International Corpus of English. Amsterdam: John Benjamins. Osimk, Ruth 2011 Decoding sounds: An experimental approach to intelligibility in ELF. VIEWS 18 (1): 64–89. Phillipson, Robert 1992 Linguistic Imperialism. Oxford: Oxford University Press. Pickering, Lucy 2006 Current research on intelligibility in English as a lingua franca. Annual Review of Applied Linguistics 26: 219–233. Pitzl, Marie-Luise 2009 “We should not wake up any dogs”: Idiom and metaphor in ELF. In English as a Lingua Franca: Studies and Findings, Anna Mauranen and Elina Ranta (eds.), 298–322. Newcastle, UK: Cambridge Scholars Press. Pointon, Graham 2012 The tentacles of spelling pronunciations. Linguicism: Language in a Word. Accessed 28 October 2012 from http://www.linguism. co.uk/language/the-tentacles-of-spelling-pronunciations Prodromou, Luke 2008 English as a Lingua Franca: A Corpus-based Analysis. London: Continuum. Pullin Stark, Patricia 2009 No joke—This is serious! Power, solidarity and humour in Business English as a lingua franca. In English as a Lingua Franca: Studies and Findings, Anna Mauranen and Elina Ranta (eds.), 152–177. Newcastle, UK: Cambridge Scholars Publishing. References 199 Quirk, Randolph 1990 Language varieties and standard language. English Today 21: 3–10. [Reprinted in Barbara Seidlhofer, Controversies in Applied Linguistics, Oxford: Oxford University Press, 2003: 9–19.] Riney, Timothy J., Naoyuki Takagi, and Kumiko Inutsuka 2005 Phonetic parameters and perceptual judgments of accent in English by American and Japanese listeners. TESOL Quarterly 39 (3): 441– 466. Roach, Peter 1998 Myth 18: Some languages are spoken more quickly than others. In Language Myths, Laurie Bauer and Peter Trudgill (eds.), 150–158. London: Penguin. Roach, Peter 2009 English Phonetics and Phonology: A practical Course. 4th ed. Cambridge: Cambridge University Press. Sailaja, Pingali 2009 Indian English. Edinburgh: Edinburgh University Press. Saito, Kazuya 2012 Effects of instruction on L2 pronunciation development: A synthesis of 15 quasi-experimental intervention studies. TESOL Quarterly 46: 842–854. Saito, Kazuya 2012 Effects of instruction on L2 pronunciation development: A synthesis of 15 quasi-experimental intervention studies. TESOL Quarterly 46: 842–854. Salbrina Sharbawi 2006 The vowels of Brunei English: An acoustic investigation. English World-Wide 27: 247–264. Scheuer, Sylwia 2005 Why native speakers are (still) relevant. In English Pronunciation Models: A Changing Scene, Katarzyna Dziubalska-Kołaczyk and Joanna Przedlacka (eds.), 111–130. Bern: Peter Lang. Schneider, Edgar W. 2007 Postcolonial English: Varieties around the World. Cambridge: Cambridge University Press. Schneider, Edgar W. 2011 English around the World: An Introduction. Cambridge: Cambridge University Press. Seidlhofer, Barbara 2001 Closing the conceptual gap: The case for a description of English as a Lingua Franca. International Journal of Applied Linguistics 11 (2): 133–158. 200 References Seidlhofer, Barbara 2009 Orientations in ELF research: Form and function. In English as a Lingua Franca: Studies and Findings, Anna Mauranen and Elina Ranta (eds.), 37–59. Newcastle, UK: Cambridge Scholars Press. Seidlhofer, Barbara 2011 Understanding English as a Lingua Franca. Oxford: Oxford University Press. Seidlhofer, Barbara, and Henry Widdowson 2009 Accommodation and the idiom principle in English as a Lingua Franca In Global Englishes in Asian Contexts: Current and Future Debates, Kumiko Murata and Jennifer Jenkins (eds.), 26–39. Palgrave Macmillan. Setter, Jane, Cathy S. P. Wong, and Brian H. S. Chan. 2010 Hong Kong English. Edinburgh: Edinburgh University Press.. Shaw, Philip, Tim Caudery, and Margrethe Petersen 2009 Students on exchange in Scandinavia: Motivation, interaction, ELF development. In English as a Lingua Franca: Studies and Findings, Anna Mauranen and Elina Ranta (eds.), 178–199. Newcastle, UK: Cambridge Scholars Publishing. Sinclair, John 1985 Selected issues. In English in the World: Teaching and Learning the Language and Literatures, Randolph Quirk and H. G. Widdowson (eds.), 248–254. Cambridge: Cambridge University Press. Smit, Ute 2010 English as a Lingua Franca in Higher Education: A Longitudinal Study of Classroom Discourse. Berlin/New York: De Gruyter Mouton. Smith, Larry E. 1992 Spread of English and issues of intelligibility. In The Other Tongue: English across Cultures, Braj B. Kachru (ed.), 74–90. Urbana and Chicago: University of Illinois Press. Smith, Larry E., and John A. Bisazza 1982 The comprehension of three varieties of English for college students in seven countries. Language Learning 32: 259–270. Smith, Larry E., and Khalilulla Rafiqzad 1979 English for cross-cultural communication: The question of intelligibility. TESOL Quarterly 13: 371–380. Sobkowiak, Włodzimierz 2005 Why not LFC? In English Pronunciation Models: A Changing Scene, Katarzyna Dziubalska-Kołaczyk and Joanna Przedlacka (eds.), 131– 149. Bern: Peter Lang. References 201 Swan, Michael 1987 German speakers. In Learner English: A Teacher's Guide to Interference and Other Problems, Michael Swan and Bernard Smith (eds.), 30–41. Cambridge: Cambridge University Press. Sweeney, Emma, and Zhu Hua 2010 Accommodating towards the audience. Do native speakers of English know how to accommodate their communication strategies towards non-native speakers of English? Journal of Business Communication, 47(4), 477–504. Tan Kah Keong 2005 Vocalisation of /l/ in Singapore English. In English in Singapore: Phonetic Research on a Corpus, David Deterding, Adam Brown, and Low Ee Ling (eds.), 43–53. Singapore: McGraw-Hill. Teschner, Richard V., and M. Stanley Whitley 2004. Pronouncing English: A Stress-based Approach with CD-ROM. Washington, DC: Georgetown University Press. Trudgill, Peter 2005 Native-speaker segmental phonological models and the English Lingua Franca Core. In English Pronunciation Models: A Changing Scene, Katarzyna Dziubalska-Kołaczyk and Joanna Przedlacka (eds.), 77–98. Bern: Peter Lang. VOICE 2011 The Vienna-Oxford International Corpus of English (version 1.1 online). Director: Barbara Seidlhofer; Researchers: Angelika Breiteneder, Theresa Klimpfinger, Stefan Majewski, Ruth Osimk, Marie-Luise Pitzl. http://voice.univie.ac.at (accessed 18 October 2011). Walker, Robyn 2010 Teaching the Pronunciation of English as a Lingua Franca. Oxford: Oxford University Press. Wells, J. C. 1982 Accents of English. Cambridge: Cambridge University Press. Wells, J. C. 2006 English Intonation: An Introduction. Cambridge: Cambridge University Press. Wells, J. C. 2008 Longman Pronunciation Dictionary. 3rd ed. Harlow, UK: Pearson Longman. Wikipedia 2013 Joint Admissions and Matriculation Board. http://en.wikipedia.org/ wiki/Joint_Admissions_and_Matriculation_Board (accessed 22 February 2013). Author Index Aarts, Bas, 20 Algeo, John, 73 Archibald, Alasdair, 5 Baskaran, Loga, 34, 65, 70, 80, 157 Bhatt, Rakesh M., 108 Bisazza, John A., 11 Björkman, Beyza, 11, 26, 120 Brinton, Donna M., 79 Brown, Adam, 100 Carter, Ronald, 102, 103, 158, 174 Caudery, Tim, 9, 11, 17 Celce-Murcia, Marianne, 79 Chan, Brian H. S., 157 Clynes, Adrian, 53, 55 Cogo, Alessia, 5, 6, 8, 12, 22, 112, 114, 119, 123, 125, 157 Coleman, John, 79, 81 Couper, Graeme, 51, 172 Coupland, Justine, 16 Coupland, Nikolas, 16 Cruttenden, Alan, 43, 47, 48, 51, 80 Crystal, David, 80 Davies, Alan, 1, 22 Derwing, Tracy M., 10 Deterding, David, 34, 40, 41, 42, 51, 53, 55, 65, 67, 70, 73, 77, 78, 80, 101, 103, 104, 108, 110, 125, 157, 170, 173, 175 Dewey, Martin, 5, 6, 8, 22, 111, 112, 114, 119 Docherty, Gerry, 56, 58 Dörnyei, Zoltán, 19 Duanmu, San, 124 Ehrenreich, Susanne, 15, 22, 169 Field, John, 70 Firth, Alan, 3, 15, 19, 123, 139 Fletcher, Janet, 81, 83 Fowler, Helen, 8, 71 Gargesh, Ravinder, 65 Giles, Howard, 16 Goodwin, Janet M., 79 Grabe, Esther, 79, 80, 81 Gut, Ulrike, 65, 80 Hancock, Mark, 176 Have, Paul ten, 14, 17, 18, 24 Holliday, Adrian, 168, 169, 174 Horvath, Barbara, 51 Horvath, Ronald J., 51 House, Juliane, 1, 11, 16, 24, 162, 164, 175 Hülmbauer, Cornelia, 98, 108 Hüttner, Julia, 12, 81 Hughes, Renecca, 81 Inutsuka, Kumiko, 41, 44 Jenkins, Jennifer, 2, 4, 5, 6, 7, 8, 9, 16, 22, 32, 33, 34, 54, 55, 63, 67, 68, 70, 73, 74, 77, 80, 98, 168, 169 Johnson, Keith, 56, 79 Kachru, Braj, 1, 4, 6 Kalocsai, Karolina, 123 Kaur, Jagdish, 12, 13, 26, 34, 131, 138, 157 Kirkpatrick, Andy, 5, 6, 9, 34, 40, 41, 42, 67, 77, 78, 168, 169, 170, 174 Kitzinger, Celia, 14, 16 Klimpfinger, Theresa, 123 Kochanski, Greg, 79, 81 Author index Ladefoged, Peter, 56, 79 Laver, John, 79, 136 Levis, John, 77 Liddicoat, Anthony, 14, 18, 19 Local, John, 166 Low Ee Ling, 80 Matsumoto, Yumi, 8, 22, 26 Mauranen, Anna, 5, 12, 22, 26, 33 McCarthy, Michael, 102, 103, 158, 174 McCrocklin, Shannon, 8 McKay, Sandra Lee, 1, 22, 171 McLellan, James, 123 Mesthrie, Rajend, 108 Mortensen, Janus, 15 Morton, Susan L., 10 Mugglestone, Lynda, 40 Munro, Murray J., 10, 11, 169, 177 Nelson, Cecil L., 2, 10, 133 Nelson, Gerald, 20 Nolan, Francis, 80 Osimk, Ruth, 8, 9 203 Ranta, Elina, 5 Riney, Timothy J., 41, 44 Roach, Peter, 40, 51, 68, 81 Sailaja, Pingali, 80 Saito, Kazuya, 172 Salbrina Sharbawi, 40, 70, 73, 101, 103, 108, 110, 125 Scheuer, Sylwia, 9, 169 Schneider, Edgar W., 6, 174 Seidlhofer, Barbara, 1, 4, 5, 20, 24, 25, 98, 111, 114, 168, 174 Setter, Jane, 157 Shaw, Philip, 9, 11, 17 Sinclair, John, 24 Smit, Ute, 5, 14, 19, 22, 23, 25, 32, 34, 123, 139 Smith, Larry E., 4, 9, 10, 11, 13, 133 Sobkowiak, Włodzimierz, 9, 169 Swan, Michael, 34 Sweeney, Emma, 188 Takagi, Naoyuki, 41, 44, Teschner, Richard V., 79 Trudgill, Peter, 169 Quirk, Randolph, 4 Walker, Robyn, 173, 174 Wallis, Sean, 20 Wells, J. C., ix, 34, 40, 51, 56, 64, 67, 73, 74, 80, 81 Whitley, M. Stanley, 79 Widdowson, Henry, 5, 98 Wikipedia, 48 Wong, Cathy S. P., 157 Wong, Jennie, 34, 41 Rafiqzad, Khalilulla, 11 Zhu Hua, 188 Petersen, Margrethe, 9, 11, 17 Phillipson, Robert, 174 Pickering, Lucy, 10, 12, 85 Pitzl, Marie-Luise, 98, 100 Pointon, Graham, 75 Prodromou, Luke, 12, 98 Pullin Stark, Patricia, 154 Subject Index accent reduction, 169 accentedness, 10, 176 accommodation, 12, 16–17, 37, 39, 55, 85–87, 90, 98, 123, 125, 157, 167, 168, 174–176 Achebe, Chinua, 10 acoustics, 56 affricates, 58–61, 89 air traffic communication, 40, 71 alpha bravo charlie delta, 144 approximants, 41–47, 89 Arabic, 125, 173 articles, 111–114, 132 Asian Corpus of English (ACE), 20– 21 asking for clarification, 14, 23, 139– 144 asking for help, 14, 136–138 aspiration, 7, 8, 52, 55–57, 63, 75, 89, 91, 97, 118 attitudes, 9, 12 Australia, 1, 6, 10, 11, 51, 64, 80 Austria, 123 backchannels, 15, 139, 147–149 baju kurung, 127, 136, 137 Berlitz, 60, 74, 91, 184 Bhutan, 42, 55, 89, 151, 179 Black Swan, 28, 43, 66, 89, 114, 179 Brazil, 6, 176 Brunei, 1, 20, 21, 40, 69, 72, 73, 80, 94, 110, 123, 125, 127, 152 Canada, 10, 169 changing the topic, 139, 149, 152– 153 China, 15, 21, 34, 51, 172, 173 Chinese (Mandarin), 56, 69, 93, 100, 101, 117, 123, 124, 126, 141, 173 Civil Aviation Authority (CAA), ix, 40, 71, 191 CMACE, 20–27 code-switching, 32, 123–128 collaborative completions, 137, 138, 157, 164–166 collage, 46, 60, 91, 93, 154, 183 communities of practice, 22 competitive overlaps, 164 comprehensibility, 9, 10, 133 consonant clusters, 7, 30, 44–51, 90, 91, 172 consonants, 34–63, 89, 91 content words, 70 continuum of understanding, 13 conventional wisdom, 12 Conversation Analysis (CA), 17–19, 24 copular verb, 117–118 core features, 7 corpora, 20 Corpus of Contemporary American English (COCA), ix, 30–31, 95, 96 correcting, 14, 139, 144–145 correcting oneself, 131–132 critical listening, 172 cutlass, 27, 93, 94, 188 dark-L, 7, 51–55 debacle, 75 Denmark, 9, 11, 17, 123 dental fricatives (see TH sounds) discourse, 17, 79, 112, 157–159 dorm, 93, 94, 95, 122, 183 echoing, 162–165 ELFA corpus 22 elite satay, 101, 105, 184 Subject index endangered fields, 5, 98 English as a lingua franca (ELF), 1– 2, 4–7 English as an International Language (EIL), 5 enhancing intelligibility, 2, 3, 9, 11, 34, 40, 51, 71, 73, 80, 112, 157, 158, 159, 170, 173, 177 epenthetic vowel, 48, 51, 172 equipments, 3, 108, 110 Erasmus exchange programme, 123 Estuary English, 51, 175 ever, 103–104 exceptions, 17 Expanding Circle, 1–2, 6, 11, 20, 34, 169, 177 familiarity, 11, 12, 67 far-away uncle, 98 feedback, 18, 26, 172 find, 107, 135, 186 Finland, 26 fixed phrases, 94–98 flaming hot, 30, 31, 45, 90, 94, 180 flexibility, 172 floating, 28, 38, 39, 45, 90, 159, 160, 182 fluency, 12, 79, 98, 173 form, 17, 25, 112 France, 8 fricatives, 57–61, 89 function, 17, 25, 112 function words, 3, 70 furnitures, 3, 108, 110 garment, 50, 89, 142, 179 generic nouns, 112 German, 173 Germany, 6, 8, 15, 22, 34, 52, 123 glottal stop, 37, 56, 58, 61, 66, 68 going native, 71, 72, 91, 94, 96, 186 grammar (syntax), 32, 91, 107–102, 171 205 great leap forward, 63, 91, 99, 100, 141, 142, 189 Greek, 173 Guangdong Province, 41 H-dropping, 40–41, 89, 91 H-insertion, 40–41, 89 Hash, 23, 101, 102, 156, 186 holes, 42, 170 hot cake, 25, 41, 68, 89, 99, 100, 134, 189 hotpot, 93, 52, 179 Hungary, 123 Hungarian, 123 hyper-articulation, 48 idioms, 5, 12, 22, 25, 98–100 imperialism, 174 India, 6, 11, 20, 65, 80 Indonesia, 21 Indonesian, 101, 111 informations, 108 Inner Circle, 1–2, 6, 10, 11, 15, 20, 34, 70, 73, 75, 76, 79, 80, 81, 115, 169 instances of misunderstanding, 13, 27 intelligibility, 1, 2, 7, 9–12, 176 International Corpus of English (ICE), 20 interpersonal conflict, 16 interpretability, 9, 10 intonation, 79–81, 169 intonational nucleus (see stress) Ireland, 34 irritation, 11 Istanbul, 5 Italian, 125 Italy, 8 Japan, 6, 8, 21, 22, 33, 41, 44, 70 Japanese, 173 joint supervision, 47, 78, 82, 90, 184 joking, 154–155 206 Subject index Journal of English as a Lingua Franca (JELF), 5 Karl Marx, 45, 61, 90, 99, 120, 185 Kirkpatrick, Andy, 20 knuckle, 52, 91, 92, 161, 189 Korea, 51, 172 L-vocalization, 51–55 Labuan, 43, 89, 141, 159, 187 language change, 73 Laos, 21, 42, 170 laughter, 62, 85, 90, 119, 139, 154– 156 laziness, 177 learneds, 101, 102, 147, 189 lexical repetition, 159–162 lexis, 32, 91, 92–107 Lingua Franca Core (LFC), 7–9 London, 34, 40, 51 Los Angeles, 17 lunchtime, 43, 59, 88, 89, 150, 151, 187 makan, 125 Malaysia, 1, 11, 20, 21, 65, 70, 80, 157, 170 Man U, 66, 90, 155, 187 Mandarin (see Chinese) meal plan, 45, 52, 66, 90, 94, 183 Miliband, David, 75 mimicking native speakers, 8, 168, 169, 177 missing verbs, 117–119 misunderstandings, 1, 12–14 mm, 15, 147 mm-mm, 148 mother tongue, 1 names, 93 native speakers, 1, 15, 16, 22, 44, 51, 80, 81, 157, 166, 168, 174, 177 natural repairs, 15 nearby, 42, 89, 151, 179 New York, 34 New Zealand, 51, 64 next, 42, 89, 113, 132, 160, 171, 180 niche, 59, 60, 74, 91, 93, 133, 134, 186 Nigeria, 1, 6, 21, 48, 65, 80, 102 non-awareness, 139, 156–157 non-core features, 7 non-understandings, 13 of, 120 orchard, 59, 60, 144, 190 one of, 110–111 Outer Circle, 1–2, 6, 10, 11, 20, 22, 34, 51, 177 overlapping speech, 24, 81, 164 P6, 55, 56, 89, 94, 97, 189 Pakistan, 23 paraphrasing, 12, 14, 132–135 pattern-drilling, 172 pausing, 12, 16, 166 pedagogy, 168–178 pehin menteri ugama, 127, 128, 148, 184 pepper crab, 94, 97, 150, 189 Philippines, 20, 42, 170 plosives, 7, 8, 55, 56, 89, 91 plural nouns, 3, 11, 108–111 Poland, 9, 169 Polish, 173 polysemes, 105–107 portion, 27, 101, 102, 188 Portuguese, 173 pound, 106, 140, 183 pragmatics, 9, 13, 175 prepositions, 119–120 prestige, 40 pronunciation, 7, 33–87, 171–173 pronunciation teaching, 172–173 quantification, 18 questionnaires, 11 Subject index recordings, 20–24 reference materials, 174 repairs, 14–16, 131–167 other-initiated repairs, 14, 138– 145 other-repairs, 14 self-initiated repairs, 14, 131–138 self-repairs, 14 repetition, 12, 159–162 re-present, 16, 162 resumptive pronouns, 3 rhythm, 7, 34, 79–80 rhythm metrics, 80 RP British English, 11, 169 Russian, 173 Sacks, Harvey, 17 Salbrina Sharbawi, 20 salmon, 73 same as, 112 Seidlhofer, Barbara, 20, 24 selecting part, 139, 150–152 semantics, 101–105 send, 100–101 silence, 139, 145–147 Singapore, 6, 11, 20, 34, 51, 65, 69, 73, 80, 97, 104, 157 six three three four, 35, 82, 83, 90, 97, 188 sleeping dogs, 98 sorry, 141–142 South China, 42, 44 South-East Asia, 22, 33, 34, 39, 53, 56, 57, 67, 69, 70, 73, 77, 78, 91, 101, 103, 104, 105, 130, 157, 158, 168 Spain, 8 Spanish, 173 speaking rate, 12, 16, 62, 67, 72, 75, 81, 82, 90, 169, 173 spelling out, 144 spelling pronunciation, 73–74, 90 standard grammar, 5 207 stress, 24, 74–79 utterance stress, 7, 77–79, 90 word stress, 7, 8, 71, 74–77, 90, 91 structure of interactions, 19 suffixes -ed suffix, 116–117 -ing suffix, 115 -s plural suffix, 3, 109–111 -s verb suffix, 114–117 suprasegmentals, 79–85, 90, 91 surprise, 154 Sweden, 9, 11, 17, 26 Switzerland, 8, 22, 33 syllables, 83, 90 syntax (see grammar) Taiwan, 1, 21 Tamil, 126, 136 teaching accommodation, 174–176 teaching materials, 174 telephone conversations, 17 teluk belanga, 127, 136, 137, 184 tenses, 11, 116 terminal degree, 31, 52, 65, 84, 91, 94, 95, 180 territorial imperative, 98 testing, 177 textbooks, 174 TH sounds, 7, 8, 34–40, 86, 118, 129, 172, 176 Three Circles of English, 1, 6 tiredness, 12, 85 tokens of misunderstanding, 13, 27, 179–190 tone of voice, 17 topic fronting, 3, 129, 157–159 tow, 37, 86, 90, 183 transcription, x, 24–27 tubers, 74, 90, 184 Turkey, 123 turn-signals, 81 208 Subject index uh-uh, 148 UK (Britain), 1, 6, 8, 56, 73, 74, 79, 80, 102, 174 University of Brunei Darussalam (UBD), 20, 21 until now, 53, 91, 104–105 USA (America), 1, 6, 8, 11, 74, 80, 94, 174 used to, 102–103 usually, 173 valley, 87, 173 verb suffixes, 114 Vienna, 22, 24, 139 vinyl plastic, 52, 91, 93, 94, 183 virgin, 53, 54, 64, 74, 76, 90, 134, 186 VOICE, x, 20, 24, 25, 115, 123 voice onset time (VOT), 56 voicing, 55–58, 68–69, 89 vowel duration (length), 8, 68–70, 73, 90, 91 vowel quality, 7, 8, 63, 71, 73, 91, 172 vowel quantity (see vowel duration) vowel reduction, 3, 7, 34, 70, 75, 80, 90, 169, 176, 177 vowels, 63–73, 90 BATH, 64 DRESS, 65, 72 FACE, 60, 64, 68 FLEECE, 68, 69 GOAT, 64 KIT, 64, 65, 68, 69 LOT, 64, 71, 72 NURSE, 7, 54, 64–65, 90 PRICE, 88 STRUT, 64, 65 TRAP, 65–67, 72, 90 weak forms, 70 weather, 38, 91, 129, 150, 158, 159, 187 word order, 11, 120–121 word selection, 94, 145 working memory, 16 World Englishes, 6 yeah, 15, 25, 148 zen, 57, 71, 72, 89, 124, 181 zero-initiated self-repairs, 15