In the past, pronunciation instruction has often been restricted to phonemic awareness and production in very limited linguistic and communicative contexts. For many reasons, this approach, in which prosodic and/or discourse-related phenomena (e.g. reduction, elision) play a minor role, has been relatively ineffective (Celce-Murcia et al. Reference Celce-Murcia, Brinton and Goodwin1996, Fraser Reference Fraser1999). The dissatisfaction with traditional pronunciation teaching has lead to a gradual paradigm shift towards the inclusion of prosodic and/or discourse-related phenomena alongside segmentals in order to improve overall comprehensibility in a second/foreign language (Chun Reference Chun2002, Trouvain & Gut Reference Trouvain and Gut2007). Indeed, pronunciation instruction does not make much sense if the learner does not know how the phonemes s/he is taught should be connected in natural discourse.
In this context, Perspectives on teaching connected speech to second language speakers aims at fostering interest in researching and teaching connected speech (henceforth CS). More specifically, this collection of fourteen papers – some of which are re-editions of previously published articles, e.g. the chapters by James D. Brown & Ann Hilferty, by Yasuko Ito, by Moana Rogerson, and by Nobuko Hasegawa – seeks ‘to facilitate and promote the research on connected speech in applied linguistics and to stimulate and encourage the teaching of connected speech in language classrooms around the world’ (p. 6). Therefore the intended audience for this book consists of second/foreign language teachers, researchers, and materials developers. The volume is thematically organized into five sections: ‘What do we know so far?’, ‘Does connected speech instruction work?’, ‘How should connected speech be taught in English?’, ‘How should connected speech be taught in Japanese?’, and ‘How should connected speech be tested?’. It is rounded off with a comprehensive list of references and a glossary of technical terms. This glossary will not only be valuable for language teachers looking for a definition of a specific CS phenomenon, but also for the reader as the editors did not opt for a unified terminological framework throughout the book. Instead, they used a system of footnotes in which terminological equivalents and cross-references to other chapters are provided.
The state-of-the-art section of the book (‘What do we know so far?’) consists of three introductory chapters, on CS phenomena, CS learning and CS teaching. In chapter 1, James D. Brown & Kimi Kondo-Brown introduce the notion of connected speech and set out to describe the set of phonological phenomena that fall under this general term: word stress, sentence stress and timing, reduction, citation and weak forms of words, elision, intrusion, assimilation, juncture, and contraction. The authors also emphasize that CS phenomena occur in all registers of any natural language.
In chapter 2, Yasuko Ito examines some of the difficulties non-native speakers encounter when they are confronted with CS phenomena. Productively, non-native speakers apply little modifications to their speech (e.g. underuse of reduced forms) until they have reached a high level of proficiency, whereas perceptively the presence of e.g. reduced forms in the input makes it more difficult for them to process utterances in the target language. These results point towards the need to integrate CS into second/foreign language teaching.
In chapter 3, James D. Brown therefore sets out to review the materials available to language teachers interested in CS, including both teacher training materials and ESL/EFL textbooks. Few ESL/EFL textbooks were found to deal with CS phenomena. The chapter discusses nine textboobs that are carefully evaluated with respect to e.g. their purpose, the organization of the lessons, the types of exercises, and their coverage of CS. It also contains a comparison chart in which the main features of each textbook are summarized (pp. 34–45).
The second section of the volume (‘Does connected speech instruction work?’) is concerned with the effect of explicit CS instruction on listening comprehension (chapter 4), word recognition (chapter 5), and input-intake processes (chapter 6). In chapter 4, James D. Brown & Ann Hilferty investigate the effect of CS instruction on Chinese EFL learners' listening comprehension. It is shown that the learners who have learnt to understand reduced forms get better results on three listening comprehension measures than a control group whose members have not received such an instruction. This suggests that CS instruction enhances listening comprehension in a second/foreign language.
Chapter 5, by Takashi Matsuzawa, examines Japanese EFL students' ability to identify target words undergoing different types of CS phenomena (e.g. assimilation, contraction, deletion) and the effect of CS instruction on the learners' listening skills. While some CS phenomena (e.g. flapping, contractions, weak forms of function words) make word identification more difficult than others (e.g. linking, deletion, palatalization), the post-test taken after CS instruction has taken place shows a significant improvement for all CS categories. On these grounds, Matsuzawa concludes that explicit instruction enhances non-native speakers' comprehension of lexical items undergoing pronunciation changes in CS (p. 64).
In chapter 6, Yasuko Ito explores to what extent CS-induced modifications reduce the perceptual saliency of linguistic entities and by doing so interfere with their processing and assimilation into the learner's developing language system (i.e. intake). The results confirm the hypothesis that pronunciation changes in CS make linguistic entities perceptually less salient to non-native ears. Ito also shows that different types of contractions influence the learners' listening comprehension differently, with phonological contractions like I have → I've being more difficult than ‘(rather) lexical’ ones like will not → won't or it does → it doesn't (p. 71). Finally, Ito reports a significant interaction between the type of contraction and the learners' proficiency level, which indicates that listeners with a greater knowledge of the language are able to compensate for reduced perceptual saliency.
The research discussed so far provides evidence for the fact that non-native speakers do benefit from CS instruction, which suggests that it may indeed be desirable to include CS phenomena into the second/foreign language curriculum. With a view on integrating CS into the language classroom, I regret that the papers in the first two sections of the volume are (almost) exclusively concerned with CS-phenomena in relationship with receptive skills. Indeed, productive skills are only discussed (partially) in chapter 2. To my knowledge, however, little work has been done on that issue so far (e.g. Bley-Vroman & Kweon Reference Bley-Vroman and Kweon2002, Kweon Reference Kweon2000). Yet, knowing to what extent (if at all) non-native speakers apply CS pronunciation changes in their own speech and which CS phenomena are the most/the least difficult for them to produce in a (more or less) native-like fashion would enable language teachers and materials developers to better fit their teaching activities to the learners' needs.
The question how CS should be taught to second/foreign language learners is taken up in the volume's third and fourth sections. Section three (‘How should connected speech be taught in English?’) focuses on CS instruction in English. In chapter 7, Moana Rogerson investigates ESL teachers' perspectives on CS. Although most instructors consider themselves familiar with CS phenomena and find them an important and helpful aspect of a second/foreign language learner's listening skills, they devote little teaching time to CS. They attribute this to a lack of suitable teaching materials as many ESL texts provide little information on the linguistic and pragmatic constraints on CS processes (see also chapter 3).
In chapter 8, Robert Cahill outlines a complete pedagogical approach to CS and exemplifies it with the case of reduced interrogatives. His method is based on the concept of mini-lessons, i.e. ‘5–10 minute exercises that will allow the teacher to include linguistic elements that would not normally be incorporated into the syllabus’ (p. 103). The first phase of instruction involves reviewing contractions in order to activate background knowledge and/or introduce the topic of reduced forms. This is achieved by means of listening activities (e.g. determining whether or not a given utterance contains (un)reduced forms, choosing the right contraction from different possibilities). The next step consists in teaching the reduced forms of modal/auxiliary verb–pronoun combinations that occur in yes/no questions. The students first receive a list of frequently used combinations in both their reduced and unreduced state. The instructor then models pronunciations for both types and the students imitate them. Next, recognition and/or fill-in exercises are used in order to train the learners' perception and production. In the last phase of instruction, reductions are examined in wh-questions. The exercises used in this stage include minimal pair exercises, cloze dialogues, information gap, and guided conversation. Cahill's approach seeks to help EFL learners to move from controlled to automatic processing of reductions. Although his method was designed for Japanese EFL learners, it should also be of interest to teachers in other teaching contexts.
In chapter 9, Kevin Varden explores ways of using the free phonetic software package WASP (▪▪▪▪) for teaching the reduction patterns of spoken (North American) English. The chapter starts with a discussion of WASP displays (waveforms, pitch traces, spectrograms) as well as of the different types sounds in English. Then, Varden sets out to investigate several types of reductions (schwa reduction, flapping, coalescent assimilation, elision), and shows how these patterns can be taught to non-native speakers with the aid of WASP spectrograms. Although the outlined approach requires a considerable amount of phonetic knowledge, computer technology can in my opinion be a valuable aid in making non-native speakers aware of their speech as well as of the way(s) in which it may differ from the target language norm (i.e. phonological consciousness raising).
The fourth section of the volume (‘How should connected speech be taught in Japanese?’) examines CS in Japanese and explores ways of integrating CS into the JSL syllabus. In chapter 10, Nobuko Hasegawa discusses sound changes in Japanese by examining casual speech processes in comparison with fast speech processes. It is shown that, while phonological changes in fast speech (e.g. ‘high vowel devoicing’) are triggered by the rate of speech, sound changes in casual speech (e.g. ‘less sonorant vowel deletion’, ‘phrase final reduction’) are determined by sociolinguistic (e.g. politeness) and morphosyntactic factors.
In chapter 11, Takado Toda first discusses patterns of elision, contraction, assimilation, and transition in Japanese CS, and compares them to their English and French counterparts. She then outlines a three-step approach to CS teaching in which she combines listening activities with an inductive, form-focussed teaching methodology so that the learners can discover by themselves the pragmatic and/or (socio)linguistic constraints on Japanese CS. Toda's approach constitutes a useful supplement to Cahill's (see chapter 8) as the latter method does not pay attention to the (socio)linguistic constraints on CS phenomena in English.
In chapter 12, Takako Sakai & Yu Igashima also advocate explicit form-focussed instruction in teaching sound changes in Japanese CS, and propose to combine such an approach with the use of computer technology. They then set out to exemplify their method with CALL (computer-assisted language learning) materials which they have developed with a view on teaching ‘moraic syllable’, ‘word coalescing’, and ‘vowel devoicing’ as these phonological changes are particularly difficult for non-native speakers (p. 209). They finally report on an evaluation study showing that the learners reacted positively to the CALL materials but also that the use of computerized exercises lead to significant improvements in the learners' perception of the sound changes under investigation.
Chapter 13, by Yukari Hirata, investigates (non-)native speakers' perception of minute length contrasts in Japanese. It is shown that non-native speakers' ability to perceive such contrasts is lower when words are presented in CS than in isolation. Hirata ascribes this finding to the fact that the duration of Japanese vowels varies with speaking rate, and that there is significant overlap between short and long vowel categories at different rates. Studying length contrasts in citation form only may therefore not be pedagogically optimal. In this respect, non-native speakers who have been taught length contrasts in CS appear to have little difficulties identifying these contrasts in isolation, while learners who have only been confronted with the citation form of words are less able to generalize their ability to CS (p. 242). According to Hirata, these results contradict the intuition that minute sound contrasts should be taught in isolation, and emphasize the need to use CS contexts in pronunciation teaching.
The papers in the third and the fourth sections provide useful suggestions for the teaching of CS in English and Japanese. In my opinion, some proposals can also be used for teaching CS in other languages and will therefore be of interest to instructors in other teaching contexts. It should be noted that the chapters in the two sections focus almost exclusively on techniques to train the second/foreign language learners' perception of CS phenomena (i.e. receptive approach), thereby neglecting the productive side of CS instruction (i.e. the actual use of CS phenomena in speech). In the language classroom, however, both aspects deserve equal attention. Finally, one may also regret that in a discussion of pedagogical issues relating to CS instruction little attention is paid to the place of the mother tongue in CS teaching (e.g. in the form of a contrastive approach) as well as to its role (i.e. L1 transfer) in the second/foreign language learners' production and perception of CS phenomena in the target language (see, however, Takado Toda's paper in chapter 11, p. 201).
While the papers in these two sections of the volume explore ways of integrating CS into the second/foreign language classroom, the final section (‘How should connected speech be tested?’) deals with the evaluation of CS instruction. To this end, chapter 14, by Brown & Kondo-Brown, offers an in-depth discussion of a wide range of techniques that can be used to assess the second/foreign language learners' mastery of CS phenomena in the target language. These include both well-known teaching activities (e.g. cloze exercises, various types of dictations, listening activities, communicative tests) and tests used in scientific research on CS (e.g. Bowen's Integrative Grammar Test). The chapter also raises the question whether similar tests can be developed for other languages, and discusses some example tests for French and Japanese CS phenomena.
Perspectives on teaching connected speech to second language speakers is a valuable book for anyone interested in the teaching of English and Japanese as a second/foreign language. The volume reflects the current state of affairs in the research on connected speech phenomena, and is particularly successful in highlighting their importance for non-native speakers. The authors also show convincingly that non-native speakers do benefit from specific connected speech instruction, and provide teachers with a wide range of techniques to integrate connected speech into their classroom practice. As some questions are left unanswered (e.g. with respect to the occurrence and the type of constraints on connected speech phenomena in languages other than English and Japanese, to non-native speakers' production of connected speech phenomena, to the role of the mother tongue in the learners' speech and/or perception), the book opens up new directions for further research into connected speech. I recommend it to teachers, materials developers and researchers alike.
