1、Sound Patterns of Spoken EnglishChapter 1 begins by noting that most people arent aware of the sounds of language. This book is written by one of those annoying people who listen not to what others say, but to how they say it. I dedicate it to fellow sound anoraks and to others interested in spoken
2、language, with a hope that they will nd it useful.Sound Patterns of Spoken English Linda Shockey 2003 by Linda Shockey 350 Main Street, Malden, MA 02148-5018, USA 108 Cowley Road, Oxford OX4 1JF, UK 550 Swanston Street, Carlton South, Melbourne, Victoria 3053, Australia Kurfrstendamm 57, 10707 Berli
3、n, Germany The right of Linda Shockey to be identied as the Author of this Work has been asserted in accordance with the UK Copyright, Designs, and Patents Act 1988. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by an
4、y means, electronic, mechanical, photocopying, recording or otherwise, except as permitted by the UK Copyright, Designs, and Patents Act 1988, without the prior permission of the publisher. First published 2003 by Blackwell Publishing Ltd Library of Congress Cataloging-in-Publication Data Shockey, L
5、inda. Sound patterns of spoken English / Linda Shockey. p. cm. Includes bibliographical references (p. ) and index. ISBN 0-631-22045-3 (hardcover : alk. paper) ISBN 0-631-22046-1 (pbk. : alk. paper) 1. English language Phonology. 2. English language Spoken English. 3. English language Variation. 4.
6、Speech acts (Linguistics) 5. Conversation. I. Title. PE1133 .S47 2003 421.5 dc21 2002007301 A catalogue record for this title is available from the British Library. Set in 10/12.5pt by Graphicraft Limited, Hong Kong Printed and bound in the United Kingdom by MPG Books Ltd, Bodmin, Cornwall For furth
7、er information on Blackwell Publishing, visit our website: http:/Contents List of Figures and Tables ix Preface x 1 Setting the Stage 1 1.1 Phonetics or Phonology? 3 1.1.1 More mind than body (fossils again) 7 1.1.2 A 50/50 mixture 7 1.1.3 More body than mind 8 1.1.4 Functional phonology and percept
8、ion 9 1.1.5 Have we captured the meaning of phonology? 10 1.1.6 Inuence of phonology on phonetics 10 1.1.7Back to basics 11 1.2 Fast Speech? 11 1.3 Summary 13 2 Processes in Conversational English 14 2.1 The Vulnerability Hierarchy 14 2.1.1 Frequency 14 2.1.2 Discourse 16 2.1.3 Rate? 17 2.1.4 Member
9、ship in a linguistic unit 18 2.1.5 Phonetic/Phonological 18 2.1.6 Morphological 192.2 Reduction Processes in English 19 2.2.1 Varieties examined 19 2.3 Stress as a Conditioning Factor 20 2.3.1 Schwa absorption 22 2.3.2 Reduction of closure for obstruents 27 2.3.3 Tapping 29 2.3.4 Devoicing and voici
10、ng 30 2.4 Syllabic Conditioning Factors 32 2.4.1 Syllable shape 32 2.4.2 Onsets and codas 33 2.4.3 CVCV alternation 34 2.4.4 Syllable-nal adjustments 36 2.4.5 Syllable shape again 42 2.5 Other Processes 42 2.5.1 -reduction 43 2.5.2 h-dropping 44 2.5.3 Palatalization 44 2.6 Icons 46 2.7Weak Forms? 46
11、 2.8 Combinations of these Processes 48 3 Attempts at Phonological Explanation 49 3.1 Past Work on Conversational Phonology 49 3.2 Natural Phonology 52 3.3 Variable Rules 53 3.4 More on Rule Order 54 3.5 Attempts in the 1990s 56 3.5.1 Autosegmental 56 3.5.2 Metrical 58 3.5.3 Articulatory 58 3.5.4 Un
12、derspecication 59 3.5.5 Firthian prosodics 60 3.5.6 Optimality theory 61 3.5.7A synthesist 64 vi Contents3.6 And into the New Millennium 67 3.6.1 Trace/Event theory 67 3.7Summary 7 1 4 Experimental Studies in Casual Speech 72 4.1 Production of Casual Speech 72 4.1.1 General production studies 72 4.1
13、.2 Production/Perception studies of particular processes 80 4.2 Perception of Casual Speech 89 4.2.1 Setting the stage 89 4.2.2 Phonology in speech perception 93 4.2.3 Other theories 104 4.3 Summary 109 5 Applications 111 5.1 Phonology 111 5.1.1 Writ small in English, writ large in other languages 1
14、11 5.1.2 Historical phonology 113 5.2 First and Second Language Acquisition 117 5.2.1 First language acquisition 117 5.2.2 Second language acquisition 119 5.3 Interacting with Computers 124 5.3.1 Speech synthesis 125 5.3.2 Speech recognition 125 5.4 Summary 126 Bibliography 127 Index 142 Contents vi
15、iFigures and Tables Figures 2.1 Map of Lodges research sites 21 3.1 t-glottalling in several accents 65 4.1 Citation-form and casual alveolar consonants in both citation form and casual speech 79 Tables 2.1 Factors inuencing casual speech reduction 15 4.1 Listeners transcriptions of gated utterances
16、 101Preface This is not an introductory book: to get the most from it, a reader should have studied some linguistics and should therefore know the basics of phonetics and phonology. There are numerous works where these basics are presented clearly and knowledgeably, and it would be an unneccessary d
17、uplication of effort (as well as an embarrassing display of hubris) to attempt a recapitulation of what is known. The following books (or others of a similar nature) should be assimilated before reading Sound Patterns of Spoken English: Clark, J. and Yallop, C., Introduction to Phonetics and Phonolo
18、gy, Blackwell, 1995. Ladefoged, P., Vowels and Consonants, Blackwell, 2000. Roca, I. and Johnson, W., A Course in Phonology, Blackwell, 1999. There are hundreds of other useful references included in the text of this book. A few of these which have formed my approach to the study of sounds (and to t
19、he authors of which I am greatly indebted) follow: Bailey, C.-J., New Ways of Analysing Variation in English, Georgetown University Press, 1973. Brown, G., Listening to Spoken English, Longman, 1977, 1996. Hooper, J., Natural Generative Phonology, Academic Press, 1976.Lehiste, I., Suprasegmentals, M
20、IT Press, 1970. Stampe, D., A Dissertation on Natural Phonology, Garland, 1979. In my opinion, these works show great insight into the study of spoken language. Preface xiSetting the Stage 1 1 Setting the Stage Most people speaking their native language do not notice either the sounds that they prod
21、uce or the sounds that they hear. They focus directly on the meaning of the input and output: the sounds serve as a channel for the information, but not as a focus in themselves (cf. Brown 1977: 45) This is obviously the most efcient way to communicate. If we were to allow a preoccupation with sound
22、s to get in the way of understanding, we would seriously handicap our interactions. One consequence of this opacity of the sound medium is that our notion of how we pronounce words and longer utterances can be very different from what we actually say. Take a sentence like And the suspicious cases we
23、re excluded. Whereas a speaker of English might well think they are saying: (a) ndvsvscpvskeszwvyksckludd what they may be producing is (b) vs:cpkessv w xsckludt This book will look how you get from (a) to (b). It deals with pronun- ciation as found in everyday speech i.e. normal pronunciation. Year
24、s of listening closely to English as spoken by people from a great variety of groups (age, sex, status, geographic origin, education) leads me to believe that there are some phonological differences2 Setting the Stage from citation form which occur in many types of spoken English. Further, these dif
25、ferences are very common within these varieties of English and fall into easily recognizable types which can be described using a small number of phonological processes, most of which can be seen to operate in English under other circumstances. I call these differences reductions (though this term i
26、s a loose one: sometimes characteristics are added or simply changed rather than lost). A citation form is the most formal pronunciation used by a particular person. It can be different for different people: for example, the most formal form of the word celery has three syllables for some people and
27、 two syllables for others. For the former group, the pronunciation csyli involves a reduction, for the latter group, it does not. csyli could, however, have been a reduced form in the history of the language of the two-syllable group, even if not within the lifetime of current speakers. That it is n
28、o longer a reduced form attests to its promotion: the word is pronounced in its reduced form so often that the reduced form becomes standard. I speak as if promotion occurs to individual lexical items rather than classes of items, because it can be shown that not all words which have a given structu
29、re will undergo reduction and promotion: raillery, for example, will presumably remain a three-syllable word for those who have only two in celery, perhaps because the former is an unusual word, perhaps because it has more internal structure than celery perhaps for other reasons. In general, the mor
30、e common an item is, the more likely it is to reduce, given that it contains ele- ments which are reduction-prone (see chapter 2). The idea of lexeme-specic phonology is not a new one: many phonologists and sociolinguists have worked under the assumption that phonological change over time occurs rst
31、 in a single word or small set of words, then spreads to a larger set what is known as lexical diffusion. (For an early treatment, see Wang, 1977.) The citation form is therefore not the same as a phonological underlying form: it must be pronounceable and will appear as such in a pronouncing diction
32、ary. Words like celery generally appear with both pronunciations cited above. Deciding what is a reduced form can hence be difcult, but there are few debatable cases in the material I present here: nearly everySetting the Stage 3 native speaker of English will agree that the word rst has a /t/ at th
33、e end in citation form, but virtually none of them will pronounce it under certain conditions. The material which I cover in this treatise overlaps the boundar- ies of several areas of study: sociolinguistics, for example, is inter- ested in which reductions are used most frequently by given groups
34、and what social forces spark them off. Lexicography may be inter- ested in reduced variants, but only in so far as they are found in words in isolation, whereas this work looks at reductions very much in terms of the stream of speech in which they occur. Rhetoricians or singing teachers may regard r
35、eductions as dangerous deviations from maximal intelligibility, and a similar attitude may be found in speech scientists attempting to do automatic speech recognition. This book recognizes reductions as a normal part of speech and further suggests that the forces which cause them in English are the
36、same forces which result in most-favoured output in others of the worlds languages. 1.1 Phonetics or Phonology? It has been demonstrated (Lieberman, 1970; Fowler and Housum, 1987; Fowler, 1988) that there is phonetic reduction in connected speech, especially in words which have once been focal but h
37、ave since passed to a lower information status: the rst time a word is used, its articulation is more precise and the resulting acoustic signal more distinct than in subsequent tokens of the same word. By phonetic I mean that the effect can be described in terms of of vocal tract inertia: since the
38、topic is known, it is not necessary to make the effort to achieve a maximal pronunciation after the rst token. We expect the same to happen in all languages, though there may be differences of degree. Phonetic effects are not the only ones which one nds in relaxed, connected speech: there are also l
39、anguage-specic reductions which occur in predictable environments and which appear to be con- trolled by cognitive mechanisms rather than by physical ones. These we term phonological reductions because they are part of the linguistic plan of a particular language. Sotillo (1997) has shown that4 Sett
40、ing the Stage these behave quite differently from the phonetic effects described above: whereas phonetic effects are sensitive to previous mention, phonological reductions are not. We speak here as if phonetics and phonology were distinct dis- ciplines, and some feel condent in assigning a given pho
41、nomenon to one or the other (Keating, 1988; Farnetani and Recasens, 1996). Both comprise the study of sounds, but can this study be divided into two neat sections? Phonology has meant different things to different people over the course of the history of linguistics. Looking at it logically, what ar
42、e possible meanings for the term, given that it has to mean some- thing more abstract than phonetics? (1) One could take the stance that phonology deals only with the relationship between sound units in a language (segmental and suprasegmental) and meaning (provided you are referring to lexical rath
43、er than indexical meaning). Truly phonological events would then involve exchanges of sound units which made a difference in meaning, either: (a) from meaning 1 to meaning 2 (e.g. pin/pan) or (b) from meaning 1 to non-meaning or vice versa (e.g. pan/pon). Phonetics would be everything else and would
44、 deal with how these units are realized: all variation, conditioned or unconditioned would then be phonetics. As far as I know, this does not corres- pond to a position ever taken by a real school of phonology, but is a logical possibility. (2) Phonology could be seen as the study of meaning-changin
45、g sound units and their representatives in different environments, regardless of whether they change the meaning, and with no con- straints on the relationship between the abstract phoneme and its representatives in speech: anything can change to anything else, as long as the change is regular/predi
46、ctable, that is, as long as the linkage to the underlying phonemic identity of each item is dis- coverable. This will allow one-to one, many-to-one, and one-to-many mappings between underlying components and surface components, as well as no mapping (in which an underlying component has no phonetic
47、realization).Setting the Stage 5 This type of phonology would look at the sound system of a language as an abstract code in which the identity of each element is determined entirely by its own original description and by its relationship to other elements. Fudge (1967) provides an early ex- ample of
48、 introducing phonological primes with no implicit phonetic content. Foleys point of view (1977) is not unlike this: his thesis is that phonological elements can be identied only through their partici- pation in phonological rules: As, for example, the elements of a psychological theory must be estab
49、lished without reduction to neurology or physiology, so too the elements of a phonological theory must be estab- lished by consideration of phonological processes, without reduction to the phonetic characteristics of the supercial elements. (p. 27) and Only when phonology frees itself from phonetic reductionism will it attain scientic status. Kelly and Local (1989) also take a position of this sort: We draw a strict distinction between phonology and phonetics. Pho- nol
