An Introduction To Psycholinguistics [PDF]

Citation preview

Contents



Longman Group UK Limited, Longman House, Burnt Mill, Harlow, Essex CM20 2JE, England and Associated Companies throughout the world.



Published in the United States of America by Longman Publishing, New York



Preface xi



0Longman Group UK Limited 1993



Acknowledgements



All rights reserved; no part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise without either the prior written permission of the Publishers or a licence permitting restricted copying in the United Kingdom issued by the Copyright Licensing Agency Ltd., 90 Tottenham Court Road, London W1P 9HE.



PART 1: FIRST LANGUAGE 1



First published 1993



1.



How children learn language 3 1.1 The development of speech production 3 1.2 Speech understanding and its importance 16 1.3 Parentese and Baby Talk 22 1.4 Imitation and correction 25 1.5 Discussion questions 27 Suggested readings 28



2.



Animals and language 30 2.1 Teaching language to the chimpanzee, gorilla



ISBN &582 05982.8 PPR British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library Library o f Congress Cataloging-in-Publication Data Steinberg, Danny D. Introduction to psycholinguistics/Danny D. Steinberg. p. cm. Includes bibliographical references and index. 1. Psycholinguistics. I. Tirle. P37.S77 1993 4011.9-dc20



a



,



and dolphin 30 i o ~the ~ wild 40 2.2 Animal c o m m u ~ ~ i c a t in 2.3 Animal comn~unicationand human language 42 2.4 Pygmy chimp reheats dcbatc 43 2.5 Conclusions regarding animals and languagc 45 2.6 Discussion questions 46 Suggested rcadings 47



I



92-38645 CIP



1 sct in Linotron Bembo 11/12pt by 8U Printed in Ilong Kong W 11 '0 1



xiii



I v 1



I



,



I



3.



Wild children and language 48 3.1 Legends, evil kings and cmpcrors 48 3.2 Victor: the Wild 13oy of Avcyron 50 3.3 Genic: raised in solitary confincrr~cnt 55 3.4 Isabcllc: confincmcnt with a mutc mother



59



viii



Contents



Contents



ix I



3.5 Helen: the famous deaf and blind girl 61 3.6 A critical age for first-language acquisition? 63 3.7 Discussion questions 65 Suggested readings 66 4.



Sign language, written language and the deaf 67 4.1 Soundless language 67 4.2 Language without speech 67 4.3 Gestures and signs 69 4.4 Sign languages 73 4.5 The sign language struggle in deaf education 4.6 The Oral Approach 82 4.7 The Written Language Approach 84 4.8 A parting note on deaf education 89 4.9 Discussion questions 90 Suggested readings 90



6.5 The psychological unreality of Chomsky's grammar 128 6.6 The anti-Mentalist skeletons in Chomsky's closet 129 6.7 Discussion questions 131 Suggested readings 131 7.



Language: from intelligence or innate ideas? 133 7.1 Where do language ideas come from? 133 7.2 Empiricist and Rationalist answers 135 7.3 Chomsky's Universal Grammar 137 7.4 Arguments re intelligence and Universal Grammar 139 7.5 Mentalism and Behaviourism contrasted 153 7.6 Discussion questions 155 Suggested readings 156



8.



Language, thought and culture 158 8.1 The arrest of the Sunday School teacher 158 8.2 Speech as the basis of thought 159 8.3 Arguments against this idea 160 8.4 Language as the basis of thought 163 8.5 Arguments against this idea 166 8.6 Wherc language does affect thought 169 8.7 Thought as the basis of language 170 8.8 Discussion questions 171 Suggested readings 173



9.



Language and the brain 174 9.1 Brain structurc and function 174 9.2 Hcmisphcric dominance and lateralization 177 9.3 Languagc areas and functioning 181 9.4 Brain n~aturation and critical age for learning language 184 9.5 Langilagc disorders 186 9.6 Mcthods of investigating brain and langi~agc 194 7 Mind and brain 197 9.8 1)iscussion qi~cstions 190 Suggested readings 200



80



PART 2: LANGUAGE AND MIND 93 5.



6.



Mental grammar 95 5.1 Grammar and psycholinguistics 95 5.2 Chomsky's competence and performance distinction 96 5.3 Chomsky's grammatical conceptions 97 5.4 Linguistic challenges to Chomsky's grammar 110 5.5 Discussion qucstions 1 13 Suggested rcadings 114 Sentence processing and psychological reality 116 6.1 Meaning, sound and syntax relations in Chomsky's grammar 116 6.2 Why Chomsky's grammar is not a performance model 117 6.3 Typcs of performancc modcls 119 6.4 Somc fcaturcs of scntcncc production and understanding 123



I



I



x



Contents



PART 3: SECOND LANGUAGE 201 Children vs adults in second-language acquisition 203 10.1 A common belief 203 10.2 Psychological factors affecting second-languagc learning 203 10.3 Social situations affecting second-language learning 210 10.4 Who is better? 213 10.5 Critical age 215 10.6 Discussion questions 217 Suggested readings 217



Preface



11. Second-language teaching 219 11.1 Dimensions and methods 219 11.2 Traditional methods 221 11.3 The Chomskyan revolution and contemporar) methods 228 11.4 Conclusions 238 11.5 Discussion questions 240 Suggested readings 241 12. Bilingualism and cognition 242 12.1 Varieties of bilinguals 242 12.2 Is bilingualism beneficial or detrimental? 243 12.3 Simultaneous and sequential learning situations 250 12.4 Transfer Effect of L1 on L2 learning 254 12.5 Discussion questions 258 Suggcstcd readings 258 Author lndex 260 Subject lndex 263



,



I



, , ,



i I



I



With this book I wish to introduce the interested reader t o Psycholinguistics, that field of inquiry where language is regarded from a psychological perspective. As such, Psycholinguistics inevitably intersects with psychology, linguistics, philosophy, education and other disciplines, as well. While this is an introductory book, the reader is nonetheless taken to the heart of matters and contemporary controversy. This is so whether the issues are of a theoretical or a practical nature. Prior knowledge of any discipline is not presupposcd. I have tricd my best to introduce readers to the essential concepts of the constituent subfields with a view to gaining knowlcdgc and understanding issucs so that rcadcrs may bc ablc to think for thcmselvcs and conduct enquiries on thcir own. Thc book is divided into threc parts: First Lanon11;1won't let go'; 'I an1 not n doctor'; "T'liis nor ice-crc;l~ii';



I '



!



16



How children learn language



First language 'Paul not tired'; 'I not hurt him'; 'Don't touch the fish'; 'Don't kick my box'



i,



I



The child has now a good idea of when 'do' must t inserted ('You didn't caught me', 'Don't touch the fish Don't kick my box') and when 'do' is not inserted ('I a1 not a doctor', 'Donna won't let go'). The child still makc errors but seems to grasp the basic notion that 'do' is nc added when there is a modal ('can', 'will': 'This can't stic [adhere?]', 'Donna won't let go') or when 'be' is the verb ( am not a doctor'). After this period, it is only a matter ( months before most of the problems in negative making a1 successfully dealt with. Bellugi and Klima found that the three children in the study (the same three who were in the Brown morpherr acquisition study) all took about 6 months to pass throug the three periods. Still, there were great individual di ferences as to when they began the negative productions ( the first period: one of the children began as early as 1' years of age while the others didn't begin until they we] around 2Y2 years. While such wide differences in speech production a1 typical of very young children, by 4 or 5 years of age it seerr that most differences level off. And, while there are st] passives and other complcx structures for children to maste by this age they arc able to produce most of the essenti. structures of their language (this is evidently a worldwic phenomenon, whatever the language). It is perhaps by 9 or 1 ycars of age that all of the structurcs of the language have bee acquired.



1.2 Speech understanding and its importance 1.2.1



Speech understanding, the basis of speech production



Wc have seen how childrcn develop thc ability to produc sentences. 1311t, wli;it is the sourcc of that ability? Conside such typic:ll childrc~l's uttcranccs as 'W;~ntbanana', 'Nc sleep', 'Why it's not working?' ;lnd 'Whcrc I can fin(



,



'



,



17



them?'. Since children are not born with the knowledge of any particular language, it is necessary that they be exposed to the language in order to learn it. It is further necessary that the speech to which children are exposed be related to objects, events and situations in the environment and to experiences in their minds. Children will not learn speech, if they are exposed only to speech sounds. Even if the child hears a spoken word a thousand times, e.g. 'dog', there is no way for the child to discover the meaning of the word unless some environmental clue is provided - in this example, a dog or a picture of a dog. Even abstract words (discussed later) must be learned in some such way. While the ability to utter speech sounds, such as 'dog', in appropriate situations (when a dog is prcscnt or is barking, for example), is a good indicator that the child knows the word, simply being able to utter the sound form is of no significance. This is like someone being able to repeat the Japanese word inu. Anyone can do this but unless that person knows the meaning of that word, which is 'dog', we would not say thc person knows that word. A parrot can learn to utter many words and sentcnccs but wc do not regard the parrot as having knowledge of language in any significant sense. What is critical is that the child know thc mcaning of what is uttered. Evcn if a child learns to imitate somc words, we would not say that thc child knows thosc words unless thc child demonstrates in somc way that hc or she knows the mcaning of the word. Wc can judgc this if thc child uses the word correctly, or if thc child responds appropriately in some bchavioural way, such as by looking, pointing or following somc command. Now, in order for a child to learn the meaning of thc sound form of a word, the child must first hear t h ~ word t spokcn by others. (The child cannot know bcforchand that onc object has the namc of 'mama' while anothcr has the namc of 'dog'.) At the sanic time that the word is spoken somc rclcvant cnvironmcntal cxpcricncc must occur (the appcnrancc of the child's mother, for cxan~plc).Thcsc being the ncccssary conditions for Ic;lrnil~g,it is clc;lr that the cl~ild n ~ u s tlearn to undcrsta~ldspcccll hcforc he or she is ablc to produce i t ( n ~ c ; l n i ~ ~ g f i ~Iltl yis) . necessarily thc case that speech undcrstnntlirig prcccdcs spccch production. Spccch



I I



,



18



First language



production, therefore, is dependent on speech understar ing and its development follows that of speech understar ing. Aside from the above considerations, there is empiri evidence that speech comprehension develops in advance speech production. Parents have always noted that child are able to respond appropriately to speech that is m complex than what they (the children) are able to say. Besi parental observations, findings from research studies wh were especially designed to compare understanding and p duction also demonstrate the primacy of understandi Huttenlocher, for example, studied four children and f o ~ that they were able to understand speech at a level well bey( that to which they had progressed in production. The child were able to respond appropriately to commands even w: these commands involved vocabulary and structures that t' had never used in their own speech. One young boy, example, responded appropriately (by pointing) to s distinctions as 'baby's diaper' versus 'your diaper', and 'bat bottle' versus 'your bottle'. (The 'baby' here referred to the boy's younger sister!) Similarly, in another study, Sachs and Truswell f o ~ that children who could say only single words ('k: 'smell', 'ball', 'truck', etc.) could understand speech st1 turcs composed of more than a single word. The chilc provided appropriate rcsponscs when given comma consisting of novel two-word syntactic combinations, s as 'Kiss ball' and 'Smell truck'. Thcrc is the observation, too, that some child completely skip over earlier stagcs of spccch productiolnlorc advanced stagcs. The physicist Einstein and 13ritish writer Carlyle arc reputed to have bccn slow starting to spcak, but when they did they spoke scntcnccs. Such a phcnomcnon is con~plctclyunderstand: when one considers that children first acquire knowlcdgc language through spccch u~ldcrstnnding and that dc occurs only with the expression of such k~lowlcdgc, productio11. Without the lal~guagc knowlcdgc ncqu~ through speech undcrstandi~~g, Einstein, C;lrlylc or any, clsc for th:lt matter would not 11;lvc bccn ablc to utter cvc word nlc:r~lingfillly,Ict nlonc ;I sclltcncc. Evidcncc with Il;~~~dicappcd childrc~11pcrliaps provi,



How children learn language



19



even stronger evidence. In m y own research, I have come across children who were mute since birth but yet, having normal hearing, were able to understand what anyone said to them. O n e such 3-year-old Japanese girl I studied could not utter a single word but was normal in all other respects. (Her other motor skills were so good she could even ride a bicycle.) O n being tested, she could respond appropriately to such commands (in Japanese) as 'Put the red paper under the table' and 'Bring me the little doll from the other room'. With the assistance of one of m y students, I was even ablc to teach this child to understand to read (understand the meaning of) nearly 100 Chinese characters. - Yet this little girl's achievement pales before that of Christopher Nolan (this case is further described in Chapter 8), a novelist of some renown in the English language. Since birth Nolan has had little control over the muscles of his body. Control is so poor that he cannot sit without support and even has trouble swallowing'food. Fortunately, Nolan's hearing was normal and so he learned to understand speech. Only many years later was he able to producc (by selecting letters using a pointer attached to his head) his first words. Like Nolan, there are a great many other people who arc born with little control of their body but who do learn to understand speech, although they may not be ablc to producc any without difficulty. Thus, after having considered the various aspects of the relationship of spccch understanding and speech production, we can conclude that language learning may occur without production but not without understanding. 1.2.2



Learning abstract words



When accluiring the n~canings of words, children soon ilndcrstand and producc somc that arc quite abstract. Words expressing fcclings (liungcr, pain, joy) and con~plcxideas (lie (untruth), honest, guess) arc just somc of those Icnrncd. Ijut, how arc they Icnrncd? It cannot be hy sinlplc ;~ssoci;~tion coch : lthe spccrh C,rm 'dog' with thc ol,jcct 'dog7 bcc;~osc fcclings ; ~ n diJc;ls arc not in tlic pllysicnl c.~lvirollnlc~lt for the child to ohscrvc, i l l the way that hc. or sllc c;111



I



20



I



i



First language



How



observe a dog. Experiences in the mind are of a diffe nature. What the child must d o in order to learn the meal of abstract words is to observe speech, along with situat and events in the physical environment and then relate the, experiences and processes in the mind. Such learning seen be so mysterious as to be almost magical, yet, the fact is children d o learn such words. Let us consider now just this might be done. For example, how might the words 'hungry' and 'I be learned? First the child must take note of when words are spoken by others and the situations in which occur. The child might cry and the mother might then 'Are you hungry?'. O r , the child might point to a ba and the mother might say, 'Do you want a banana? Are hungry?'. The mother might use the word 'hungry' bec she might estimate that it has been some time since the last ate. Considering 'hurt', perhaps the child falls ar bruised on the knee. The father might say, 'Poor girl. I it hurt?'. O r , on receiving a needle injection, the child c and the father says, 'It hurts, doesn't it? H o w about s ice-cream?'. It is up to the child to remember what words spoken ('hungry' and 'hurt') and to rclate them to partic feelings (hunger and pain) that the child has cxpcrienct the mind. After a number of such instanccs whcrc cc words and certain fcclings arc cxpcricnccd togctlicr, child will have cnough information to make a guc: which sound form relates to which fccling. When thc ( then ,expcrienccs further instances, the child can tcst wht hc or shc is correct, i.c. whcthcr the sound form of 'hun rclatcs to the fccling of hunger, and, the sound forn 'hurt' rclatcs to thc fccling of pain. Words like 'lie' (falsehood) and 'guess' will also 1 cnvironmcntal situations and mental states for the chi1 bring togctlicr. Tlicsc gcncrally will he more difficul identify than fcclings because they involve piirc (I fccling) ideas. The child will have to make infcrcnccs f complex situations in order to extract such ideas. cxanlplc, after telling his nlothcr sonlcthing that hc kn is ilntrilc ('I didn't hit her'), the littlc hoy's niothcr 111 say ; ~ ~ i g r i l 'Tliat's y, a lie!'. What the littlc boy must dc discovc.r the nlcnning for this s o ~ ~ nfor111 d of 'lie' is: rc



I I



(



I



21



that he has told a lie (he can review this by comparing what he said with what he knows to be the actual situation, and note this discrepancy - the discrepancy between saying and situation being the concept of lie), and then assign this concept to the sound form. Whether this hypothesis is accurate or not is something the child will test when he hears the word 'lie' again.



I



1.2.3



(



I



children learn language



I 1



1, I



1



Memory and language acquisition



Underlying all of the remarkable accomplishments of the child in language acquisition is one crucially important psychological factor, that o f memory. For, in the course o f learning to identify the words of the language, devising rules for their use, and relating speech to the environment and mind, the child utilizes a phenomenal memory capacity. The child must remember a multitude of particular words, phrases and sentences, along with the contexts (physical and mental) in which thcy occurred. Such data provide the basis for structural analyses. If, for cxamplc, children did not rcmcmber many of the words, phrases and scntcnccs that thcy heard, thcy would havc littlc basis for discovering abstract meanings and rules. The various syntactic structures that wcrc mcntioncd and discussed carlicr, negation for instance, rcquircs that the child rcn~cmbcrmany ncgativc scntcnces. If the child could not rcmcmbcr ncgativc scntcnccs that had bccn cxpcricnccd previously, thc child would havc nothing with which to compare a prcscntly occurring scntcncc, and, thus could not makc significant infcrcnccs as to its structiirc. Without a good nicnlory, language learning would not bc possible. Aside from tllc common observation that children often rcmcmbcr, word for word, stories which they arc told, childrcli also learn a host of idioms in phrasc and scntcncc form. Thcrc is n o rcason, tlicrcforc, not to believe that children also store in mcnlory n n~ultitudcof ordinary phrases and scntc~iccs,which call scrvc them for analysis later. Such a prodigious memory capacity, it is worth noting, is not uniquc to lan!:uagc. For, in Illany other arcas of lifc - in rcmcmbcri~igfnccs, ol?jccts, nlusic, past events and v;lst qua~ltiticsof knowledge in a variety of fields - the cxtclit of ;I child's nlclnory is simil;lrly rc~n;lrknblc.



I !



22



How children learn language



First language



observed in many language communities. People who wish to communicate will naturally use speech that is at a linguistic level they think the hearer will understand. The characteristics of Parentese evidently are ones which serve to make the acquisition of language understanding and production easier for a learner. This is not to say that if parents did not use Parentese, their children would not learn language. Undoubtedly they would learn language anyway. But, given the obvious facilitating nature of Parentese and the way it naturally arises, it may well be that children who receive such language input learn to understand speech faster than children who do not.



1.3 Parentese and Baby Talk



1.3.1



I I



I



i



Parentese



Parentese (a term which I use to replace the oft-u 'Motherese') is used to refer to the sort of speech t children receive when they are young. The speech wlparents and others use in talking to children has a numbe distinctive characteristics. For example, parents' gener talk to their children about what is happening in immediate environment and not about abstract or rew objects and events. A sentence like 'The dog wants wa rather than 'I might start reading that psycholinguis book tomorrow' is what a 1- or 2-year-old is likely to hc Also, sentences tend to be short and the structures sim] e.g. 'The dog wants water' rather than 'The dog which been running a lot wants to drink some water'. Vocabul: too, tends to be simple, 'see' rather than 'notice' and 'h; rather than 'difficult'. When sentences are spoken children, the speech tends to be slower, the pitch higher , more pauses inserted than would occur if the same senten were spoken to mature speakers. Also, more words given stress and emphasis. Such exaggerations undoubte serve to highlight and to focus the child's attention important sentence constituents. (More on this on page 1~ Parents, too, use granlmatical speech when talking their children. Research by Ncwport, Labov and 0th show that few of the sentences arc ungrammatical in natL Such consistency undoubtedly is useful to the child whc searching to discover thc structures which undcrlic sf tcnccs. lncidcntally, thc findings that scntcnces are mail grammatical lays to rcst Chomsky's mistakcn claim that child rcccivcs scntcnccs which arc largcly ~ungran~matit 'dcgcncratc' being the term he uscs. I t is interesting that not only adults, but childrcn, tc tend to use I'arcntcsc when talking with younger childrt It has been found, for example, that Cycar-olds produc simplified speech w11c11 talking to Zycar-olds but I to adults (some of the 4-year-olds even did not ha youngcr siblings). Although n ~ u c hof this rcscarch has bc done with English, thcrc is good reason to suppose tl si~ch;I phenomenon is ul~ivcrs,llsince the s;llnc has be



1



23



t



Baby Talk is different from Parentese. While Parentese uses regular vocabulary and syntax, Baby Talk involves the use of vocabulary and syntax that is overly simplified and reduced. Curious though, from a psycholinguistic view, is the fact that most of the features which Baby Talk adopts are thosc which have their basis in the early speech of childrcn. Parents and others evidently believe that thosc features, when rcintroduccd back to the child, scrvc to foster communication. Most Baby Talk involvcs modifications in vocabulary. Therc arc already cstablishcd words likc 'bow-wow' (dog), 'pec-pce' (urinc) and 'choo-choo' (train) in English and, it1 Japancsc, 'wan-wan' (dog: thc standard word for which is inu), 'shcc-shcc' (urinc) and 'bu-bu' (car: the standard words for which arc jidosl.ta or kurumn). From such cxamplcs, wc can scc that thc main sound structurc of such words tcnds to d a Consonant Vowcl syllable unit which be d o m i ~ ~ a t cby is rcpcatcd (duplicated). Another construction principle for mnny 13aby Talk words, is that they rcprcscnt (somewhat) the sounds which varioi~sthings n1;lkc. Thus, 'bow-wow' ;lnJ 'wan-wan' arc simulations of the b:irking of dogs, 'bil-bu' is supposed to be the sound 111ndc by a car c n g i ~ ~ c: ~, n d'cIioo-c11oo' tlic sound madc by ;I tr;~in.That such ;I sound :IS'choo-choo' is Incant to ;lppro?cin~atc to 11ow 1;irgcly extinct steam loconiotivcs hothcrs 11c.ithcr parent nor c.hild. I-lcrc thc word h;ls hcco~nc.part of stnnd:lrd 1l;lby Talk voc.;lbulnry.



+



I



1



4



I



24



First language



Besides standard Baby Talk vocabulary ('standard' the sense that the item has already been coined and adop by others), it is not uncommon for a family to create . use its own words, words which are not used outside of family. Often these words derive from mispronoun words which their child produces. For example, attempting to imitate the word 'vomit', one child I kn said 'vompo'. After that the parents used 'vompo' insteac vomit in talking to the child. Sometimes a- few s vocabulary items might be retained by parents for sentim tal reasons. In English Baby Talk, it might be mentioned passing, it is common to add the 'iy' sound to words end in a consonant, e.g. 'birdy' for bird, 'horsie' for h o ~ 'kitty' for kitten. This provides the vowel for completion of the paradigmatic Consonant + Vosyllable. Since the 'iy' suffix also serves a diminutive ; affectionate function in English, this may also serve promote its usage. Syntax plays a less prominent role in Baby Talk tl does vocabulary. Parents seem only occasionally to Baby Talk syntax. When they do, their utterances strikingly similar to those in the children's telegraphic st. of speech production. A mother might say, for exam1 something like 'Mommy givc Tony banana' instead of syntactically proper 'I will givc you a banana'. In such utterance, neither the modal 'will' nor the article 'a' has bc included. And, the proper names 'Mommy and 'Tony' h; been substituted for the personal pronouns 'I' and 'yo Ccrtainly, fixcd proper narncs arc easier for the young ch to understand than are items involving shifting speaker-hca rclations. It is latcr that the child learns to cope with I coinplcxitics of 'I' and 'you'. Such proper name substit tions, it should be noted, also occur in Parcntcsc and tl-, arc not solcly fcatures of Baby Talk. Whcthcr Baby Talk should or should not be used sometimes a conccrn of parents, with intensity of concc varying grcatly from country to country. In Canada and t US, there sccins to be little such concern, whilc in Jap thcrc sccms to be a lot. Since 13aby Talk is a transitic phase, in that it is not continued for very long (childra thcit~sclvcswill soon replace it rathcr than being rcgardcd



How



children learn language



25



babies b y older children), and since Baby Talk (like Parentese) appears to be a universally occurring natural phenomenon, it would seem to me that Baby Talk could be beneficial t o some small degree for the child in learning language. Certainly, there is no good reason t o think of it as being harmful. That most parents (and grandparents) derive special enjoyment from using Baby Talk with their children m i g h t well serve to reinforce social solidarity.



1.4 Imitation and correction 1.4.1 The role of imitation



It was once widely believed by theorists that children acquire language entirely through imitation, i.e. by copying the speech that they hear. It is undoubtedly true that children d o imitate a great deal of what they hear. They d o learn to say such words and phrases as 'dog', 'Papa', 'run', 'happy', 'no', 'Why not?', 'bread and butter', 'Not now', etc. They do imitate the intonation pattcrns and sounds of thcir language, and thcy d o tend to approximate the proper ordcr of words in a scntcncc. O n the other hand, whilc somc languagc lcarning docs involvc imitation, this principlc is inadequatc to cxplain thc fundamentals underlying thc acquisition of language. Bccausc imitation involvcs thc rcproduction of speech, it thcrcforc cannot cxplain how spccch is undcrstood, thc knowlcdgc of which is thc basis of spccch production. N o r can it cxplain thc child's acquisition of i~lorphcmcssuch 2s thc I'lural or I'ast Tcnsc, or structural nlanipulations such as thosc of Ncgation and Question. T o illustratc how in~itationcannot account for rulcs, let 11s consider, in somc detail, crrors that childrcn makc. English children c o ~ ~ ~ m o producc nly words like 'mouses', 'shccps', 'gooses', 'gocd', 'comcd', 'fallcd' and 'hrcakcd'. Why d o cl~ildrcnproducc such words? C:lcarly, this cannot be duc to inlitntion since children gcncrally d o not hear pcoplc say such words. T h c most satisfnctory answer is that children have for~nulatcdrulcs in thcir nlinds, and c o ~ ~ s t r u c t such words o n the llasis of these rules. The s;lmc must be



'I I



'!



26



First



language



How



said for children saying things like 'No heavy', 'No tho shining', 'When we can go?' and 'He is doing what?'. these are not utterances made by adults, and since regularly appear, they cannot be ones that children imitated. They must be ones created on the basis of which the children have devised. While exceptions to rules must be learned, such pluralizing nouns ('mice' and not 'mouses') and in m: the Past Tense ('went' and not 'goed'), and while fu aspects regarding negation and question-making mun developed, the errors in themselves provide strong evic that rule-learning (Regular Plural, Regular Past) has t place. Rules, by their very nature, cannot be imit Rather, they are abstract constructions o f the mind. I do not appear in the physical environment - only sy sounds representing words and sequences of words do. on the basis of these observables that abstract and observable principles and rules are created in the mind then applied later in production. Thus, while imitation play an important role in language acquisition, it is a lin one - limited to certain aspects of speech production 1.4.2



The role of correction



Like imitation, the role of correction in language acquisi has been widely misconceived. Correction is iiot important factor in that process. While it used to be tho1 that correcting children's speech is essential for irnprc mcnt, research has shown that such is not thc casc. In ac fact, parcnts pay little attention to thc grammatical corr ncss of thcir children's spccch. Rather than correcting child's grammar, parcnts arc more intcrcstcd in respond to: the truth va114e of what is said (thcy will scold a child v says 'I didn't docd it' when he or shc is lying), the sc clpproprintnrc.ss of what is said (a child who says to a visit aunt, 'lladdy no like you' will be givcn a talking to later not on syntax) or t l ~ rcli~r~clri~c.ss o f what tllc child says ( parent will praise the child). When parcnts do attempt to corrcct tllcir childre spcccli, the rcsults arc oftCli fruitless ;111d frustr;ltil Consider the f o l l o w i ~ ~exchange g 1)c.twcc.n ;I nlothcr a11d s (notcd hy McNcill): I



children learn language



27



Son: Nobody don't like me. Mother: Nobody likes m e . I



T h e above sequence is repeated by mother and son seven more times. Mother (in desperation): N o w , listen carefully. N o b o d y likes m e . Son: Oh! Nobody don't likes m e . While some progress was achieved (the son added 's' t o 'like'), the major concern of the mother, the improper occurrence of the auxiliary 'do', was not noticed by the child as needing correction. Undoubtedly, there are instances where a parent's correction does result in improvement. Still they are not great in number. In any case what serves typically as correction is the mere repetition of the child's utterance in corrected form. T h e child is really givcn no direct clue as t o exactly what is wrong with the utterance that he o r she has produced. And, to give the child a direct explanation (even if a parent were able to formulate the problem) would often be absurd. The mother in the example above could hardly say to the child, ' N o w , Johnny, you don't add "do" in that sentence to make it ncgativc because the scntcncc is already ncgativc; the word ''nobody" is the negative of "so1ncbody". ' Children naturally corrcct thcir own mistakes over time, without the intervention of othcrs. 1.5 Discussion questions



Which is primary: spccch production o r spccch undcrstanding? Why? Elow is it possible for childrcn to begin by speaking whole sentences without going t h r o i ~ g lthe ~ one- ; ~ n d two-word production stagcs? Why d o cliildrcn product, 'tclcgr;lpliic.' spccc.li? Why arc sonic niorpllcliics, such ;IS tllc I'osscssivc~ ;IIIJ the I1:lst, Ic;~rncdfaster th;r11 otllcrs, such ;IS thc' 'l'llircl I'crso~l S i ~ ~ g t i land ; ~ r tlic Auxili;lry 'hc'? What is tllc soulid rule, th;lt governs thc 1 1 s of ~ :ldiiillg



I



28



6. 7.



8.



9. 10.



11. 12.



First language



's', '2' or 'iz' to make regular plurals, possession and third person? Hint:Consider whether the final (last) sound in word to which the suffix is to be added is a conson or a vowel, and what type of consonant it is. 1 might use the following words as examples: crc crows; Bob, Bob's; cat, cats; Chris, Chris'; pic pieces; judge, judges. A child says, ' I no want some candy.' What must child still learn in order to make a proper negat sentence? How might a child learn the meaning of 'idea' as 'That's a good idea!'? Do you know some words in Baby Talk of a langu other than English? Generally, what is the form of sound structure of these words? Does it agree u what has been said in the chapter? Would you recommend the use of Parentese or B, Talk in speaking to a child? What evidence is there that children learn rules wl they learn language? In your discussion present d concerning one morphcrne, such as Plural or Past, : also one sentence structure rule, such as negation. Can rules be learned by imitation? How might a child gct rid of errors without be. corrected by others?



Suggested readings I3rown, R. (1973) A First Lar;qrragc: The Early Staps. Cambrid) Mass.: Harvard Univcrsity I'rcss. Dc Villicrs, J. G. and Ilc Villicrs, P. A. (1978) I,anGqrrl Arqrrisition. Cambridge, Mass.: Harvard Univcrsity I'rcss. I)ulay, H., I3iirt, M. and Ic.rt 'T'crr:~



31



Vicki: the speaking chimp



T h e first known comprehensive scientific attempt to teach language to our nearest evolutionary relations was carried out in the 1940s by a husband and wife team o f psychologists, Cathy and Keith Hayes. T h e Hayes raised a baby female chimp, Vicki, along with their o w n baby son, Donald, with the hope that Vicki would learn speech as Donald would. Vicki was treated as a full member o f the family; she ate her meals at the table, played games at home and went. o n outings. .However, despite all their efforts, after three years Vicki had only learned t o utter four words: 'mama', 'papa', 'up' and 'cup', and these were so poorly pronounced they were hard to understand. Yet during the same amount of time, Donald had become fluent in the language. In the face o f such disappointing results, the Hayes felt obliged to abandon the project. 2.1.2 Washoe: the signing chimp In 1966, another husband and wife team o f psychologists, Allen and Beatrice Gardner, attempted to teach sign language to a baby chimp, a female they called Washoe (rhymes with 'show'). They reasoned that any attempt to tcach chimps to speak was doomed to failure because of the simple fact that chimps d o not possess the necessary vocal apparatus for human spccch. Vicki's failure t o learn to talk could plausibly bc said to bc duc to a simple physiological failurc arid not a mental onc. Since chimps are so adcpt at using thcir hands, the Gardncrs conccivcd o f the idea o f tcaching them a modified form o f Amcrican Sign Languagc (ASL), a language used by the hearing-impaircd in the Unitcd Statcs. It might bc noted hcrc that ASL is a gcnuinc and conlplctc langliagc, as arc many othcr sign languages, such as 13ritish Sign Languagc and French Sign Language. It has a vocabulary and a syntax just as an ordinary speech-based lal~guagedocs; anything that can be said in English can be cxprcsscd in sig11 I a ~ ~ g i ~ a M g eO. W C V Cit~ differs , from linear sl>ccch in that it involves three dinicnsions with hci;ll and othcr body movcnlcnts as well as hand movcnlcllts. A wllolc word call bc cxprcsscd by a hand cot~figurntion.ASL is not h;lscd o n fi11gc.r spclli~lj:, which is ;I Icttcr-hy-letter



I



I



32



1



', 1



\



First language



system, nor is its grammar very similar to Engl grammar. (For a more detailed discussion o f sign langua see Chapter 4.) O n e o f Washoe7s early signs was 'open', which expressed by a throwing out o f the arms. After about fi years with the Gardners, Washoe learned a vocabulary about 130 signs and, according to the Gardners, displa. two- and three-word utterances, such as, 'Go sweet', wl she wanted to be taken to the raspberry bushes, and '0; food drink', when she wanted something out o f the refrig ator. T h e two- or three-word length of utterance is simila that produced by human children around the age of 2 yea The Gardners givc the impression that the language the signing ape and that o f the signing child is very sim and that differences can only be found through cl analysis. In one study, they go so far as to claim the apt be superior to the child. However, as Premack and o t researchers have noted, after three years Washoe's achie ment never advanced beyond its very elementary level. contrast, by the age o f 3, ordinary children have lean thousands o f words and construct sentences o n the basis an abstract syntax, including the making o f such coml cated negatives as, 'I don't know his name', 'Paul can't h: one' and 'I am not a doctor'. (Whether o r not to add 'do' a ncgative sentence or where a ncgative rnarkcr likc 'not to bc placed is not a simple matter.) Hearing-impail childrcn w h o have learned ASL from infancy also acquir~ similarly high lcvcl of linguistic knowledge. Ilcspite the Gardncrs' claims that Washoc's cal acquisition of signs is analogous to the development language in human childrcn, the truth of the matter is tt Washoc ncvcr progressed beyond the linguistic lcvcl of t average 1 V2- or Zycar-old child. The many years o f trainii and exposure to sign language which Washoc expcriciic~ could not advance her beyond thc most clcmcntary lcvcl human achicvc~iicnt. 2.1.3



Son of Washoe and a chimp signing communit



Aftcr a nuinbcr of ycnrs, Waslioc was moved to a Licility i tlic state of Washirlgto~~ : ~ n dbcc;lmc part of tlic rcscnrc conciuctc.d hy Ilogcr and Ilchhy Foi~ts (anothcr couple:



Animals and language



33



w h o were working with a number of chimps. T h e Fouts regard their chimp subjects fondly and are currently involved in a project to establish a primate reserve in that state, where chimps would be given some o f the rudiments o f human culture, such as tools, for example, so that they could start their o w n community. T h e Fouts' primary interest is in looking at h o w language may o r may not develop in the social context o f that community. O n e of their major focuses is o n Washoe's 'adopted' son, Loulis, w h o , they say, is acquiring signs n o t only from Washoe but from the other chimps. T h e Fouts have witnessed Washoe demonstrating signs for Loulis and even helping to mould Loulis' hands into the proper configurations. They say, too, that they have observed three-way chimp conversations. For them, all of this demonstrates that language in the chimpanzee can advance, once given a start, without the intervention o f humans. To date, though, little further progress has been made in this regard. 2.1.4



Lana: the computer chimp



T h e Rumbaughs (another husband and wife team!) taught the chimp Lana a simple artificial language called Ycrkish (aftcr the Yerkes primate ccntrc). This language consisted o f scvcn colours and nine gcomctrical shapes which rcprcscntcd mainly objccts and actions. These items wcrc displayed on a large keyboard. Lana had to press certain keys in the right scqucnce to make requests, likc 'Plcasc machine givc milk' o r 'Please T i m givc ball'. Lana Icarncd hundrcds of sentences in this fashion. She had nanics for people, food, objccts and even a special phrase 'that-whichis' to name things she did not know the name of. Once she cvcn asked the trainer to lcavc the room (with a polite 'please'!) aftcr he had purposely niixcd 11p one of I ~ c r sentences to test her rcactioii. Sue Savage-IIu~iibaughlicrsclf believes that apes hnvc but a li~ilitcdIanguagc acquisition ability. She has cxprcsscd the opinion thnt pcrhnps the media raised hopes too higli for a ~ ~ i n i aInnguagc l rcscnrch, nlthough it might he snit1 thnt animal rcscarchcrs thc~nsclvcshave hnrdly hccn modcst o r cai1tion:lry in tlic clniins they 11:lvc mndc. Sllc says it is not



U C ([I



h



g



0



CJ



-z 2 $ s E>



Y*



2-,2 &&-C



b m



e cd 2 $ rn.50 E Y



g = $5 m



5



3



C 0 &



so,, -c . C .=



L 2 . z ~ : xu 3 ,'



" '"m CJ



3



ZZ2 0y Y



5: 5 3 ' a u =



g $,



- C ) e



V)



'JC



0



L "



Z Z C Y 3



-



.



rn I Y :



k-'JC



v l l



vl



2 4 L0. P" ' CU ' f . 2 w



% 5 . as 3 -Fi c E " 0 s LI .2~ Sa T,c sY-CJ



"



-2



4 Y



a> S : 2 > z P



vl



"



-



6 2 2 = .,." p L C :



& I 2:



gi " = "I:= %s



0



bl?



u



x"or 5 22 "5 " " z2 ;. :+.& g L.z- 5 - - " 2 .=;



0



c



24



Y



'"



a



h



.- = 52s 2 5 e -



a



5



%,::



40



Animals and language



First language



should be understood, providing, of course, that meaning of particular words are already known. Thus, a acquiring the notions of direct and indirect object, Akeakaresponded correctly on her first exposure to the sente 'person left frisbee fetch' (take the left frisbee to the persc Herman rejects the criticism that the dolphins are me, carrying out the same kind of stimulus-response-sha behaviour as do many marine park dolphins and whales. correctly points out that it could not be simple stimul response-shaped behaviour because the dolphins respc appropriately to specific commands which they have nc received before. It might be noted here that although the dolpl responded to commands, this cannot be taken as evidence they have learned the grammatical structure for comma This is because all dolphin 'sentences' had the m e a n i n ~ commands. In other words, the dolphin does not know syntactic difference between an imperative, a question declarative sentence form because there is nothing wl marks a sentence as being a command or not a command. sentences were of one form. There is, thcreforc, nothing command syntactic form for the dolphins to learn. Besides the lcarning of other syntactic structures, it , rcnlains for further rcscarch to demonstrate w h c ~ dolphins would bc able to express in production what t already have learned in tcrms of language understandl Givcn what we know about child languagc acquisition . the relationship of spccch u~~dcrstanding and spccch p duction, we might expect that the dolphins would dcvc the production capacity with little difficulty, provided tl arc give11 a convenient I I I C ~ I I S for physical expression. (



2.2 Animal communication in the wild



H u ~ n n nspeech is only one sm;lll part of the com~nunicat illvcntory of chirps, hisses, growls, s~lorts, whist1 gcsturc.s, lx~rksand bilzzcs wl~icliwc find i l l the rcst of t n~~inlnl kingdon). Aninlnls con11nunic;ltc tllrough ;I wi varicty of means, and though not cvcrythi~lgc o ~ n i n gfro cini~nalsis c o n l ~ ~ l u ~ i i c ; ~i l lt inilturc, vc :lt ;I Ixlsic lcvcl nnini; Il;lvc I I I ; ~ I I of ~ the s;lnlcb rc;lsons ns wc d o for pnssi~



41



information: to get food, to find a mate, t o warn and threaten others, etc. Many animals use sound signals, but many also use other modalities. Substances involving smell may be used as signals, as in the case of ants, which leave chemical trails for nestmates to follow in finding food. Visual signals may, for example, be used by dogs to threaten o r attract; the baring of teeth and tail-wagging serves to convey such intentions. T h e most complex type o f communication in the wild, unsurprisingly, involves that of the higher primates. Richard Seyfarth and Dorothy Cheney report that wild vervet monkeys make specific sounds that are more cpmplicated than hitherto believed. For example, these monkeys' alarm calls seem to be predator-specific. Thus, while one type of grunt indicates 'Beware, here comes an eagle!', another type of grunt indicates 'Beware, here comes a leopard!'. The calls of birds can announce, among other things, a readiness to mate, give alarm and defend territory. As researchers have determined, the bird calls and songs of particular spccics are largely inborn, although there are certain aspects of these calls and songs which will not develop unless the young bird is exposed to the voice of the adult bird. O f SOIIIC intcrest to the studcnt of languagc is the ability of certain spccics of birds, most notably the parrot and thc mynah bird, to imitate human spccch with amazing clarity. O n e bird can have a repertoire of a ~ ~ u n ~ bo fc r phrases and scntcnccs, such as, 'Hello, how arc you?' and 'Get away from thcrc!'. Rarely though docs the bird prodi~ccan utterance in a relevant contcxt. It can produce speech, but has no understanding of it, nor docs it ever produce uttcranccs other than thosc i t has been spccifically trained on. O f great interest, too, is the way honey bees inform other bccs of the presence of food in the vicinity. They use hot11 vision and toucl~.A bcc reporting back to the hive will go through n scrics of movements, n kind of dance, to tell other bees if n nectar sourcc is nc:lr o r distant, and, moreover, if i t is distant, jr~st how far ;~w:lyand i l l wll;lt clircction. Otllcr bccs can get the n1css;lgc not o ~ l l y by sccing thcsc movcmcnts but hy t o u c l l i ~ ~ gthat , is, hy ;ipproaching tllc scoi~thcc and fCclillg its m o v c ~ l l c ~ l with ts



.A 7:



-



2 2 ' I :



44



II i(



I



i



I



First language



Animals and language



words and learned to use grammatical rules. This started the whole animal and language controversy once again. an article in the book 'Langtlage' and Intelligence in Monk and Apes, UCLA psychologist Patricia Marks Greenfi and Sue Savage-Rumbaugh (of Lana fame - see abo claim that tiny Kanzi, when around 5 years old, lear: over a period of 5 months to use grammar equal to that 2-year-old human child. According to Greenfield, makes short, telegraphic sentences talking about relati ships between actions and objects, objects and locations, so forth' and has acquired grammatical rules that allow 1 to produce an infinite number of sentences and even ink his own symbols and use them consistently. All cc munication is done with hand gestures and s y n ~ b o l so computer keyboard. Kanzi is said to have a vocabulary about 250 printed words. Even complex relations arc said to be diffcrc~ltiatcd Kanzi. If Kanzi wcrc going to bitc his sistcr Mu playfully, he might sign 'bitc Mulika', but if his sistcr him, he would sign 'Mulika bitc'. Thus, order, important syntactic feature, would signal a differcncc meaning. Well, as might be cxpcctcd, neither I-Icrbcrt T e r ~ nor Noam CIio11~skyhave been much convinced by tlrcsults. Tcrracc says,



45



have used that capacity, until humans from some universities came along to show them how.



2.5



Conclusions regarding animals and language



(



s this bcforc and o n closcr scrutiny Wc'vc had c l a i ~ l ~like animals wcrc found t o be responding t o cuing (sul~lc,inndvcr gcsturcs by carcgivcrs) or rewards. Th;lt kind of rcsponsc is terribly pertinent to l a n g u q y .



Chomsky is less gcntlc in his criticis111. A cliimpnn using grammar 'is not a logical impossibility, but i t IS outlandish t1i;lt I don't know of any biologist who has tnl the possibility seriously'. I-lc pcrsu;lsivcly ;lrgucs that pygmy chimps rcnlly hnd the ;lbility to 11sc grnlnn.lnr nlillions of ycnrs past, tllcy surcly would hnvc iiscd it now hccnusc I;~ngii;lgcis so hio1ogic;llly ntlv;l~~tngcoiis. ' complctc,ly unknow11 hiologic;~llytl1;lt :Iny orgnnisnl c o ~ h;lvc n c-.lp;lcity tli;lt woulil hC highly v;lli~;~l>lc. for survi hilt wor~lcl 11ot 11sc it1. It clocs i~ldccclscclli rnthcr odd think that ;III nninl;~lwould Il;ivc c.volutio~~;lrily tlcvc.lop tllc, Ilighly c-o~ilplcsc;lp:lc.iry ti)r I : ~ I ~ ~ : ~ 1 ~I ;1Iwol~ld t~ c n



i



I



)



,,



T h e research with animals clearly shows that animals have only a rudimentary language ability, whether in the wild o r through training. What is puzzling and requires explanation is why their language ability is so low when their overall intellectual ability is so much higher. Apes exhibit, for example, intelligent complex behaviour regarding social organization, food acquisition and problem solving. Why, then, are they not ablc to learn more o f the sign language taught to them? After all, human children learn sign language in all o f its complexity. And w h y couldn't they at least learn to understand human speech given that they havc a hearing acuity which is as good as o r better than human hearing? After all, thcrc are human beings w h o arc born without thc ability for spccch production, yct they can learn to understand human language in all o f its complexity. And why, too, cannot otherwise intclligcnt chimps be ablc to dcal with numbers and d o elementary arithmetic? T h e chimp Ai at I. M. (1'978) Symbolization, languagc, and chimpanzcc: A thcorctical rccvaluation based o n initiill languagc acquisition process in four young 1'211-23 troglodytes. 13rcliri nrrd Larqrrnqc*, 6, 265-300. Tcrracc, M. S., I'ctitio, L. A., Sanders, ll. J. and 13cvcr, T. G. (1070) Can an ape crcatc a scntci~cc?Srirrrrc,, 206, 8'91-900.



Wild children and language



CHAPTER 3



Wild children and language



I It seems that people have always wondered about whet1 language is instinctive, something that is as natural humans as walking and smiling. They have also wonder too whether, even without experiencing language, childr would be able to produce it on their own. Even as recent the sixteenth century, we have such a brilliant thinker Montaigne saying,



I



\



I believe that a child brought up in completc solitude, far from intercourse (which would bc a difficult cxperiment to carry o would have somc kind of spccch to cxprcss his idca, for it is I likely that naturc would dcprivc us of this rccoursc whcn she I given it to many othcr animals. I



Some havc even thought that children who had not bc exposed to spccch, would speak in the original language humankind. Montaigne bclicvcd this too, and hc a1 bclicvcd that many animals have language. As the chaptc in this book on child language and 011 animals and langua testify, pcoplc arc still very miic11 interested in thcsc and othcr related questions as well, such as wl~cthcrthcrc is age beyond which a person would be iinablc to lea lan~iiqc.



3.1



Legends, evil kings and emperors



According to Icgcnd, some cxpcrimcnts have :~lrcadyhcc carried out to dctcrniinc what I;lngtiagc, if any, childrc~ would develop if they h;ld ncvcr hccn cxposcd to spcccll Thc ancient (heck 11istori:in I lcrodoti~swrote of ;I stor!



49



which he heard from priests in Egypt about one of their kings, Psamtik I, w h o reigned in the seventh century BC. According t o the tale (which was already more than 300 years old when Herodotus heard it), Psamtik gave t w o infants to a shepherd with the instruction that they be raised without anyone speaking in their presence. T h e king assumed that, without outside interference, the children would eventually speak the original human language. T h e children's first word (and perhaps only word!) was reported by the shepherd to have been becos, o r something sounding like it. After inquiring of his learned advisers as to what what language that word might be in, the king was told that such sounds meant 'bread' in the Phrygian language, a language then spoken in what is n o w central Turkey. Psamtik felt that hc had his answer regarding original languagc, although sceptics havc suggested that the sounds could havc come from the children imitating sheep o r goats! Akbar the Great, the Mogul Emperor of India in the sixteenth century, is also reported to have carricd out a similar experiment. However, in this instance, after years of confincmcnt, the infants did not speak at all. Likewise it is said that James IV of Scotland also conducted such an cxpcrimcnt with infants. When he heard their first uttcrings, the king declared that they wcrc in perfect Hebrew! After the passage of ccnturics and even millennia, thcrc is no way of knowing whether these monstrous cxpcrin ~ c n t swere rcally done in thc way that is reported or that the results claimed were what was rcally found. Whatcvcr the case they do reflect an overall h u n ~ a n fascination with Iangi~agc2nd the for111 that a supposed 'natural' language would take. Mcdiacval Europcan scholars, too, spcr~tmuch timc talking about and cvcn trying to reconstruct the langungc that tlicy bclicvcd was spoken by Adam and Eve. 'Natural n1:11i' was one o f the topics o f tllc Enlightcilmcnt, too, and from liousscaii's 'noble savage' to Edgar Iiicc 13iirroi1gl1s' 'T;lrzan' we 11;lvc bcc~idrawn t o the idca of ;I hiimn~ibeing untouched hy civilizntio~l.A l t h o i ~ g lcducntctl ~ pcoplc today do not bclicvc that ;I child deprived o f I:l~lgii;~gc fro111birth woiild start spc:lki~lgthe. originirl I;~~i~:iingc o f I i u ~ i l a n k i ~or ~d, ;lily Iangliagc a t a11 for th;lt ~n:lttc.r, tl1c.t-c, is grc;lt scientific intcrcst in origii1;ll I:lngu;igc(s) ,lnd i l l the cffccts of I;l~lgii:~gc dcprivatio~l.I )cprivatio~~ or isol;ltio~lcnscs may hC iihlc ro tc'll



50



First language



us something about the nature of human language, h o learned, and when it is best learned. Since ethical considerations deter scientists frorr ducting language deprivation experiments with chl scientists have been on the lookout for cases which naturally, so to speak, i.e. without their intervention as through peculiar circumstances o r the pervers human behaviour. Over the past few centuries thert been a number of reported cases of children raisg wolves, pigs, sheep and other animals. (A fasci collection of such cases is described in Malson's boo1 Children.) Then, too, there are childrcn who have appa survived on their own for years in the wild; g r o w i ~ even without the aid of animals. O n a different level, are the cascs of children who have been kept in confink or isolation by their parents o r others, and conseql were not exposed to language. But there arc also a number of childrcn who have bccn cared for by 1 parcnts, but who, bccausc of a physical disability, SL dcahcss, havc noncthclcss bccn dcprivcd of lang Studying thc dcaf, too, thcrcforc might also provide u: insight into certain psycholinguistic questions in m u c same way as docs the study of childrcn who havc grov in the wild or in isolation. Unfortunately, as far as the cascs involving childrcn arc conccrncd, most of the rcports arc not adc, for scrious scicntific analysis. Too oftcn thcy arc basc fragmentary data usually with no details other than such and such a child had been found in such and suc environment. The exact naturc o f their Ia11gi1agc \ found, and the cliildrcn's subscqucnt Iangiiagc dcvclopn were not properly studied, niost having lived ill an bcforc thcrc was widespread scientific psycl~ological li~~guistic intcrcst in the matters r;~isc.dhy tllcir condit I-1owcvCr, S O I I I ~cases / I ( I I J ( ~I ~ C C I Iwell ~ O C I I I ~ ~ and CI~~C~, the most important of thcsc th;it will bc prcsc.~itcd licrc 3.2 Victor: the Wild Boy of Aveyron S c i c ~ ~ t i i~~vcstig:ltio~i fi~ into tllc n1;lttcr of wilci c-llild illcrc;\scd dram;ltic:\lly in j;11111;1ryof t l l ~yc:lr 1800 w11e1



Wild children and language



51



boy was captured in the woods near the village of SaintSernin in the Aveyron district of France. H e appeared to be 11 o r 12 years old, was naked except for what was left of a tattered shirt, and he made n o sounds other than guttural animal-like noises. His general appearance and behaviour were typical of the wild men of popular legend and he seemed to have survived on his o w n for years in the wild. Probably he had been abandoned originally but at what age or by whom could not be ascertained. Attempts to trace his personal history failed and nothing could be uncovered of his life before his being discovered. Fortunately for the boy, the France o f 1800 was alive with a spirit of scientific inquiry and a sympathy for lost children. When Sicard, the noted director o f the Institute for Deaf-Mutes in Paris, heard of the boy, he made efforts and was successful in getting custody of him. Sicard's mission in life was the education of the deaf and he had already had considerable success with deaf children, many of whom had bccn discarded as retarded by the community. H e demonstrated that, to a significallt degree, many such childrcn could bc taught to communicate in sign languagc and could also learn to read and write. Sicard delighted in showing off the childrcn, for thcy proved thcmsclvcs quitc capable of cngagi~lg (through writing) in intclligcnt and oftcn witty banter with his scientific collcagucs. Sicard perceptively noted that thcrc sccmcd to be strong siinilaritics bctwccil this newly found wild child who had been dcprivcd of language by isolation and those cl~ildrcnwho had bccn dcprivcd of language and normal social co11t;lct bccausc of dcafi~css.He was cngcr, therefore, to apply to the Wild I3oy (as he was the11 called) sonic of the methods he Iiad dcvclopcd in educating the dcaf. It w;is not long, though, bcforc Sicard's optimism fi~cicdin the face of the lack of progress with tlic Wild I3oy of Avcyron. N o t only did the boy f i i l to learn nny Iangungc 1>iit liis bchnviour was cl~litcunlike th:~tof thc other childrcn in the i~lstitutc.After O I , I ~ a few n i o ~ ~ t hthe s , instltutc ~ s s ~ ~ ~ d a report stating that thcrc Ii;~dhcc~ln o progress with tllc hoy and that none could rcaso~~:\hly hc expcctcd. 'They rcgnrdcd hini ;is bci~lguntc;lch:~hlc;111d g;lvc LIP 011 Ilini. Forrull:lccly, Ilowcvcr, the boy's c'duc-;ltiol~w ; ~ stllc11 t;1kc11 ovcbr hy n~iotllcrcngcr c.duc;lror, tllc, c-rc.~tivc: I I I ~~~~~~~~:ltc~cI.



52



1 I



I



I



j/



I 1 I



First



Wild children and language



language



Jean-Marc Gaspard Itard. Itard set up an ambit. programme for the boy, with goals which included soci, well as language training. Itard's success with the clearly shows that the assessment made by Sicard and others at that institute had been quite incorrect. T h e Wild Boy was given the name Victor by Itard his education began with intense work that involvc variety o f games and activities which Itard designel socialize Victor and make him aware of the world arc him. These had a dramatic effect. Victor took pleasu long walks, taking baths, dressing himself and setting table. Where, at the time o f his capture, Victor had example, been virtually insensitive to temperature an( with his hands, he now insisted o n the bath water being right and having utensils when eating. While earlier he not reacted at all to the passing countrysidc durii carriage ride, he now took obvious pleasure in the cha of scenery. He enjoyed guessing games, cspccially o shcll game varicty, such as whcrc onc trics to follov quick movements o f tlic hand in which a chestnut is and thcn guesses under which of thrcc cups tlic chcstr hidden. H e cxprcsscd a widc rangc of emotions and dc through movcmcnt, or 'action language' as Itard callc His scnscs appcarcd normal, from his rcaction to the t around him. Spcccl~training with Victor, however, proved t vcry frustrating for Itard. I t centred around simply t r y i ~ get Victor to rcpcat sonlc words and spcccli sounds. TI: consistently failed to do, and ltard concluded that it ilnrcalistic to expect speech from an adolcsccnt w h o liac spent virtually his cntirc life in the wild. Ncvcrthi Victor could disti~lgtlishspeech sounds from other so in the cnviroiimcnt ;ind hc was cvcn ;iblc to diffcrcntiat~ sounds of ~lornlal spcccli fro111 the poorly pronou spcccli sounds nindc by tlic deaf children in the inst whcrc he now resided. Victor's spcccli tr;lining first rcsultcd ill his hcing to rcpcat the sound 'li', :~pparcntlyhis pcrso~l;llcontrac of the name Julie, the n;llllc of the dal1glltc.r of ;In ;issista~ tllc i n s t i t ~ ~ t c111 . additio~l,hc would rcpc:rt the ~>hr;isc IIicyl!' ('Oh i~lary.Incomplctc sign I a n g ~ ~ a g c ~ tyl7ic;llly found i l l dcvcloping cou~ltrics,altllougl~in cv



69



some developed nations, sign language may suffer from deficiencies. In Japan, for example, where the national government prohibits the teaching and use o f sign language in public schools, standardization and vocabulary are problems. (The rationale for this anti-sign position is considered in a later section.) Returning to the language criterion, not only can a fluent signer o f a complete sign language such as American Sign Language (ASL) sign whatever a speaker can say, but the signer communicates at the same speed as a speaker does. T h e speed at which signers produce sentences (more precisely the propositions which underlie sentences) in a sign conversation tends to be the same at which speakers produce sentences in a speech conversation. This occurs even though a signer, as does a speaker, has the ability to exceed this speed. There seems to be an optimum speed at which humans are able comfortably to process language information, whether that information is in the form o f speech o r sign. Bcforc considering the essentials o f sign language, it will bc uscful, and intcresting, to examine a related means o f communication that is used by hearing persons: gestures. Although gestures may bc cornplcx, they arc but collections o f signs and d o not form a true languagc; there is no grammar with which gcsturcs may be combincd to form thc cquivalcr~to f scntcnccs, cxcept of thc most rudimentary sort ('You comc', 'I hungry'). Ncvcrthclcss, they d o play an important part in communication both in conjunction with spccch and as an altcrnativc for spccch.



4.3



Gestures and signs



4.3.1 Gestures without speech I'coplc use n vnricty of body nlovcmcnts to convey mcssngcs. Most o f these nlovcmcnts, cnllccl g c s t i ~ r ~ s , mainly involve thc face and hands although the posture o f thc body is i ~ n p o r t a n t ;IS well. Wc irsc gcstIlres to coniniunic;~tc ;I variety o f types of mcssngcbs, ;IS, for cx:lmplc~:greetings (hcllo, goodl>yc - by lnoving the 11;11lds



70



First language



Sign language, written language and the deaf



and arms); requests and commands (come, go, stop moving the hands); insults (the sticking out of the ton by children, the raising o f the middle finger by adu answering (yes, no, I don't know - by moving the he evaluating (goodlperfect - by making a circle with thumb and index finger; and victorylsuccess - by ma1 the V sign); descriptions (tall, short, long - by use of hands and arms), referring (self, other, this one, that 01 by pointing) and scolding (a facial scowl). These are some o f the categories for which w e have gestures that be used for communication independently o f speech. Some gestures are almost universal, such as the moof the hand or arm towards the body to indicate 'cc (Americans make a sweeping motion of the arm from elbow towards thcir own body with the hand hcld vertical position while Japanesc havc thcir arm outstrct, horizontally but movc only their hand with the palm fa downwards), or in pointing to onc's body to indicate ' (Americans point to thcir chest whilc Chincsc point to I nose - to m y knowledge no c o n ~ n ~ u n i tpoints y bclow waist). Most gcsturcs, howcvcr, arc specific to cult1 linguistic or geographic areas. Thus Sri Lankans shake I head in a way which, whilc indicating 'ycs' or 'agrccm for them, indicates 'no' or 'disagrccn~cnt' for En! speakers. (I once was the adviser of an MA student fron Lanka, and cvcn though I knew that her shaking of the 1 indicated agreement with what I was saying, I ncvcr cc get used to it.) And, while the Japancsc place their in fingers sticking upwards on the sides of tllcir head indicate that someone is 'angry', a person from Frs visiting Japan might, in searching for a meaning, inter] the gcsturc as indicating 'cuckold', after tllc French I;111gu expression 'wearing of the horns'. Facial nlovcmcnts, in p;lrticular, arc ilscd cvcrywl to express a wide range of cnlotions :~ndfeelings. We do need actually to utter the words 'I :In1 . . . {hap surprised, disgusted, di.;nppoilltcd, cscitcd, ;Ingry, ctc wI1c11 we have i l l our non-vc.rIx11 rcpcrtoirc tllc :Imnz. flcsihility to roll our cyc.; in cxnspcratio~l, contract ( hrow i l l c o ~ ~ s t c r n ; ~ t and i o ~ i haughtily raisc our cycl>ro (from which, hy the way, wc get anot11c.r 11;lughty sort word, 's~lpcrciliour', nic;111ing, from Idatin, 'r;liscd c)



,



(



i



1 Ii 1



I



71



brow'). States o f aversion, confusion, attention, distress, love, annoyance, superiority, belligerence, doubt, stupidity, bewilderment, determination, and so on, can all be conveyed by various combinations o f facial expression, hand movement and body posture. In examining gestures, it becomes obvious that some gestures are more related to, o r suggest, the ideas that they are intended t o represent than are others. T h e hand and arm gesture for 'come', pointing to your o w n body for 'self or a smile gesture for 'friendliness', for example, have a certain closeness. These kinds o f gestures having a close relationship between gesture and meaning are called iconic gestures. These iconic gestures contrast with ones that are more abstract, such as the shaking o f hands to signify agreement in the closing o f a business deal. This would be a non-iconic gesture. It gives less clue as to meaning than the iconic gcsturc. Such being the case w e would expect iconic gcsturcs to be easier to acquire than non-iconic ones. Some gcsturcs, howcvcr, are not so easy to categorizc. Would you bc ablc to guess that a listener's noisy sucking in of breath whilc the eyes look upwards signals 'dccp consideration for what thc spcakcr has said' in certain segments o f Japancsc socicty? I was baffled when I first canlc upon it! Bcsidcs thc gcncral gcstures uscd in a culturc, thcrc are also rcstrictcd gcstures which are known and uscd by small groups. Tllcsc arc typically to bc found in spccializcd fields of work. Stock trading Wc havc all sccn pictures o f stocks being bought and sold on the floor o f an exchange. With thosc f ~ ~ r i o uhand s and fingcr signals, stocks arc named, prices quoted and dcals arc closed. i r At a racc-track in Britain, you might scc a nlan putting one right fingcr in his left car. Hc is not rclicving a11 itch but, as a bookmaker (a bet taker), 11c is indicating that the odds on a certain horse arc 6 to 4. Mltsir A symphony conductor often pi~llsthe pal111 back towards the body to request less v o l t ~ n ~fro111 c the orchestra. Sports Itcfcrccs and j t ~ d g c silsc elaborate hand and arm gcsturcs to indicate tllc state o f play and the assignn~cnto f points and pc~~altics. 1 1 i i o i If you wcrc an announcer prcscnting the



72



First language



news, and the person in charge drew his or her index fing across the throat, you would bring your talking to a clo: O n the other hand, if you were on a dark deserted street ir dangerous part of town with two rough-looking charactc approaching you, and one signalled the other with tl gesture while looking your way, you would start runnin 4.3.2



Gestures with speech



Despite the great number of gestures which are available 1 use, it is clear that most of the gesturing that people eng: in when they communicate are coordinated with speec While some of these gestures have an iconic sign functiol i.e. can be used by itself to convey a meaning - others not. A good example of the latter is beat, where one's ha or finger is kept in motion and is synchronized with wha person is saying. Beats are constant in form and do not change with I content of the sentences. In making beats, pcople will mc their hands up and down or back and forth. This tends to done in the periphery of gesture space, such as to the sic not in the central portion. The purpose of bcats, accordi to McNcill in his insightf~llanalysis of gesturcs, is basica to emphasize the discourse function of concurrent spcec Beats do not add to the content of a description or story t rather serve to emphasize the introduction of new char; ters, the setting of a scene, the O C C L I ~ ~ C I I C Cof SOIIIC eve1 and the likc. McNcill presents the following case cxalnp A person (A) has been shown a film and is asked to t 2 about it. A says that the charactcr in the film llns girlfriend, and as A says 'his girlfriend' he makes n bent. then says that her first name is Alicc and, as hc says 'Alicc he nlakcs anothcr bcat. A then goes on to say that 11 family 11amc is White and, as he says 'White', hc mnk, anothcr bcat. Thrcc bcats wcrc performed succcssivcly I this little bit of n;~rrntivc, o11c bcat per piccc of n c informatio~l.However, i t sllould be noted that many bcamay occilr with :I si~lglcS C I I ~ C I I C Cnnd that 11cw illfornlntio is not :llways involved. For c s ; l ~ ~ ~ pMcNcill lc, dcscrihcs on 5-yc;lr-old child saying, in response to being nskcd whn somctllil~gis, 'It's solnc.thi11gc.lsc.' As he was saying this th, child's hand rose lip and dowll thrcc timcs on the arnlrcst o



Sign language, written language and the deaf



I



'



I



,



;



, 4



1 I



the chair he was sitting in. Thus, even young children acquire quite early the gestures which accompany the speech of their language. The use of beat, however, is more pronounced in some cultures than in others. Italians and Jews, for example, seem t o d o it more than Japanese and Britons. O n e Jewish man talking t o another man might even tap the other on the stomach. (I doubt that a Japanese could even be trained to d o this!) Besides beats, people make another, perhaps more important type of gesture along with speech. This is the iconic or content gesture which, according to McNeill and his colleagues' research, occurs just once within each clause. Such gestures occupy the central gesture space and can add to or make more explicit some part of a description or a story line. Thus, for example, when people are asked to describe something they see and they utter sentences likc 'He is trying to go u p the inside of the drainpipe' and 'He is going up through the pipe this time', in both cases the speakers make an upward gesture, either with the finger or the hand. The gesture is made while that important portion of the sentence (italicized) is being uttered. Making 11ote of what pcople d o when thcy talk, such as their production of icon and beat gesturcs that thcy make, can be a very interesting pastime. You will be surprised at what you learn, if you can stiflc that urge to smile or laugh aloud.



4.4



)



73



4.4.1



Sign languages



Types of sign language



Sign Innguagcs use hand, facc or other body movc~ncntsin a three-dimc~lsional space as the physical mcnns of conlmunicntion. 1 r i 1 c i : I l y thcrc arc two types of sign Inngilagc and thcsc ciiffcr as to whcthcr or not the signs rcprcscnt ordinary (spccch-based) Iangu:~gcs.Thus, thcrc arc sign Inngungcs which rcprcscnt the words (tlirough signs) nnd their order ns they appcnr i l l ordin:lry Innguagcs, such ;IS Swedish, E~~glisli and Frcnch, and there arc sig11 I:lng~~;lgcs such ;IS Amcricnn Sign La11gu;1gcn11d Ijritish Sign Ln~lgu;~gc



74



First language



Sign language, written language and the deaf



which have their own words and grammatical systems fc the generation of sentences. I



4.4.2 Sign languages representing spelling or speech



,



Sign language based on ordinary language can be of t n different kinds. O n e such kind represents words by spellir them out in terms of individual signs, where each s i ~ represents a letter of the alphabet. Hand and fing configurations are used to indicate letters, such as making V with the index and middle fingers or an 0 with tl thumb and index finger. Thus, a word such as enough wou be signed letter by letter, e, n, o, u, g and h, followir English spelling. Words and entire sentences are commun cated in this letter-by-letter method. The ordcr of letters exactly the one that occurs in the writing of the ordinal language. There are both one- and two-hand systems of fing spelling (Figurc 4.1). The Americans and Swcdcs, fi example, use one hand, while the British use two. Uscrs



i



+



,



,



I



,



Figure 4.1



Fingcr spcllillg: 1 :111d2 I1:11ldcd



75



both systems can sign relatively quickly but both processes are rather laborious. The two-handed system is faster and provides more easily identifiable letters but it does not allow a hand free for other uses. Only a few deaf schools (like the Rochester School in America) rely wholly on finger spelling to express all communications. (The sentence 'The boy coughed' is expressed as the series o f individual letter signs: t, h, e, b, o, y, c, o, u, g, h, e, and d.) Signers of all systems, though, must learn to use finger spelling since most proper nouns like Manila, Caroline and Kensington, d o not have their own special individual signs. Such proper nouns must be finger spelled. - More popular than The Rochester Method of finger spelling is a kind of sign language which uses whole signs for each speech word or meaningful word part (morpheme). For 'coughed', for example, there would be one whole sign for 'cough', and another for the past tense. Seeing Essential English and Signing Exact English are typical of such sign systcms. Thcse language systems, which, for want of a name, I shall call Signing Ordinary Language (SOL), follow in signs the exact linear flow o f spoken words. Thus, 'I askcd John for the cards' would havc a sign for each English n~orphcmcin that sentence, signed in the same ordcr as the spoken sentence: I ask P A S T Jolzn +for + t h ~ card + P L U R A L . The word 'askcd' has a separate sign for the root 'ask' and a separate sign for the suffix marking past tense. Similarly, 'card' and a sign which marks the plural havc scparatc signs. These systcms arc relatively new, having been devised by educators in the past 40 years. A SOL system has certain important advantages for the Icarncr. Ijy learning it, not only will the person be ablc to coi~~municatc with other hcaring-impaired pcrso~ls (who k11ow the systcnl) but the Icarncr will have knowledge of the syntnx and vocnbulary of the ordinary language as well. The ordinary languagc would then not hnvc to be Icarncd as a remote second Iangtiagc (as far as vocabul;iry and syntax is concerned) by pcrsons whose nativc 1a11gu;1gc is an incicpcndcnt sign Iangi~agc. For the s:lnic reason, too, Ic;~rningto read will not he as difficult. In nddition, the SOL system has rhc advantage of bcing c:~sicrfor hearing pcrsons



+



+



+



+



76



First language



to learn than independent sign languages since S O L is base o n the grammar which hearing people already kno. through speech. This is particularly advantageous to tlparents o f deaf children w h o naturally want t o establish means for communicating with their children as quickly : possible. Most deaf children, it is worth noting in th regard, are born to hearing parents. Despite its many advantages, S O L suffers from 01 serious, perhaps fatal, disadvantage. Deaf signers general d o not like it. They find it m o r e cumbersome and le natural t o use than a sign language that has evolved throuf extensive use in the deaf community.



4.4.3



{



.



Sign language, written language and the deaf



77



same w a y as are spoken w o r d s s o as t o provide variation i n grammatical classes and meaning. 1



!



n



C



a. Present



5-4



Independent sign languages (SL)



L



GO



WIN



Some characteristics of S L T h e signs o f a sign language that is indcpcndcnt o f a spccc based languagc (hercaftcr, SL) can bc b r o k c ~ i d o w n in three basic componcnts: lzatld cor!fi,quratiorr, h o w thc hand formcd; place of articmlatiorl, whcrc tlic hand is formcd; ar movement, h o w thc hand moves. At the word level o f ; ordinary language, thcrc arc not only words which diff; coniplctcly in meaning from onc another, but also wori which arc very ~ n u c hrclatcd, differing only in morpliolog: o r tlicir sound form. For cxamplc, in E~lglish from tf word 'conlparc', words likc 'comparcd', 'comparcs', 'con paring' and 'comparison' arc derived. Such morphologic; changcs also have thcir equivalents in SL. Acljusting th movcmcnt o f a sign by changing the speed or tension o rate o f repetition, gives ;III SL the ability to dcrivc noun from vcrbs, such as 'coniparison' from 'comp;lrc', ;IS well a produce derivations which arc unicluc to the SL. Fo example, in ASL the signs for 'church', 'pious', ant arrow-mi~ldcd' (!) differ only in t l ~ cninnncr of movcmcn involved. Thcrc ;Ire, then, ullinflcctcd forms o f signs whicl call hc. dcfi ncd hy the fc;lti~rcso f place, confi gu1-.ltiol11rtio anc nlovCnicnt, with v;lri;ltio115 i l l movcmcnt providing the nic;llis for ~liorphologicxl v;lrintion ;111d c l ~ a ~ i g iel l- ~; I S I > ~ C ~ . S1, words ;111d 111orl>hc111cs arc. m ; l ~ ~ i p i ~ l ; i tinc d 1li11ch the



I WENT



Figure 4.2



:



I



WON



1)ritish sign Ia~igiingcvcrbs



Just h o w signs tcnd t o be made when comparing ASL to say, Chincsc Sign L a ~ ~ g i ~ ((:SL), wc w e find not uncxpcctcdly that not only ;ire the signs tl~cmsclvcs complctcly clifkrcnt, but, also, that ASL ilscs :I slightly marc pi~lchcdh:lnd configuration, with tllc fingers curling under, fist-likc, into the p;llnl, much nlorc s o th;ln i l l c.cli thc proponc*lits of tlic tc.;~ching ( spcccli, gclic.r:illy c-;~llcdtlic. 0r;ll Appro:~cli. 'I'lic O r . Ap~>ro:~cIi h;ls n wortliy aim, t o tc~;~cIi t l i c hc;lri~ig-ilii~>nirc



Sign language, written language and the deaf



83



to produce and understand speech so that they can communicate with the hearing community. Unfortunately, historically, its supporters, w h o controlled education in the schools, advocated the use o f speech to the exclusion of any other means of communication. T h e use of SL was proscribed even for communication among deaf persons. While the Oral Approach advocates, such as Daniel Ling and the Ewings, may even admit that sign language is a language, they argue not only that the learning and use of sign language negatively affects the acquisition of speech but that without speech there will be defective thinking. (The teaching of reading and written language are attacked for similar reasons, which is w h y the teaching of reading in so many deaf schools was - and often still is - delayed until children are beyond the second o r third grade.) These contentions, which have n o basis in cmpirical observation o r psycholinguistic theory, are false. If anything, knowledge of SL and reatling facilitate the acquisition o f speech. And, as far as thought is concerned, deaf people without speech are found to test nearly as highly in intelligence as hearing people. It is unfortunate that such erroneous ideas continue to bc hcld in so many places. Thc Oral Approach focuses on thc teaching of spccch production. Childrcn from thc the agc o f 2 o r 3 years onwards arc spccially trained in thc skill of articulating spcccll sounds. Also, it is not uncommon at prcscnt to havc somc computcrizcd cquipmcnt that displays sou~lds and assists in thc teaching. Many children d o respond and d o acquirc n fair ability to spcak. For the most part, howcvcr, thcsc arc children who havc only a modcratc hcaring loss. Those with profound impairment typically d o poorly. The understanding of speech is usually fostcrcd through both exploiting any residual hcaring that learners may have and tlic tc;lching of 'spccchrcading', c o ~ n m o n l yknown as 'liprcnding.' With spccchrcnding, ,111 ndcpt person can interpret nhout l ~ a l fof w l ~ n tis snid, wllich, given the g-rc;lt nniount of rcdundnncy in ordi~laryspccch, is enough to guess nioc;t of the content. Many sou11cls, howcvcr, ;lrc p;lrticul;irly dil'ficult to diffcrcntintc vis~l;~lly; thc'sc- includc most vowels, c.g. 'a', 'c', 'II', 'i', nlid rn:lny c o r ~ s o ~ ~ ;sucli l~its ;IS 'k', Lgl,'I), r , 's , sll', 'c.11' :111tl'j.' 111 ;~clciitioii,in ccrt;li~i I:11igua~:cs, s11c.11;IS J ; ~ p ; ~ l ~ cwliicli sc, csliil>it rcl;itivcly little L



1



1



84



,



', I



I\



I



1 1 \



I



First language



lip and facial movement, the articulation of speech especially difficult to deal with. As was noted above, a great problem with the O r Approach is that it tends only to work for a portion of tf hearing-impaired population. Research shows, unsurpri: ingly, that the less people can hear, the less they will be ab to produce and understand in terms of speech. Thu relatively few children who are born with a severe ( profound hearing loss (over 75 or 80 decibels in their bett ear) acquire any significant degree of speech. Even tho with a lesser hearing loss often do not acquire sufficient clear pronunciation such that they are understood I ordinary hearing persons. As a result of a pure O r Approach education, many hearing-impaired persons a not only unable to communicate with the hearing con munity but are unable to communicate adequately wig their hearing-impaired colleagues. It was this tragic situ. tion, one that continues in many placcs today, th convinced many educators of the deaf, that cducation programmes should include sign languagc in thcir cu riculum along with speech training. Thcsc programme which generally go by thc namc of Total Con~municatiol spread in thc 1970s in the US, Canada and othcr countrie While Total Communication is now widely acccptcd I many countries, nevcrthclcss, thcrc is still grcat rcsistanc (as in Japan) to admitting sign language into the cducation. curriculum for thc hearing-impaired.



4.7 The Written Language Approach



4.7.1 The importance of literacy a n d essentials of t h e approach Although Total Coi11nlii11icatiollhas in~provcdthc lot of thc deaf in a significant way by providing sign Ial~g~lagc i i ~ addition to spccch training, one great educational prohlc~n rcmains, that of literacy. On the average, hcaring-i~npaircd persons, still (cvcn aftcr a Total Cominu1lic:ltion cd~ication) graduate from high school with a reading lcvcl ccliiivalc*nt only to th:lt of ;I hcaring cllild in (;r:~tic 5 of clcmciitary



Sign language, written language and the deaf



85



school. (Their writing ability is poorer.) This is true for the US, Japan and other developed countries and is even lower in less-developed countries. The problem is that reading and writing are dependent on knowledge of ordinary speechbased languages; we use our knowledge of the grammar of the ordinary language both to understand what we read and to produce what we write. Since the hearing-impaired person's knowledge of speech-based language is usually quite limited, the ability of that person to acquire literacy based on that knowledge is similarly limited. Given the overwhelming necessity of being able to read and write in order to function well in modern society, it is not surprising that we find that most hearing-impaired people are able to secure only low-level jobs, when they are able to secure jobs at all. A high level of literacy is essential if the hearing-impaired are to realize their potential. Towards this end, I believe that an approach, one that I call the Written Language Approach, can be of benefit. The essential idea of this approach is that the meaningful written forms of an ordinary speech-based language such as English or Spanish (its words, phrases and sentences) are acquired through direct association with objects, events and situations in the environment. Thus, just as hearing children learn language by initially associating the speech sounds that they hear with environmental experiences, hearing-impaired childrcn can learn language in a similar way, but through an association of written forms with cnvironmcntal expericnccs. As a rcsult, hearing-impaired children will acquire csscntially thc samc vocabulary and syntax of the languagc of hcaring childrcn because, in the writing of a languagc such as English, Danish or Chincsc, virtually all of thc vocabulary and syntactic structures that appcar in spccch also appcar in writing, c.g. subject-verb relations, objcctvcrb relations, ilcgation, question, rclativc clausc formation, passives. This is not to say that thcrc arc no diffcrcnccs i l l spccch and writing. Esscntially, though, o11c basic grammar undcrlics both forms of cxprcssion. And, because virtually any sci~tcnccor idea that can bc cxprcsscd in spccch can be cxprcsscd in writing, we can say, by analogy, that written Iang~iagccan bc rcgardcd as a complctc language. Its 111ain diffbrcncc with spcccll concerns thc physical nlcai~s of



86



First language



transmission sound.



- writing involves light,



Sign language, written language and the deaf



while speech involvc



4.7.2 Historical perspective



Actually, the ideas which I propose here are by n o meal new. Further, they have a remarkable history although fe in deaf education are aware of them. When I first conceivt of these ideas, I truly believed they were original with m It was only after successfully completing original researc with American and Japanese deaf children using tE, approach, did I discover that a hundred years earlier, no1 other than Alexander Graham Bell had taught writtt language to a 5-year-old deaf boy with some success ar that 200 years before him a thinker by the name of Dalgarr had, in 1680, formulated the same approach a t Oxford! Bt was well aware of Dalgarno's conception, for, in 1883 Bc stated, I believe that George Dalgarno . . . has given us the truc princip



to work upon when he asserts that a dcaf person should be taugl to read and write in as ncarly as possible the same way that youn ones are taught to speak and understand their mother tongue. R should talk to thc dcaf child just as wc do to the hcaring onc, wit the exception that words arc to bc addrcsscd to his cyc instcad ( his ear. Perhaps Itard may have been aware of Dalgarno's ideas, fo he, too, used a written languagc approach with thc Wilc Boy of Avcyron and was quite successful in that approacl as coniparcd to his cfforts in using spccch (Chapter 3). Somehow in the 1970s the idea of the Writtcr Langi~agcApproach bcgan to sprout once again. I t camc tc mc and it came to Shigctada Suzuki, a dcaf educator in Japan; I liavc heard runiours of so111c work in Europc but havc not been able to vcrify these. Other than 111y work with English and Japanese, and Suzuki and his collcagucs with Japa~~cse, no otl~crserious rcscarcl~as bccn done ~lndcr this conception. In fact, to my knowledge (and dismay), it SCCIIIS that 110 one cvcn talks ahout tllcsc ideas.



87



4.7.3 Written language and reading distinguished At this point it would be well to consider the distinction between written language and reading. T h e main difference is that written language is learned directly from the environment without the use of any prior linguistic medium, such as sign language o r speech. Reading, by contrast, is learned through a linguistic medium. Thus, when w e say that a hearing person has learned to read, we presume that that person already had language prior to learning to read. We furthermore presume that reading was taught through the medium o f that language, i.e. speech. Typically we point t o a written word and say it ('dog'). And, in reading a book we point to written words and say them ('The little dog ran to the girl'). T h e child interprets the written words by means of the vocabulary and syntax which the child has already learned in speech. This is the essence of reading. However, suppose w e consider a hearing-impaired child who does not know speech or sign, and we point to the written word 'dog'. That child will not be able to understand the meaning of this word. If, however, a picture of a dog is placed alongside the written word, then thc child will havc the opportunity of learning the meaning of the word. A child who learns languagc in this way can bc said to havc learned writtcn languagc. T h e writing itsclf is the primary mcdium for languagc concepts. The child must. discover the meaning of the writtcn vocabulary items and then induce thc syntactic relations that pertain to those items, just as hcaring children do with spccch. Learning to read is a much casicr process than Icarr~ing to interpret writtcn languagc, since, in learning to read one will not havc to acquirc the grammar of the languagc. Thc grammar is already known bcforc the first rcading lesson begins. T o Icarn to read, onc siniply has to lcarn how visi~al writtcn forms correspond to known spccch sound forms. T o Icarn writtcn languagc, however, the vocabulary, ~norpliology,syntax and other aspects of tlic grammar must all be ;~cr~uircd on the basis of the visual writtcn forms and their rcl;~tionship to n~caningf~llobjects, situations and cvcnts. Thc hearing-impaircd cl~ildmust undcrgo tllc samc 1voccss of Inngii;~gclearning that the hearing child did in



88



acquiring language. Such a process can be expected to more time-consuming than learning to read.



1



I



I I



1



Sign language, written language and the deaf



First language



4.7.4 Assessment of the Written Language Approacb There are a number of distinct advantages to the Writt Language Approach: 1. T h e learning medium is appropriate. Perception of writ! items depends on vision, a medium in which normal hearing-impaired have full capability. 2. Written language knowledge need not be acquired by instructors. Parents and teachers of the hearing-impai do not have to learn the written language in order teach it; they already know the writing system and hi to use it. (Hearing parents who wish to acquire slanguage which they can pass on to their children m spend years in learning it, just as they would any o t second language.) Only relatively simple instructio methods and techniques need be acquired. 3. Instruction can begin early. Parents of hearing-impai children can teach their children written language homel during the children's most formative ye; Children as young as 6 months can be exposed written language in a natural way, and learn in supportive comfort of their own home. Children hearing parents need not have to wait for two or th ycars beforc their parents learn somc sign langu~ which can be passed on to them. Nor would child1 have to wait at home before they can be takcn to schc for what is nccessarily a limited amount of si languagc or speech instruction, givcn thc nurnbcr hours that the child can spend in school. 4. All hearinpimpaired children can henqfit. Effort dcvotcd teaching written languagc is never wastcd sincc wh; cvcr is lcarncd will improve thc child's lcvcl of literal in the filturc, whcther thc child lcarns spccch or sit languagc. 5. Written lan L rl-:



--



2-



C c



5L.S $ ka L)



194 Language and mind



with substitutions made in terms of the three characteristics of signs (configuration, location and movement), the result is the production o f many meaningless individual signs or of signs which have meaning but are nonsensical in sentences. And, just as in Wernicke's aphasia for speech, much substitution occurs within the same lexical category: nouns for nouns, verbs for verbs, etc. Furthermore, substitution occurs, too, along semantically related lines, such a: 'daughter' for 'son' or 'bed' for 'chair', producing similarl) bizarre results. O n e signing patient displayed Wernicke's aphasia symptoms in his writing as well as sign language. leading researchers to conclude that there is a genera linguistic dysfunction at work and not just a mot01 dysfunction impairing the ability to sign. Signers who have suffered damage to the righi hemisphere generally do not display aphasia symptoms ir the production of signs. Their sign production sccms tc remain grammatical and unimpcdcd. Howcvcr, thcsc samc persons are likely to suffer some impairment in thci~ comprchension of signs. Why it should be thc casc that thcii ability to understand signs is disturbed but their ability tc producc signs is not, as yet remains uncxplaincd. In any case, thc sanlc gcncral pattcrns of aphasi: cmcrgc in sign as with speech. l'aticnts with lcft-hcmisphcrt damagc suffer the same aphasias whcthcr their language is : spoken one or a signed one. Thus, cvcn though the rich, hcmisphcrc is disposed to process spatial tasks, when thc task involvcd is a linguistic onc and spatial syntax i: involvcd, thc lcft I~cmisphcrc asscrts dolninance. Clcarl) this shows that it is the lcft hcmisphcrc that is specially ccli~ippcdto handlc Ianguagc, whcthcr the modality of thc languagc be that of spccch or sign.



9.6



Methods of investigating brain and language



9.6.1 Established m e t h o d s : post-mortem, injured people, electrical stimulation co~iipnr;~tivclylittle undcrst;i~idi~igwc 1i:rvc of the ~ici~rologicnl basis of I ; ~ n g \ ~ nigl lctllc I,r:~iliis the result of tllc ;~pplicntionof ;I rc~l;~tivcly sni;~ll~ i i ~ l i i h cofr ~ ~ i c ~ t l i o Tlic ds.



The,



Language and the brain 195 '



oldest method, that used b y Broca himself, is the postmortem examination o f the brains o f patients w h o had displayed language disorders while they were alive. T h e abnormalities he found in certain areas o f their brains in post-mortems correlated with the language symptoms they displayed while alive. Another method involves observing the language o f patients w h o have had brain operations. A person might require - because o f an accident o r a tumour for example - the removal o f a lobe o f the brain (lobectomy) or even o f an entire hemisphere (hemispherectomy). Then, too, the study o f the language of living patients with severe brain damage caused by accidents or war-time injuries was and still is a fruitful method o f investigation. A more recent method, pioneered by Penfield in the 1950s, involves the electrical stimulation o f the cerebral cortex in patients w h o are conscious during brain surgery (electroencephalography). O n being stimulated, patients would report, for example, that they remember childhood cvcnts or old songs. ( H o w to verify what the patient says about thc past is a problem.) T h e use of this procedure has been very limited since it is restricted to the open brain areas o f pcrsons w h o arc undergoing surgery without the use of anacsthcsia.



9.5.7 N e w high-tech m e t h o d s : CAT a n d PET In rcccnt years, revolutionary ncw mcthods have bccn dcvclopcd which Icnd thcmsclvcs nicely to thc study of language and thc brain. Thcsc involvc powerful ncw tcchniqi~cs in radiological imaging. CT (bcst known as C A T , Computcrizcd Axial Tomography) and P E T (I'ositron Eniission Tomography) arc the most widcly used in this rcgard. 130th o f these tccl~niqucs use the brain as it is, witl~outsurgery or any othcr radical proccdurc. As such they may be used with normal persons as wcll as those with 17r;lin problc~iis. A C A T scan involvcs usi~iga11 X-ray sotlrcc so ;IS to m;lkc Ilunicrous slice sc;lns, thc images of whicli arc integrated I>y computcr to construct ;111 image o f tlic wliolc I7r;lili or portio~iof'it. (:urioiis scientists rccc.ntly used (:AT to cs;~niilic;I sccrioli o f thc Imin o f 1jroc;l's origi11;il p:~ticnt,



Language and the brain 197



196 Language and mind I



I



I



I i \



I



I



I



Leborgne, who is better known in scientific literature as 'Tau'. (He was nicknamed this because that sound was the only one that he could utter.) The brain has been preserved for over a 100 years in a medical museum in Paris! Modern researchers were able to re-examine, as it were, the patient, to determine just which areas of the brain had been affected. Tomography has shown that Broca was essentially correct in concluding that the language deficits of the patient had indeed involved trauma to the area of the brain which bears his name. Fascinating though the C A T method may be, undoubtedly the most exciting one to come along to date (in 1972: is PET, Positron Emission Tomography. Unlike C A T which images slices of the brain and integrates them into : whole by computer, PET allows for the dircct observatior of the brain as a whole. Like CAT, it also allows for thc study of language in both the normal or damaged brain The PET procedure involves injecting a mildly radioactive substance into the blood and then tracing the blood flou patterns within the brain by means of special detector* surrounding the person's head. Thcsc detectors ther provide a colour image. With PET, areas of thc brain ligh up in different colours whcn there is an incrcasc in blooc flow (an indication of increased brain activity). As subjcct perform various linguistic tasks givcn to them by rcscarch crs, it becomes possible to map the areas which undcrlic language use in the brain in a way that was ncvcr possiblc before. In reading, for example, the PET scan shows that ligh signals from the eyes (as we look at the printed word) arc scnt to thc visual area of the cortex (in thc occipital lobc' thcn forward to the visual association arcas. Whcn spccch i: hcard, on the othcr hand, thc acoustic signals from thc cal go to thc auditory cortcx (in thc temporal lobc). PET' scan! arc ablc to dctcrinine how closcly inodcls of spccch production and understanding a11d models of rcading ant writing conform to reality. PET has already providcd cvidencc that counters ollc vicw of rcading which holds that thc printed word IIILIS~ always bc soundcd out in ordcr to bc undcrstood. I t sl~owcd that visual forms of words may bc scnt directly to the scrnantic arcas in thc frontal lobc for con~prchcnsion.Acccsq



I



(



:'



' .



,



'



4



(



I



to the stored auditory form in the angular gyrus for mentally sounding out words is not necessary in the recovery of meaning. This direct semantic connection occurs mainly with common, familiar words. However, even when people have learned to read by a method of sounding out letters, like Phonics, after repeated exposure to the written words, the sounding-out activity will be bypassed and the semantic areas will be directly activated. Only when a special task is presented, such as trying to determine whether 'blue' or 'go' rhymes with 'shoe' (the words are presented in written form), will a portion of the brain near the auditory cortex light up, indicating that internal sounding out is taking place. New discoveries by using PET are of such widespread interest nowadays that they are being reported frequently in the newspapers and magazines. In October 1991 I read where researchers, investigating the recall of a word, discovered that both the frontal and visual lobes of the right hemisphere are activated. Such functions of the right hemisphere were previously unknown. Then, on 17 February 1992, thc New York Times News Service reported that, by mcans of PET: 1. The brain distributes language processing over a few or many cerebral areas. According to researcher George Ojcmann, many additional areas of the brain, besides Broca's and Wcrnickc's arcas, are involved in languagc. 2. Second languages arc rathcr looscly organized in thc brain. A sccond languagc can evcn be localized in the right hcmisphcre. Thc casc of Carla, 22, is cited. She grew up spcaking both hcr nativc Italian and English. Whcn shc bcgan training to bccomc a simultancous translator, hcr languagc abilities in both Italian and Englisll wcrc localizcd in thc lcft hcmisphcrc. Aftcr tra~~slationtraining, l~owcvcr, English shifted to thc right l~cinisphercalthough Italian rcrilaincd on the lcft! Such findings, if replicated, will s o ~ n c h o wliavc to be incorporated into gcncral thcory. 9.7



Mind and brain



W11:lt is thc relationship bctwccn mind nnd hrain? Certainly,



198 Language and mind



without brain, there would be no mind. But does th necessarily imply that the mind is under the complel control of the brain? These questions take us back t Chapter 7 where the relationship between mind and bod was discussed. What was said for body, of course, applies 1 brain since brain is a part of body. The issue is this: Is there a pevfect correlation between person's experiences and the events which take place in the brai While there is strong evidence of a general dependence mental occurrences on the functioning of the brain, it h never been shown that the correspondence is so exact tlfrom observation of a person's brain one could arrive at knowledge of the person's experiences in every detail. Ma1 theorists believe that consciousness plays a role in determi ing events in the mind and, hence, in the brain; and that, it were not for conscious control, we would find ourselv in a continual dream-like state where events would occ and be experienced in the mind but where we would ha no power to act in ordcr to control such events. We have now arrived at another philosophical problci the issue of frec will and determinism. Arc cvcnts in t mind wholly determined by other evcnts? Thcsc can events which had occurred in the brain or cvcnts which h occurred in mind. T h e determinisn~ of cvcnts can physical (from thc brain and othcr parts of thc cent1 nervous system) or it can be mcntal (fro111 the operations the mind), or both, depending on whcthcr one takes dualist or monist view of thc iinivcrsc. According dualists, thcrc are two kinds of stuff in the univcrsc, ti physical and the mental. For monists, thcrc is just one kina with the physical usually being prcfcrrcd. Othcr rcccl approachcs, such as Functionalism, have tricd to avoid suc distinctions altogether by offering pragmatic analysc These, l~owcvcr, havc not bccn succcssfi~l. Even tk principal foundcr of Ful~ctionalism,Jcrry Fodor, has foun it ncccssary to nbandon that philosophy. Altllough the free will V C ~ S L I Sdctcrminism issue hn cngagcd thillkcrs for ovcr thousancis of yc;irs nnd m;ln solutions have hccn proposed concerning its rcsoli~tion,i h;ls yct to bc solved to thc s;~tisfactionol' ally but tllc 111os zcalous. Whct11c.r future ckncluiry into the fi~nctionill!: of thc



Language and the brain 199



1



i



brain can provide a resolution to the issue, remains to be seen. Discussion questions



Are you right-handed o r left-handed? Are you a mixture o f both? Why might you have the hand preference that you have? Does the size o r weight o f the brain o r the cerebral cortex have anything t o d o with language in humans? What functions are typically handled by the left hemisphere? By the right hemisphere? What is dominance and h o w does it differ from lateralization? Why might left-handed people have reading and writing problems? What happens to people's functioning when the connection between the hemispheres is cut? Also, how docs this relate to lateralization? If a person produces fluent but nonsensical sentences likc 'Wc down and to the other' o r substitutes words for thc correct word, such as 'pool' or 'wet' for 'water', what kind of aphasia might this person have? What aphasia is characterized by spcech such as 'Go store milk'? Why might right-handed peoplc gcncrally prefcr listcning to speech with thcir right car? If a fricnd of yours was in a car accident and his o r hcr brain was damagcd, how would you tcst for thc following languagc f~inctions: (a) spccch production; (b) spccch iindcrstanding; (c) reading; (d) writing? 110 you think thcrc is a critical agc for first- o r sccondlanguage learning? Why is it necessary for a comprchcnsivc thcory of brain structure and fi~nction to include localization and holistic approaches? What advantagcs docs PET have ovcr traditional methods of brain invcstigation? Is brain diffcrcnt from mind? C;in n complctc understanding of brain providc a con~plctcunderstanding of mind?



200 Language and mind



Suggested readings Blumstein, Sheila (1988) Neurolinguistics: an overview oflanguag brain relations in aphasia. In Newmeyer (1988). Caplan, D. (1987) Neurolinguistics and Linguistic Aphasiology: / Introduction. New York: Cambridge University Press. Lenneberg, Eric (1967) Biological Foundations of Language. Ne York: Wiley. Newmeyer, F. (ed.) (1988) Linguistics: The Cambridge Survey, v III, Language: Psychological and Biological Aspects. Poizner, H., Klima, E. and Bellugi, U. (1987) What the Hat Reveal About the Brain. Cambridge, Mass.: MIT Press. Sasanuma, S. (1985) Surface dyslexia and dysgraphia: how ; they manifested in Japanese? In K. E. Patterson, J. C. Marsh and M. Coltheart (eds), Deep Dyslexia. London: Routledge Kegan Paul. Sperry, R. W. (1982) Some effects of disconnecting the cereb hemispheres. Science, 217, 1223-6.



Journals Brain and Cognition Brain and Language Neurolinguistics Studies in Neurolinguistics



i



PART 3



Second language



i



CHAPTER 10



Children vs adults in secondlanguage acquisition



\



i



,



10.1 A common belief



Most people believe that children arc better than adults when attempting to learn a second language. That seems to be backed up by common observation, since young secondlanguage learners d o seem to pick up another language rather quickly, just by being exposed to it. Whether this belief is justified is the focus of this chapter. As w e shall see, all of our psycholinguistic knowledge (and then some) will i bc nccded in order to approach a reasonable answer to this problcm. Factors involved in second-language acquisition can be i dividcd into t w o kinds, thc psychological and the social. i Undcr 'psychological' wc shall considcr intellectual processi r i ~ ,which is involved in the detcrn~inationof grammatical structures and rules, memory, which is essential for lcarning to occur, and lwotor skills, which involve thc use of thc I articulators of speech (tongue, lips, vocal chords, ctc.) for t11c prodiiction of thc sounds o f a second languagc. Undcr 'social' wc shall considcr thc types of situations, scttings and intcractions which affcct our ability to lcarn a sccond languagc, in particular thc watrrral and clnssroorn situations. I



10.2 Psychological factors affecting second-language learning I



10.2.1 Intellectual processing Tlicrc ilrc only two ways to Icarn the stroctiires 2nd rules of ;I scco~ldlanguage: somconc can cxpl;~inthcnl to y o u or you ciin figure tllcnl out for yourself. T h e first wily niay bc tcrmcd 'explication', the sccond 'induction'.



, ' 204 Second language



Children vs adults in second-language acquisition 205



Explication



\



\ 1,



I I



I 1 I



I



Explication is the process whereby the rules and structure: of a second language are explained to the learner in his 01 her native language. The person is then expected tc understand, learn and apply them in the second language N o second language, however, can be learned entirely b! such means. Although many second-language teacher, assume that the rules of a language have all been discoverec and written down, and that all they have to do is reac enough to find them, this is not so. One cannot go to . bookstore and buy a book or any number of books whicl come close to explaining completely the grammar of an: language. Even for a language such as English, the mos studied of all languages, one still finds linguistic journal discussing the concepts involved in such commonplac~ features of English as the article and tense. Explaining is rarely done by parents or others w h e ~ children acquire a native language, yet children by the ag of 4 or 5 understand and speak most of their native languag. quite well. Parents do not even atterllpt to explain . relatively simple morpheme rule, likc that of the plural You don't hear them saying: 'Now, Mary, to make thc plural of "dog" you add a "z" sound to the end of the word while with "duck" you add an "s". You do this, Mary because the last sound of "dog" has a voiced consonant an( the last sound of "duck" has an unvoiced one.' Even i parents were able to formulate the explanation, which mos cannot, they know that their childrcn would not be ablc tc understand it. Similarly, parents do not tell their cliildrcr that there is a Subject Verb Object ordering in t1ici1 language, or that, in order to negate a scntcncc likc 'John wanted some cliocolatc ice-cream', do must be inserted, thc tense on tlie verb must be shifted onto tlic do, the ncgativc marker not lnust be illscrtcd after do, and rornc must change to n n y , so that the scntcncc 'John did not want ally chocolate icc-crcaln' will be the rcsult. It is ollly with :I high degree of intcllcct~i;rl maturity that n person can understand such explicit cxplar~ntions. I-lowcvcr, while some syntactic rulcs niay be so conlplcx ;111d;~bstractthat few pcoplc otllcr than students of



+



linguistics can comprehend them (let alone remember them so as to use them correctly!), there are simple rules which can be taught by explication to adults and older children without much difficulty. For example, a mature Korean speaker studying English could be told that there is a Subject Verb Object order of constituents or that English requires count nouns to have a plural marker added when more than one object is involved. O n the basis of these descriptions, a learner can learn relevant usable rules. In such cases, explication may even be a faster means of learning than induction, since induction requires that a learner be repeatedly exposed to words, phrases and scntences along with relevant situations that give some indication as to their meaning.



I



+



'



\



1



+



Induction ,



: j



'



+



I



Learning rules by self-discovery is the essence of the process of induction. The child who is exposed to second-language speech and remembers what he or she has heard will be able to analyse and discover the generalization or rule that undcrlics that speech. Actually, not only must the learner devise, co~~sciously or unconsciously, the rule based on the speech that has been heard, but lie or she must also figure out how those rulcs are to be applied in other cases. For example, given the scntences 'John danccd then John sang' and 'John danced and then he sang', spoken in a relevant situation, the lcarncr can determine that the two sentences arc related, with hc being a replacelllent for Jolin. The lcarncr 111ust also figure out that while /.re can rcplaccJolzrr in tlie second of the colljoincd scntcnccs, i t cannot do so in the first (c.g. in 'I-Ic danccd then John sang') since in that case the pronoun Iic must rcfcr to sonleone other than John. With such a rille, the lcarncr is 011 the way to being ablc to i ~ s c and understand increasingly complicated structures i ~ ~ v o l v i n gpro~lon~inalizatio~l. Such p h c n o m c ~ ~ ans pronon~inalizatio~~, ncg:ltion and thC plural arc Ic:~r~lcdby inciuction and hcconlc part of ;I young nntivc speaker's language knowledge quite early, long before the child cntcrs scl1001. .



.



m o m z .$



y d



n



$4



5 Eo ,-, o



g z



U O P I



s



s o m



.2



e,



u



5



E'"



0



d U ?



3



Y Z E



Q 0 0



u



.3



2 % ".



Cd



rO



bar=



r



e



+ o



0



6s



-2*-2 r o o-



a



.z CJ



.$!



2(



E



2



+



.rp.z a .-E



2



.3



UAzc a U e c u m aG8 I f c



("



L



0



a



c *9z2 .s 8 A $ 2 .e r e ,



o > 8c.s



s ac 0 e



CJ



0



-d



b



L%p 0



r,2



208 Second language



Children vs adults in second-language acquisition 209 I



world, built up over decades, remaining intact. O n the other hand, the ability to deal successfully with material such as lists of new names and words (clearly relevant to learning vocabulary in a second language) is affected.



10.2.3



Motor skills



Good pronunciation, which is related to the ability to control the organs of speech, is clearly an essential part of learning a foreign language. Jaws, lips, tongue, vocal chords, etc., are controlled by muscles, all of which are under the general control of the brain. The organs of speech have to do the right thing at the right time if one is to speak, and especially if one is to speak a second language like a native. Evidence shows that the particular motor skill of speech pronunciation is best developed at a younger age. In areas other than language, one can obscrve relatcd phenomena. Few people who start some new sport or involve themselves in such disciplincs as gymnastics or violin playing at the agc of 20 are ablc to attain thc samc level of proficiency as those who start 10 or so ycars earlier. Somewhere around the agc of 10 and 12 ycars the ability to acquire new motor skills begins to decline. The reason for this dcclinc is as yet unknown, although since the decline is of such a gcncral nature, involving all parts of the body, it seems likely to bc duc to some changc in central functioning in the brain. This is not to dcny, howcvcr, that there may not be other, secondary rcasons why a foreign accent persists in a second languagc. One's pcrceptual ability to hear foreign speech sounds accurately (cspccially thosc which differ only slightly from sounds in thc nativc languagc), may also contributc to incorrcct pronunciation. Children lcarning a sccond language typically lean1 to spcak i t with a pronunciation that is indistinguishable from that of a nativc speaker. Few adults, on thc other halid, arc ablc to achicvc suc11 a level. What is puzzlillg, howcvcr, is that thcrc appear to be cxccptiolls among adults. I an1 of the vicw that solnc adults (mature persons beyond t l ~ c; ~ g cof pubcrty) do Icarn to spcak a second language with nativc



pronunciation. (A counterviewpoint on whether there are truly exceptions is discussed in Section 10.5, 'Critical age'.) / Is this simply a matter of individual differences or are some persons somehow exempt from the barrier experienced by most adults in learning the pronunciation of a second ' l a n g u a g e ? There is no way of answering this at our current stage of scientific knowledge. Research studies that have been done on pronunciation in second-language learning are unanimous in their findings (e.g. Oyama, 1976; Tahta et al., 1981; and for a book length consideration see Scovel, 1988). Younger children in immigrant families, for example, are 1 found to acquire perfect or near-perfect accents, while their older siblings or parents generally do not, even when these older people have mastered other aspects of the language such as its syntax and vocabulary. While there is general agreement that the ability to acquire pronunciation declines, there has been little discussion as to whether there is a ' decline in the ability to acquire syntax and other aspects of language. Most theorists seem to think, and I would agree, that there is little or no decline in this respect. A summary of the various psychological factors and their functioning according to age is shown in Table 10.1. Thus, for example, children undcr the age of 7 years are ; ' ratcd 'high' on all psychological factors except 'explicative' I proccssing, whilc, adults are rated 'high' on 'inductive' and 'cxplicativc' proccssing but 'low' on 'mcmory' and 'motor skills'.



Table 10.1 Psychological and social factors affecting secondlanguage Icarning for childrcn and adults I



I'sychological fnctors



Social hctors



Intcllcctu:il



Situ:~tion



Ir~tlt~crivcExplicative Memory



Motor skills



Natural



Classroom



C:lliltlrc~~ Untlcr 7 t-lil:l~ 7-12 IFlil:l1



Low Mcdiuni



High I-ligll t-lil:l~ Low Mctlll l i ~ l i Mctlll ligh Mi-tliu~n M c d i t ~ l l ~



Adllltz v I



t lip,li



1.ow



1I



Low



Low



111):I1



Children vs adults in second-language acquisition 211



210 Second language



10.3 Social situations affecting second-language learning



What are some of the social factors which might explai why children and adults seem to differ so much in the ability to acquire native speech in a foreign language? Such discussion can be divided into t w o parts, learning in 'natural' setting and learning in the 'classroom'. 10.3.1 The natural situation



A natural situation for second-language learning is 01 where the second language is experienced in a situation th is similar to that in which the native language is learnel That is, language is experienced in conjunction with tl objects, situations and events of everyday life; it is 11, taught in a classroom. The paradigm case would be that of young child going to live in another country and learni~ that country's language, not by any explicit teaching, but t interacting with playmates, as, for example, an Englisl speaking 4-year-old girl from London who goes to Bciir with her parents. Through playing with Chinese childre, she soon learns Chinese. Generally speaking, as one gets older there is a dccli~ in the kind of social interaction which promotes languak learning. Adult second-language learners will have sii nificantly fewer good language-learning opportunities in new culturc than will children. O f course, if thc adults sta at home, thcy will not be able to meet and talk much t nativc spcakers. If they work, then, because of their lack c second-language ability, thcy will not be hircd to do wor that requires nativc spcakers to linguistically interact wit them. Whether tlicir work allows them to use tlicir nativ language (as business people, language teachers, ctc.), o whctlicr their work involves a minimal amount of scconc languagc (construction work, disliwashing), in cithcr cas learners will Ilavc but a limited opportunity to cxpcricncc appropriate scco~~d-languagc dat;l i l l tlic 11ati1r:lls i t i ~ ; ~ t i o ~ l Except for situations involving love or nloney, it is ;~lnio.; iml9.ossihlc to imagine n situatio~lin wl~ich;ldults would bi conti~iu;~lly cxposcd to the s;rnlc good qu;llity ;111di1~1;111tiry of IanguagC tliat a child rccc.ivcs.



For adults, social interaction mainly occurs through the medium of language. Few native-speaker adults are willing to devote time to interacting with someone w h o does not speak the language. T h e adult foreigner, therefore, will have little opportunity to engage in meaningful language exchange except for picking up bits o f language that are experienced in the workplace o r in shopping. In contrast, the young child is often readily accepted by other children, and even adults. For young children, language is not essential to social interaction. So-called 'parallel play', for example, is common among children. They can be content just to sit in each other's company and just speak occasionally. Adults rarely find themselves in similar situations. The older the child, however, the greater the role that language plays in social interaction and the more the person will experience difficulty in being accepted. Peer group acceptance becomes a problem, especially around the age of puberty. Even children w h o speak the same language but ' come from a different school o r town have difficulty in gaining acceptance. Without such acceptance, secondlanguage lcarning in a natural situation can hardly begin. Because language is essential for social interaction and , because people gcncrally cravc such interaction, foreign ; adults oftcn tcnd to stick together in a new environment. 1 Friendships for the adult arc easier to form in the old language and even business can oftcn be conducted at least partially in thc old languagc. T l ~ c n ,too, many largc cities , with sizable foreign populations have radio, television and newspapers in thc forcign languages. All of these things tcnd to reduce thc amount of second-language exposure for adults in a way that does not occur for cl~ildrcn.



,



, '



10.3.2 The classroom situation



The classrooni for sccond-languagc Icarni~igis a planned, o r some might say, a11 ;~rtificiallyconstr-uctcd, siti~ation.As we all know, physically, tlicrc is n roo111 which is ;~rr;lngcdso thnt i t is iso1;ltcd fro111 the rest of life. In the room tl~crcis a tc;ichcr ;111d;I 1111nil>cro f students. Tlic tc;lchcr is thc one cvho Icnow.; tlic l:i~lgu;igc~ ;111dtllc S I L I L ~ C I ;1rc ~ ~ S tlicrc to Ie;lr11 t11c I ; ~ n ~ ; r ~ : ~111 g c .tlic. cncloscd sp;lcc of tllc cl;lssroom, 11othi11g 11;lppuis i ~ ~ i l c s stlic tcncllcr n1:tkcs i t 11nppe11.



Children vs adults in second-language acquisition 213



212 Second language



Students do not act on their own but follow the direction of the teacher. All other aspects of life are suspended c subordinated to language learning. This, of course, is ver different from the home or street where the learner eats at table, walks around doing things, bathes, plays outdoc games, etc., all while hearing and using language i conjunction with these activities. In the natural situation, language is but one aspect ( life, an aspect which accompanies other life events. In tl: classroom, however, language itself becomes the prim aspect of life around which all else revolves. The languag that is to be experienced by the students is planned. Whi there are degrees of planning with more or less emphasis c speech, literacy, spontaneity, etc., nonetheless, the course events is planned and the teacher is the planner. In physically isolated room, where only onc person, tl teacher, is the prime sourcc of the new language, planning unavoidable. This is true even for methods which attcml to simulate the natural situation (discussed in some detail the next chapter) by attempting to reproduce in tl classroom some of the natural language expcricnccs whit occur outside the classroom. Exposurc to good natil speech, role playing and gamcs arc somc of the dcvicl employed to allow for the natural self-discovery of languat and its use. Still, it is the tcachcr who plans and contrc such activities. There are other characteristics of the planned classroo situation which distinguish it from thc natural situatiol Thcsc includc social adjustment to group proccsscs (ind viduals must subordinate their behaviour and follo* classroom proccdurcs for thc bcncfit of all), the 11ccd t attend class in order to lcarn, the nccd for long periods ( conccntration and, whcn rcquircd, having to do how study. As far as language is conccrncd, the explicit tc;lchin of granlnlatical structi~rcs and rules niay be involvcc dcpcnding on thc method used. Using books and takin notes arc often cxpcctcd of the student. Students have to gc used to learning language as a acadc~nicsul$cct. Thus, whc considcring overall the demands of the classroom situntio~i it is clear that the oldcr one is, the hcttcr one is able to ac!ju~ and function within th;lt siti~;ition.Yoi111gchilcirc~~ will cluitc poorly in c o ~ n p r i s o nto oldcr childrcn ; ~ n d; ~ d i ~ l t s (



Whether the classroom is in a school that is in the community where the second language is spoken is a matter i of some importance, for this will determine whether ' students will have access to a natural situation outside of the class and thereby may supplement their classroom learning. Thus, for example, Pakistanis learning English in a classroom in London will have beneficial language experiences outside of the classroom that Pakistanis learning English in a classroom in Karachi will not. The former (learning English in Britain) is an English as a Second Language (ESL) context while the latter (learning English in Karachi) is an English as a Foreign Language (EFL) context. Because the ESL context provides more language-learning opportunities for the second-language learner through exposure to natural situations outside the classroom, such learners, unsurprisingly, will generally progress more rapidly than learners living in an EFL context. Furthermore, in comparing children and adults, we may say that, given that the natural situation benefits children more than adults, the ESL context will benefit children more than it will adults. O f course, the ESL context will benefit adults too, but to a lesser degree. A summary of the social situational factors, natural and classroom, and their functioning according to age is shown \ on the right sidc of Table 10.1. Thus, for example, whilc childrcn under thc age of 7 ycars are rated 'high' concerning the qi~ality and quantity of language interactions in the , natural situation, thcy arc ratcd 'low' on adjustment and skills in t l ~ cclassroonl situation. Thc opposite is the case for the adults who ratc 'low' in the natural situation but 'high' in the classroo~nsituation. I



I



10.4 Who is better?



13ccausc t l ~ canswer to this qi~cstiondcpcnds o n whether we ;Ire dealing wit11 the natural or the classroom situation, each situntion n ~ u s tbc considered scpnratcly in relation to the psycllologicnl f;lctors wllich ;lffcct tllc I c a r ~ ~ of i ~ Iangungc. ~g With tllis nppronch, we sh;lll then be able to rc:~chsomc co~~clusions i n cornpilring the ;~chicvcnlcntsof childrcn ; ~ n d ;~dultsi n scco1ld-I;111gu;1gclearning. The summary data



214 Second language



presented in Table 10.1 will serve to make it easier to dra conclusions in this regard. 10.4.1 The natural situation



In the natural situation, younger children will do best. First all, the natural situation is more favourable to childr because adults undergo a marked decline in the quality a1 quantity of the social interaction conducive to go( language learning. There is no question that, in a natu situation, the social activities o f children, especially you. children, expose them to massive amounts o f good, natu language. This does not occur for adults, and, in many ca: not even for older children. In extreme cases, members these groups may even find themselves in social conditio which are hostile - conditions which discourage secor: language learning. Still, the older child will have advantage over the adult. Psychologically, while both childrcn and adults ha optimal powers o f induction, and are able to inducc t grammar of a second language more o r lcss equally wc nonetheless, it will be easier for children to Icarn synt than it will be for adults. This is because adults undcrgc decline in memory, and, without rcmcmbcrcd data, therc nothing to analyse. Adults and cvcn oldcr childrcn longer have thc formidable powers of rote Icarning tl young children do. Although adults may devise mcmc strategies and can seek out nlorc practice, ncvcrthclcss, tl places an additional burden on them, onc that the child dc not have. Therefore, children, particularly younger ch drcn, will have an advantage ovcr adults in learning t grammar of a sccond languagc. For the same reason, ola childrcn can bc cxpcctcd to 1car11faster than adults, bccau of a bcttcr 111cnlory. As we agc and as our ability to acquire new mot skills dcclincs, our ability to conlmnnd our o r p n s of spec to carry out the new movcmcnts of n sccond I;lngu;~gc negatively affected. Thcrcforc, bccausc c h i l d r c ~possess ~ tl flexibility in nlotor skills which adults d o not hnvc, childrc will d o much better in nccluiri~lg11ntivc pro~luncintio~l in sccond Inngungc. Althoi~gliothcr Cictors may be involved speaking likc ;I nntivc, no~~ctlic*lcss, motor skills is u11doul



I



Children vs adults in second-language acquisition 215



,



tedly the major contributing factor. Again, as with the problem o f memory, adults can attempt t o overcome this deficit, such as through practice. But again, as with memory, that adds an additional step to the process and makes learning more difficult for them. Thus, it can be concluded that in all respects o f language learning, for the natural situation, children will do better than adults, with younger children doing better than older children.



; 1 I ,



1



i 1 i



'



1 ('



i



10.4.2



The classroom situation



In the classroom situation, adults will do better than young children, because, not only are they better in explicative processing but, simply put, they k n o w h o w to be students. They have sufficient maturity to meet the rigours o f a formal learning environment, where concentration, attention and even the ability to sit still for a long time, all play a role in learning. Because the older child's memory and motor skills are better than the adult's, the advantage in explicative processing enjoyed by the adult may not be sufficient to overcomc thc disadvantages experienced in these areas. Thus, an oldcr child will probably d o better than an adult in thc classroom situation. T h c best age to learn a second language in a classroom situation is probably that age where thc individual retains much o f the memory and motor skills of thc vcry young, but whcrc thc individual has bcgun to reason and understand likc an adult. That agc would probably be somcwhcre around 12 years.



I I



10.5



Critical age



Once hcforc, in a diffcrcnt co~ltcxt (Chapter 3 on Wild Children), the c o ~ ~ c c poft a critical age for first-language Ic;lrning was discussed. Rccall, if you will, that altl~ough cvidcncc was lacking, some tl~corists hypothcsizcd that thcrc was a11 ;1ge (puberty, for cxnmplc) h c y o ~ ~which d it would hc impossihlc t o ;~ccluirca first I;~nguagc. I3r;li11 cl1:lngcs were suggested as ;I possible cxpl;lnntion for such a psycllolo~:icnl barrier.



216 Second language



, , I



I 1'1



I



I



I



1 I



1



1



! I



Children vs adults in second-language acquisition 217



It is reasonable to ask the same question about thc acquisition of a second language. Is there any barrier to th learning of a second language and, if so, at what age doe this barrier become operational? As far as adults learning second language is concerned, we have the commo observation that a very great number of adult speakers dc in fact, learn the syntax of other languages perfectly. Ther are those who speak second languages so well that, on th basis of the grammar alone (not the pronunciation, whic we shall deal with shortly), they would be judged natik speakers. There is no evidence, for example, that a speak6 of a Subject + Verb Object language such as Englis cannot learn a different word ordering of sentence cor stituents, such as the Subject Object + Verb ordering th: occurs in Japanese. O r that negating a sentence by changin a word internally, such as is done in Turkish, preseni insuperable problems to speakers who negate sentences i their native language with unattached negative marker. such as 'not'. Complicated systems of grammatical case such as occur in Russian and Finnish, can bc learned by normal adult Chinese (whose language, like English, has n cases) who is willing to devote the time to learning them. is safe to affirm the view that there is no critical agc in tcrm of acquiring thc syntax of a second language. This brings us to pronunciation. Is it possible to Icarn second languagc so wcll that one truly sounds likc a nativ speaker? One psycholinguist, Thomas Scovcl, rcccntly ha claimed that no adult can ever be successf~~l in that rcgard 'Thc critical period for accentless spcech simply mcans tha adults will ncvcr lcarn to pass themselves off as nativf spcakcrs phonologically. . . .' He describes this as 'Thl Joscph Conrad Syndrome', after thc famous novelist an( nlastcr of English prosc, who, a nativc spcakcr of I'olish did not cvcn bcgin to study English until hc was 20. Scovc has in mind a certain category of adult sccond-languagc spcakcrs: thosc who have mastcrcd thc granimatical and co~nmunicativccomplcxitics of another langi~agcbut still spcak with an accent. Wc could add forn~crUS Sccrctarics of Statc, Hcnry Icissingcr (who lcft C;crlnany and wcnt to thc US a t age 14) and Zbignicw 13rzczinski (who lcft his nativc I'oland and wcnt to Canada a t agc 10) to this class. While 1 would agree with Scovcl thnt in sccol~d-



+



(



i



language acquisition there is not a critical age for syntax, I cannot agree that there is an absolute critical age for pronunciation. There is a critical age for most people, but not for all. Based on my own personal experience and the observations of others, I believe that there are persons whose pronunciation can pass as native speakers in a second language, a language which they have learned as adults. However, only if documented cases are provided can Scovel's absolute critical age notion be disproved.



10.6 Discussion questions



+



1.



'



2. 3.



1



4.



i



5.



,



! 6. I



7.



Can you think of evidence in your own life that your memory ability and your ability to acquire new motor skills are not as good as when you were a pre-teen? Why is memory so important for language learning? Are children better second-language learners in the natural situation? Can adults be better second-language learners than younger children? Do think thcre is a critical age for second-language acquisition? Consider the cases of both syntax and pronunciation. Why, concerning thc classroom situation, is an English as a S C C O I -Languagc I~ (ESL) situation morc advantagcous to second-languagc learners than an English as a Forcign Languagc (EFL) situation? 110 you personally know any counter-cvidencc to Scovcl's claim that no adult can learn thc pronunciation of a second languagc wcll enough to pass as a nativc spcakcr?



Suggested readings I)ulny, I-I., Ilurt, M. and I o11 crcativity tchsts. 111 111;llly



3



cases, their mathematics and science scores were als higher. S i m i l w research has strengthened these findings. t h e r e is some question, however, as to the validity o f these findings. Principally, this concerns whether the parents o e children may have in some way affected the outcome. ,For, even when the pdrents o f the monolinguals and bilinguals are matched socio-economically and educationally, n o control is allowed for the attitude and motivation o f the parents regarding bilingualism3 Perhaps parents w h o wish to be involved in a bilingual programme somehow would tend to provide a more advantageous home environment for their children, intellectually and linguistically, than would other parents. There is no way this issue can be resolved unless researchers are allowed to randomly assign children to monolingual o r bilingual programmes regardless o f the wishes o f their parents.



12.2.2



I



Effects on the development of intelligence



Docs Icarning a sccond language at an carly agc, while the child is still in the process o f acquiring the native or first language, 11:lvc n negative cffcct on a child's intclligc~lcc, thinking ability, crcativity o r cognitive arcas such as mathematics? Somchow, thc burden o f Icarning an additional Iangi~ngcis co~isidcrcd to havc an adverse cffcct o11 the child's :~hilitics.As was the casc in considering cffccts on the dcvclopmcnt of I:ingi~agc,most c;lrly rcscarch tended to find n ~lcgativccffcc-t. 7'hc possibility t11;lt Icarning a sccond I;lngt~;lgecoi~ld i l l sonic wny havc n positivc cffcct o n



I



-



ga



X



2g-g 5



.0- uc rfnl a f l * a c 0 sr=*= c EwC.- 9 L(



M c.2 2 -c 2ua4 2 2 $ 5 w .2 m



5



bl:



m



.S



0



e S a3 7



o g c c



&



0 0



E



r= 3 g;> . $2 b o



fle a a



a



o



5 s : ~ 2 e8'0&.-E



-



m & $



g



b



age



o



E



2 o g o Q)='



2.2 2.; a



at+,



2



0.g



0 Q)



agg



Sgu



.-C =0z



a



> a



t: g - d a X , a



g2



A 3



2



X



Yfl 3



0 rd



a



'6, ~ Q & P



5 3 m uIJ , fl



o 5 L "



a m o c



S! 3 0 Li.=



250 Second language



their part. Broader measures o f cognitive abilities along with later age follow-ups is what is needed. \



Conclusion regarding effect on intelligence



Thus, w e find that there is n o evidence that early bilingualism will harm the intellectual o r cognitive development of the child in any way. N o t only that, but there is evidence, although not strong, that it may perhaps even benefit the child intellectually. As for future research, given the view I expressed earlier (in Chapter 1) that thought and intellectual processing develop independently of language, I do not expect that researchers will ever be able to demonstrate that language affects intelligence in any important way. Unfortunately, because editors o f scientific journals tend not to accept (for publication) research studies that d o not result in statistically significant differences o r correlations (called 'negativc results' by experimentalists), substantial experimental cvidencc that would favour this view is not likely to be forthcoming.



I



1 11



:I .I



\



1



1 I



12.2.3



I



I



Conclusion regarding effects of early bilingualism on language and intelligence



A consideration o f the research evidence shows no harnifiil 'effects eithcr regarding language (first o r sccond) or i n t ~ l l i ~ c n c eIn! fact, some research suggcsts tllcrc may cvcn be beneficial cffccts. f ~ i v c nthe advantages of knowing anothcr languagc and of young children's propensity for spccdy languagc acquisition, wc must concludc that thcrc is cvcry rcason to favour early bilingualism.



,



'



12.3. Simultaneous and sequential learning situations



Tlicrc arc csscntinlly two conditions according to wliicl~a pcrson may bccomc bilingual: tlic two Ianguagcs can bc accliiircd at the same time or in scclucncc. Thc simultnnco\is Icnrning of t w o Inngungcs occurs only with childrcn, sillcct



Bilingualism and cognition 251



in only the most abnormal o f circumstances would a child when exposed t o a language not learn it. O n the other hand, sequential learning can occur with both children and adults; the second language can be learned in childhood o r after the person has become an adult. 12.3.1



Simultaneous acquisition



There are t w o situations in which a child may acquire more than one language at the same time. O n e is when speakers of the different languages use only one language each when talking to the child. For example, a mother might speak only Spanish while the father might speak only English. T h e one parent using one language only situation ('1P-1L'). T h e other situation is when the same person uses t w o different languages while speaking to the child. For cxample, the mother and father use both Spanish and English when talking to the child. T h e one parent using t w o languagcs situation ('1P-2L'). It secms that children are so flexible that they can bccomc bilingual in both languages by the age o f 3 or 4 ycars, rcgardless of their language situation. Although cvidcncc bearing on this issue is not available, it seems likely that thc child in thc 1P-1L situation will learn the t w o languagcs fastcr than thc child in thc 1P-2L situation simply bccausc o f consistcncy. O n hcari~lgsome speech, the child would not havc to puzzlc ovcr just which o f the t w o scts o f language knowlcdge is to bc rcfcrrcd to. I d o not think, however, that the diffcrcnce in spced o f acquisition bctwccn the 11'-1 L 2nd 11'-2L situations would bc great; it would be only a ninttcr o f months, pcrhaps, bcforc the 11'-2L child would cvc~itiially sort it 311 out. O n tllc other hand, it may be that 11'-2L childrcn product more miscd Inngungc scntcnccs (the view o f McL;~ughlin),wlicrc vocabulnry and syntax of tllc diffcrcnt Inllgungcs nrc used in the s;111ic sc~itc~lcc, c.g. 'Ope11 thc rc,izoko' (wlicrc rc~izoko is rcfi-igcrator in Jnpuncsc). O f coursc, i l l tinic the cliild will ovc,rconic thc, nlisccl inpiit and get things right. Ovcrnll, it w o i ~ l dsccln th;lt tlic 11'-1 L iitu,itio~iis hctrcr si11cc ;ic'clliisitio~~ Iiiny hc fastcr n11d less 111ixi11g111ij:Iit oc-cllr.



252 Second language



One of my favourite 1P-2L examples which I like to cite involves a friend who lives in the US. (The example is mentioned in another context in Chapter 8.) My friend's wife is a Japanese speaker, while he is an English speaker, although they know each other's language. From the time of the birth of their two sons, who are about three years apart in age, the mother spoke Japanese and the father spoke English to the children. T o add to the situation, there was the live-in grandmother who spoke only Russian to the children. The result? By the age of 3, each of the boys, in turn, became trilingual in English, Japanese and Russian. Because the family lived in an English-speaking neighbourhood (in Honolulu), English is what they used when they went outside to play. Whether the bilingual child must be conscious of the existence of two languages in order to make progress in acquiring them is virtually impossible to determine. One cannot interrogate a 1- or 2-year-old on such a matter. However, once the child has acquired some degree of fluency in languagc, cven very young children may indicate an awareness of knowing two languages. (For a considcration of the issue of consciousness in languagc learning, see Schmidt, 1990.) One Estonian-English bilingual child was only 2 years old when he exprcssed embarrassment on becoming aware that he had spoken in Estonian to his English monolingual cousin (Vihman, 1982). And, De Villicrs and De Villicrs (1978) citc hearing a 4-year-old say, 'I can spcak Hcbrcw and English', to which his 5-year-old American fricnd responded with 'What's English?'! I'crl~aps the monolingual English-speaking child nccdcd somc knowledge of two langi~agcsbefore hc would understand what a language is. Likewise, if a pcrson saw things only in red (through rose-colourcd glasses), you would not cxpcct that pcrson to be able to undcrstnnd what is meant by colour. A contrast is csscntial for such n rcnlizntion. (Relating to this point is the dolphin discussion in Chapter 3. 13ccausc the dolphins hnd lcnrncd Innguagc only in the impcrntivc format (13ring this, Tnkc that, ctc.), and bccnusc thcy knew n o other contrnsting sy11t;lctic forms (dcclnrntivc, question, ctc.), i t was nrgucd t1i;lt the dolphi~lscoulti not k11ow the synt;~cticforrn of the impcr;ltivc.) I$cforc closing this scction, I would ndvisc parents (or



Bilingualism and cognition 253



potential parents) who have bilingual abilities and have decided to raise their children in a bilingual situation to use the languages in the 1P-1L fashion right from the start. It is inadvisable to wait until the child is 1 or 2 years of age before introducing the second language. If such parents wait, they may not be able to carry out their plan. For, in the course of a year or two their use of one language will become so set that they will experience great psychological difficulty in changing to another. When a social relationship has been established with the child by means of one language right from birth, it will not be easy to switch to a different language. 12.3.2 Sequential acquisition



The sequential kind of bilingual situation occurs for a child when parents speak one language and the community at large speaks another. The parents could be immigrants, foreign residents or simply people who have moved from one part of a country to another part (from Englishspeaking Toronto to French-speaking Quebec City). The parents speak one language at home, which is different from the one which the child is exposed to outside the home. Clearly, sequential acquisition of the second language may take place at a variety of ages and under a variety of situations. Consider, for example, an immigrant couple to America who come from China. They spcak Chinese at homc with their infant daughter. Then at the age of 3, the child starts attending an English-speaking nursery school. By the time the child is 4, she will be speaking English with her .playmates and othcrs while continuing to spcak Chincsc at homc with hcr parents. Thus, thc child learns two languagcs scqucntially, i.c. where a sccond language is introduccd after the first languagc has started being learned. O f coursc, somc of the acquisition of the two languages will occur si~~~ultancously. What is scqucntial is the different starting times for the introduction of each language. In acquiring the sccond language, spccd, proficiency and fluency will be d c t c r ~ ~ ~ by i~~ certain c d psychologicnl and social vnriablcs. A detailed discussion of such varinblcs is



254 Second language



Bilingualism and cognition 255



presented in Chapter 8, where child and adult acquisition of a second language are compared.



Polish, respectively). In other respects, they excel in the English language. Good syntax with good vocabulary is the best duo combination for language success.



12.4 Transfer effect of L1 on L2 learning



12.4.2



12.4.1



Similarity of syntax, vocabulary and sound system



While linguists agree that no one language is more complex overall than any other language, and psycholinguists agree that no one language is easier to learn than any other language, nonetheless, not every pair of languages can be expected to be acquired at the same rate. For example, after having learned English, learning French will not be as difficult as learning Japanese. There are differences between English and French syntax but these differences are small in comparison to the monumental differences between the syntax of English and Japanese. Furthermore, there are significant similarities between English and Frcnch in terms of vocabulary. A learner would not be starting at zero as he or she would if learning Japanese. The Japanese spcakcr learning English is placed in a comparable position. Yet, acquisition of Korean by a Japancsc spcaker is rathcr casy because those syntaxes are vcry similar. O n thc other hand, English spcakcrs would not havc as much trouble with Chinese syntax as they would with Japanese syntax because Chinesc syntax is morc similar to English. Yet, whcn it comes to thc sound system, Chincsc, with its toncs, is so diffcrcnt from English and Japancsc that the English spcaker will find thc Japancsc sound systcni casicr to handle. iUd ~ u s ,wc may concludc that the grcatcr the similarity bctwccn two languages in tcrnls of their syntax, vocabulary and sound system, thc 1iiorc rapid thc rate of acquisition in thc two languages,@ If wc had to scale the importance of thcsc variables, ~ ~ j d o u b t c d l ywc would give syntax thc greater wcight. Good pronunciation or a wide vocabulary can liardly compcnsatc for poor syntax. 111 this rcgard it is intcrcsting to note that in its rccclit history Amcrica h;ls h;td two forcign-born Sccrctarics of Statc, I, I).,I I5



Coltheart, M., 200 Connolly, J., 113 Conrad, R., 90 Cook, V., 115 Coren, S., 174 Cromcr, R., 143, 156 Curtiss, S., 66



Hakes, D., 127 Hakuta, K., 245,249,258-9 Halle, M., 114 Halpern, D., 174 Hayes, C., 31, 47 Hayes, H., 31 Heider, E., 165 Herman, L., 38-9, 46-7 Herodotus, 48-9 Heschl, R., 182 Hockett, C., 47 Horrocks, G., 115 Humboldt, W. von, 159, 173 Huttenlochcr, J., 18



Dalgarno, G., 86, 90 Dechert, H., 132 Descartes, R., 55, 136-7, 152, 154-5 dc Villicrs, J. & de Villiers, P., 28, 252, 258 Diamond, M., 178 Dik, S., 113 Drostc, F., 115 Dulay, H., 10, 28, 217 Duran, R., 259 Einstein, A , , 5, 18 Engclniann, S., 245 Euclid, 134, 151 Ewing, A. & Ewing, E. C., 90 Exncr, S., 182



Filln~orc,C . , 110-1 1, 113, 115 Fodor, J., 127, 132, 100 Foss, I)., 127 Fouts, 1). & Fol~ts,It., 32-3, 35 Fric~,A , , 24 1 Frics. (:., 228, 241 Frilkow, E., 01



:



varictics, 242-3 bilirigualisrn and intclligcncc ncgativc rcports, 248 positivc reports, 248-50 bilingualism and language ncgativc rcports, 245-6 positive rcports, 246-7 Binding Thcory, 110 hirds and spccch, 41 hr:iin ft~nction,174-7 Imin strl~cturc,174-7 Ilritish Sign L : I I I ~ I I(I%SL), : I ~ ~ OH, 77



I%ritishS i p L:~ngil:igcvcrhs, 77 1lroc;l's aphnsin. I8(+8



Broca's Area, 176, 180-4, 187-8, 192-3, 195, 197



Brzezinski, Z., 216 Bush, President, 175 Carla (second language brain location), 197 cerebral cortex (cerebrum), 175 children vs. adults in second language acquisition, 213-15 chimpanzees, 31-8 Chimpsky, Nim, 30, 34-5 Chomsky as an anti-mentalist, 129-31 competence and performance, 96-7



effect o n teaching methods, 228-9



Government and Binding (GB) Theory, 107-10 grammars, 97-1 13 linguistics as psychology, 95-7 meaning, sound and syntax relations, 116-17 Standard Theory, 98-107 Syntactic Structures, 97 systcni of rules, 97 circle idea, 13.%5 classrooni situation, 21 1-13, 215 clinib (Chonisky's innatc idcn), 138-0



Clinton, I'rcsidcnt, 175 Cognitive (;r;inlmnr, 113 color tcrms, 165 < : o n ~ m ~ ~ ~ ~ iLangu;lgc c:~tivc Tc:lching (CL'T'O, 2.7 1 , 233, 240



Subject lndex 265



264 Subject lndex



, I



,I



'; I



i



I



i I



'



,



competence-performance distinction, 96-7 complex sentence production, 14-16 Computerized Axial Tomography (CAT), 195-8 Conrad, Joseph Syndrome, 216 correction, 26-7 Creoles and thought incompatibility, 168-9 critical age for first language acquisition, 63-5 critical age for language learning and brain maturation, 184-6 critical age for second language acquisition, 215-17 deaf children without language have thought, 166 deaf education, 80-9 Deep Structure (and D-structure), 100-5, 111-13 determinism, 198 dimensions of teaching methods, 219-21 Direct Method, 225-6 dolphins, 37-40 dominance of brain hemispheres, 177-81 Donald (raised with Vicki), 31 Double-Trouble phenomenon, 257 dyslcxias, 19C-1 earcdncss (preference), 179-80 casc and spccd of child acquisition arguments, 140-2 clcctrical stimulation of brain, 195 Elvar (dolphin), 37 Empiricist I'hilosophy, 135*, 152-3, 171 error analysis, 256 Eskimo (spcv Inuit) ESL vs. ESL Icnrning context, 213 cxplicntion, 203-5 facilitation in sccontl I;111gi1:1ge Icnrning, 255 finger spelling, 74-0 First Llngu;~gcstr;ltcgy, 257 Flctclicr, L. (;lctrcss), X I



free will, 198 French Sign Language, 68



Japanese language and aphasia, 191 justice (Plato's innate idea), 137



General Multi-Purpose Learning Strategies (GMPLS), 136 Generative Semantics Grammer, 110-13 Genie (raised in solitary confinement), 55-8,61, 63-6, 185 gestures with speech, 72-3 without speech, 69-72 God (Descartes' innate idea), 137 gorilla (see Kolo) Government and Binding (GB) Theory, 107-10 Grammar Dimension: Induction-Explication, 220 Grammar-Translation (GT) Method, 219-20,222-4, 239-40 Guerin, Madame, 52, 55



Kanzi (see pygmy chimp) Keller, Helen, 53, 61-5, 167, 185 Kissinger, H . , 216 Koko (gorilla), 36-7, 43, 149 Lana (chimp), 33, 43, 44 language affecting thought, 169-70 language and thought, true relationship, 171 language areas of brain, 181-4 language as basis of thought, 16% arguments against, 166-9 language disorders (see aphasias) Language Mode Dimension: Speech-Reading, 219-20 lateralization and brain hemispheres, 175, 177-81, 185, 197-8 Leborne (Tau), 196 left-handers, 174-9 Leonardo da Vinci, 175 lexicon, 99-103, 112-13 linguistics as psychology, 95-6 Linguistics Dimension: MentalistStructuralist, 221 lipreading (see speechreading) lobes of brain, 177 localisni and holism, 191-2 Logical Form, 108-9, 111 Loulis (chimp), 32-3



handedness, 174-9 hemispheres of brain, 175-81 holophrastic speech, 7 Hopi language, time conccpt issuc, 164 Hypothesis 1 vs. Hypothcsis 2 argument, 145-9 imitation, role in acquisition, 2 5 4 inadequate data argument, 142-4 induction, 203, 205 innate k~iowlcdgc,136-9 intclligcncc and bilingualisn~, 247-50 intclligcncc and language, 245-7 intclligcncc irrclcvant argument, 149-50 intclligcncc vs. Univcrsal C;ranini;~r,130-55 intcrfcrcncc in second Innguagc Icnrning. 255 Inuit (Eskimo), 165 1s:tbcllc (confined with n ~ u t c motllcr), 50-00, 03-5



c



rnathcmatics argument, 150-2 ~ncaning,non-linguistic origins, 170-1 memory role in languagc acquisition, 1 206-8 ' in first Iangungc acquisition, 21 ~ncntalism.153-5 Mcycr, It. (;lrrcstcd schooltcnchcr), ! 158-0, 244 mi11d ant1 hr:~in,108-0 nlinutc s;~mplc :rrgtlnlcnt, 144 Monitor I lyyothcsis, 234, 730 ~llonkcys(vcrvct), 41 ~llorl>hcnlc :~ccluisitionortlcr, 0-14



:



Motherese (see Parentese) motor skills and pronunciation, 208-9 Natural Approach (NA), 233-6 Natural Method, 224-5 natural situation, 210-1 1, 214-15 Nebraska law against second language teaching, 158-9 negation and production, 15-16 N i m (see Chimpsky, Nim) one-word utterance, 4-7 Oral Approach, 82-4 paralysis and thinking, 162 parameters, 110, 138 Parentese (Motherese), 22-3 perfection (Descartes' innate idea), 137 performance models Chomsky's, 117-19 types, 119-23 Perot, Ross, 175 Phoenix (dolphin), 38 Phonetic Interpretation (and Representation), 1 0 5 4 , 111 Phonological Rules, 105-6 Phrase Structure Rules, 99-103 Positron Emission Tomography (PET), 195-8 post-mortcm examination of brain, 194-5 poverty of stimulus argument, 144-9 Process Graniniar performance modcl, 122-3 I'rojection Principle, 109 I'salmtik I of Egypt, 49 psychological reality, 1 16-31 psychological unreality of Chomsky's grammar, 128-9 I'sychology f>imcnsion: Mcntnlist13chavio~1rist. 221 pygniy chimp (Ki~nzi),4.3-5 l'yth;lgorns, 134 Ilation;~listI'hilosophy, 135-7, 1 53-3 rcading process in brain, 183, 107 reason 2nd Rationalist thcory, 137



CURTIN UN~VERS~TY OF TECHNOLOGY



L I B R A R Y ,



266 Subject Index Resource Grammar performance model, 119-22 Rochester Method, 75



speechreading, 83 split brain, 180-1 Standard Theory, 98-107 Suggestopedia, 237-8 Surface Structure (and S-structur 103-5, 111-13 Syntactic Structures, 97



Sarah (chimp), 35-6, 43 schizoids and bilinguals, 167-8 second language acquisition factors, 203 talking and thinking second language locations in brain, simultaneously, 162-3 197-8 Tarzan, 49 Second Language strategy, 257 Tau (see Leborne) Seeing Exact English, 75 telegraphic speech, 8-9 Semantic Interpretation (and thought as the basis of language, Representation), 106-7, 111 170-1 Semantic Rules, 1 0 6 7 , 111 Total Communication, 84 sentence production process, 123-6 Total Physical Response (TPR), sentence understanding process, 229-3 1,240 12transfer effects of Language 1 on sequential bilingual learning Language 2,254-7 situation, 250-1, 253-4 Transformational Rules, 103-5, sex hormones and hemispheres, 110-1 1 178-9 triangle (Leibnitz' innate idea), 12 sign language two and three-word utterances, as a complete language, 67-8 7-1 4 independent type, 76-82 representing spelling or speech, U.S. Supreme Court, 159, 173 74-6 understanding precedes syntax, 78-80 production, 160-1 types, 7 . 4 Universal Grammar (UG) (also sign language aphasia, 193-4 LAD and Languge Faculty), Signing Exact English, 75, 78 137-9 Silent Way, 236-7 arguments against, 137-52 sirn~~ltancous bilingual learning situation, 250-3 Vicki (chimp), 31, 43 snow words. 165 Victor (Wild Iloy of Avc?ron), social situations in second language 5G9, 61, 6.341, 89 acqi~isition,21(&13 vision fields, 178 speech acts, 7 vocnlization (dcvclopmcnt), .3-4 speech areas of brain, 1x3 spcccll as ba.;is ofthought, 159-160 Washoc (chinlp), 31-3, 43, 46, 14' Wcrnickc's aplinsi;~,1HH argumcnt~against, 160-3 speech product~on( d c v e l o p ~ n c ~ ~ t ) , Wcrnickc's Area, 176, 180-4, 186, 188, 101-4, 107 -3- 10 Wild I3oy o f Avcyron (sc~cfVictor) syccch understanding, 16-21 a.; has~so f trr Written Laligu;~gcApp~.onch,8 4 ' Ycrkish, 33



aa-q



rn



'> I1;