Setting the Narrative Scene: How Children Begin to Tell a Story

Volume Children’s Developing Narrative and Discourse Competence This page intentionally left blank Volume Children’s la n g Developing Narrative and Discourse Competence E d ited by KEITH E. N E L S O N Pennsylvania State University, University Park A Y H A N AKS U-KO<J Bogazigi University, Istanbul, Turkey CAROLYN E. J O H N S O N University o f British Columbia , Vancouver, Canada J Routledge ^ laylor & I rands Group New York London The final camera copy for this work was prepared by the editors and therefore the publisher takes no responsibility for consistency or correctness of typographical style. However, this arrangement helps to make publication o f this kind of scholarship possible. Copyright © 20 0 1 by Lawrence Erlbaum Associates, Inc. All rights reserved. No part of this book may be reproduced in any form, by photostat, microform, retrieval system, or any other means, without the prior written permission o f the publisher. First published by Lawrence Erlbaum Associates, Inc., Publishers 10 Industrial Avenue Mahwah, New Jersey 07430 Reprinted 2009 by R outledge Routledge 270 M adison A venue N e w York, N Y 10016 2 Park Square, M ilton Park A bingdon, O xon O X 14 4 R N , U K Cover design by Kathryn Houghtaling Lacey Library of Congress Cataloging-in-Publication Data isbn: issn: 0-8058-3292-0 0163-2809 10 9 8 7 6 5 4 3 2 1 Contents C ontributors VI I Preface IX Introduction AY H A N A K S U - K O £ , K E I T H E. N E L S O N , A N D C A R O L Y N E. J O H N S O N XI 1. Setting the N arrative Scene: H ow Children Begin to Tell a Story R U T H A. B E R M A N 1 2. Representation o f M ovem ent in European Protugese: A Study o f C hildren’s N arratives H A N N A J A K U B O W I C Z BAT OR ^O A N D I S A B E L H U B FARI A 31 3. W hy Young A m erican English-Speaking Children Confuse A nger and Sadness: A Study o f G ram m ar in Practice MICHAEL BAMBERG 55 4. A C rosscultural Investigation o f Australian and Israeli Parents’ N arrative Interactions With T heir Children G I L L I A N W I G G L E S W O R T H AND ANAT STAVANS 73 5. The Acquisition o f Polite Language by Japanese Children KEIKO NAKAMURA 93 6. Interactional Processes in the O rigins o f the E xplaining C apacity E DY V E N E Z I A N O 113 C h ild ren s A ttributions o f Pragm atic Intentions and Early Literacy KENNETH REEDER 143 A uthor Index 165 Subject Index 171 Contributors AY H A N A K S U- K O£ Bogazigi University, Istanbul, Turkey R U T H A. B E R M A N Tel Aviv University, Israel MICHAEL BAMBERG Clark University, Worcester, Massachussetts H A N N A j A K U B O W I C Z BATOR^O Open University, Lisbon, Portugal I S AB E L H U B FARI A University o f Lisbon, Lisbon, Portugal C A R O L Y N E. J O H N S O N University o f British Columbia, Vancouver, Canada KEIKO NAKAMURA Keio University, Tokyo, Japan K E I T H E. N E L S O N Pennsylvania State University, University Park KENNETH REEDER University o f British Columbia, Vancouver, Canada ANAT STAVANS Hebrew University and Bjeit Berl College, Israel E DY V E N E Z I A N O Universite Nancy and Universite Paris V-CNRS GILLIAN WIGGLESWORTH Macquarie University and University o f Melbourne, Australia This page intentionally left blank Preface The chapters in this volum e deal with discourse developm ent, w ith an em phasis on narrative, from ages 1 1 / 2 to 10 , and ranging over seven languages. T h ey were d evelo p ed fro m 7 o f the 2 76 p resen tatio n s at the Seventh In te rn a tio n a l C o n gress o f the In te rn a tio n a l A sso ciatio n fo r the S tu d y o f C h ild Lan gu age (IA S C L ) in Istanbul, Turkey, in Ju ly 19 9 6 . That m eeting was a b ro a d ly in ter­ national assem bly o f 3 5 0 p articipants from 4 1 coun tries, representing m ore than 30 languages, w h o are c o n trib u tin g to the d evelo p m en t o f a scientific tradition in the fields o f general linguistics, sociolinguistics, psycholinguistics, and in te rcu ltu ral c o m m u n ica tio n , as th ey p erta in to ch ild re n ’s lan gu age acquisition. D urin g the Congress, participants shared exciting research projects and exch anged a b ro a d sp ectru m o f view p o in ts w ith honesty, dep th , and respect. O ur Turkish hosts were outstanding in their w arm th and organization. IA S C L continues to invite international applications from all professionals who share an interest in enhancing the understanding o f ch ildrens acquisition and use o f languages. K E I T H E. N E L S O N AYHAN AKSU-KO CAROLYN 9 E. J O H N S O N ix This page intentionally left blank Introduction AY H A N A K S U - K O f Bogazigi University K E I T H E. N E L S O N Pennsylvania State University C A R O L Y N E. J O H N S O N University o f British Columbia The chapters in this volum e reflect recent directions o f thinking in the area o f c h ild re n ’s d isco u rse d evelo p m en t, w ith an em p h asis on n arratives. E ach contribution shows that em pirical w ork in the last decade has focused on finer distinctions regarding the effects on developm ent o f discourse genres, different elicitation techniques and com m unicative contexts, literacy and schooling and, o f course, age, language, and culture. Each chapter addresses issues concerning the in te rre la tio n s betw een social, cogn itive, and affective cap acities and processes in discourse. Finally, each raises theoretically challenging questions regarding how and when new representations are constructed to su pp ort new co m p lexities in n arrative and d iscou rse m ore generally. A co m p reh en sive theoretical fram e calls for a conceptualization o f discourse as an interactional space th at p ro m o tes the d evelo p m en t o f h igh er level m etalin gu istic, m etarepresentational, and m etapragm atic operations. The chapters by Ruth Berm an, H anna Batoreo and Isabel H ub Faria, G illian W igglesworth and A nat Stavans, and Ken Reeder focus on aspects o f ch ild rens n arrative ability, their p rod u ction s in d ifferent genres, in different m odes o f expression and in relation to the input they get in different cultures. T h ey all look at form :fu n ctio n relations, that is, the interface betw een specific devices w ithin a language and narrative developm ent. Chapters by Edy Veneziano and K eiko N a k a m u ra , on the oth er h and, deal w ith m ore specific d isco u rse capacities such as justifications o f oppositions and use o f politeness indicators as these d evelo p in the m ore gen eral con text o f con versatio n . M ich ael B am b erg’s contribu tion cuts across the two groups, as it deals w ith children’s narrative accounts o f em otion, but from a dynam ic discursive perspective. From a m ethodological point o f view, issues relating to task conditions, levels o f assessment, cultural differences, and the like are brought under consideration, xii a k s u - k o <;, n e l s o n , a n d Jo h n s o n suggesting that some o f the factors typically identified as m ethodological should perhaps be approached from a different perspective: Th ey m ight reveal u n d er­ lying processes or themselves be determinants o f acquisition. INTERFACES BETWEEN LANGUAGE-SPECIFIC FORMS, NARRATIVE FUNCTIONS, GENRES, AND COGNITIVE D E V E L O PM EN T With high appropriateness, the b o ok begins w ith a chapter on children’s ow n beginnings. B erm an (chap. 1) analyzes and discusses how children start their o ral n arratives. She argues that such sto ry b egin n in gs d ep en d h eavily on general cognitive d evelopm ent, and p rovid e insight not on ly into ch ild ren ’s knowledge o f narrative structure, but also on their story planning abilities and their representations o f their listeners and task. The stories exam ined are told by children at ages 3 , 5 , and 9 and adults. N arrative structures, including the specific lin g u istic fo rm s used to express scen e-settin g and m o tiva tio n a l elements, are com pared for narratives based on a wordless picture book (Frog, w here are you?, M ayer, 1 9 6 9 ) and fo r a sto ry con stru cted ab ou t a p erso n al experience (a fight) in two m odes: speech and w riting. These com parisons are supp lem en ted w ith results from an ad d itio n al b o d y o f H ebrew and E nglish data from varied storytelling tasks. Berm an observes age-related developm ents in setting the spatiotem p oral fram e for the story, the am ount o f backgrou n d and m o tivatio n al in fo rm a tio n abou t the characters, the d em arcation o f the setting from the storyline events, m etatextual com m ents relating the narrated events to the storytellin g situation, and the form al m eans used to serve these functions. For exam ple, the proportion o f setting clauses increased with age in the p erso n al co n flict n arratives, b u t n ot in the p ictu reb o o k n arratives. Furtherm ore, use o f specific linguistic form s varied w ith narrative genre and m ode o f elicitatio n , su ggesting the effects o f sch o o lin g and literacy (H icks, 19 9 1; Kuntay, 19 9 7 ; M cCabe & Peterson, 19 9 1a ). Berm an addresses the in trigu ­ ing theoretical question o f why children require a long time after acquiring cer­ tain form s, includin g past tense verbs and tem poral term s such as afterwards, before they use them effectively and flexibly in their dynam ically constructed on -line narratives. The findings overall show that speakers rely on the form al options m ade available to them by the typological structure o f their language, and that these are first effectively exercised at the local clause level, then at the global discourse level (Berm an & Slobin, 19 9 4 ; H ickm ann, 1 9 9 1 ,1 9 9 5 ) . B ato reo and F aria (chap. 2), w h o also investigated ch ild ren ’s sto ry openings, start w ith an interesting review o f h ow narratives can be begun in w ays that utilize the m ost available, m ost reliable features o f a p artic u la r language, in this case, European Portuguese. In accord with the central themes o f this volum e, they exam ine the interfaces between language-specific devices, cognitive developm ent, and the initiation and elaboration o f narratives, using data from preschoolers, school-age children, and adults. T heir m ethods em ploy Introduction xiii o p en -en d ed , textless sto ryb o o k s no w fam ilia r to n arrative d ev elo p m en t research ers w o rld w id e, the C at and H orse stories (H ick m an n , 19 8 2 ) . T h e ir 5 -year-olds proved to have available all the lexical and m orphosyntactic devices used by skillful adult narrators. However, they did not sensitively deploy these devices for the spatiotem poral fram ing o f the stories, introducing the ch arac­ ters, or p roviding new inform ation about them as the narratives proceeded. B y age 10 , the ch ild ren ’s ad ap tation o f lexical and m o rp h o sy n tactic d evices to n arrative p u rp o ses w as m u ch closer to the ad ult m od el, revealin g m ature fo rm :fu n ctio n relations. T he authors affirm the im po rtan ce o f con co m itan t advances in the cognitive capacity to represent the listener in term s o f his or her in fo rm a tio n a l needs and the c ap acity to organ ize events in a go a l-d ire cte d tem poral-causal sequence (Berm an 8c Slobin, 19 9 4 ; H ickm ann, 19 9 5 ; M cCabe 8c Peterson , 1 9 9 1 a ; S lo b in , 19 9 6 ) . O pen fo r fu ture sp ecificatio n are h ow p articular d iscourse-learning contexts and p articular cognitive advances after age 5 contribute to advances in the flexibility and sophistication o f narratives. W igglesw orth and Stavans (chap. 4) deal w ith the less studied question o f the nature o f narrative input children receive from their parents in two different lan guages and cu ltu ra l b ack g ro u n d s— E n g lish -sp ea k in g A u stralian and H ebrew -speaking Israeli— nicely com plem enting B erm an ’s chapter. Age- and la n gu age-related analyses o f h ow the p arents n arrated Frog, w here are you? (cf. Berm an, this volum e) to their children revealed differences in the use o f a variety o f interactive strategies and linguistic acts such as rhetorical questions, personal digressions, affective and evaluative com m ents and the like, in ad d i­ tion to actually advancing the action o f the story (M cCabe 8c Peterson,19 9 1b ; N in io 8c Snow , 1 9 9 6 ; S co llon 8c Sco llo n , 19 8 1) . Fo r 3 -, 5-, and 7 -y e a r-o ld children and their parents, there were clear cultural differences in interpreting the sto rytellin g task. A lth o u gh all o f the sto ry sessions w ere collabo rative to som e degree, the A u stralian E n glish -sp eak in g parents had a m ore co n versa­ tional, dynam ic style, suggesting that they took the task to be an interpersonal exch ange a c tiv ity as w ell as sto ry co n stru ctio n , w h ereas the Israeli p arents displayed a m ore strictly literacy-oriented approach in their storytelling. This difference is paralleled in the perform ance o f the children; Israeli 7-year-olds c o n trib u te d to the actu al sto ry line w ith a lite ra ry -o rie n te d p e rfo rm a n c e , whereas their English-speaking agemates conversed w ith the parents as well as actively ad d in g to the p lo t line. C u ltu ra l styles, thus, have a b e a rin g on the n arratives ch ild ren e ven tu ally p rod u ce, and in p u t is im p o rta n t not o n ly in exposing children to language-specific form :fu nction relations in narrative, but also to c u ltu ra l exp ectation s and specific m odes o f execu tin g the task. W igglesw orth and Stavans also show that despite these differences, universal patterns can be identified in the behavior o f parents o f 5-year-olds w ho, in both gro u p s, d istin c tly em p h asized sto ry stru ctu re and sto ry o rg a n iz a tio n in tellin g th eir ch ild ren stories. N o tin g the critica l status o f 5 years o f age fo r developm ents in d iscou rse-organ izatio n al capacities, as reflected in the shift xiv AKSU-KOq:, NELSON, AND JOHNSON fro m a fragile p lo t-stru ctu re to a h ie ra rc h ic a lly o rgan ized one (B erm a n & Slo b in , 19 9 4 ) , the auth ors suggest that parents in both cu ltu res are eq u ally sensitive to the co g n itive-lin gu istic readiness o f their ch ildren and respo n d accordingly. W hat rem ains to be fu rth er specified is the p articu lar in terp lay between universal and culturally specific determ inants o f behavior. Reeder’s contribution (chap. 7) also has children’s narrative com petence as a m ain focu s, bu t in the w ritten m ode. R eed er investigated sch o ol ch ild re n ’s w riting skills in two genres (story retell and expository description) as a function o f their ability to attribute pragm atic intentions to others. A ttribu tion ability was determ ined by interview ing the children about speaker’s intent after they view ed a p u p p et scenario. R eed er fo u n d that advances in u n d e rsta n d in g p ragm atic intent su p p o rted advances in w ritten n arratives at aro u n d 7 to 8 years in both genres. The effects were stronger for personalized narratives than expository descriptions, which have less interpersonal and contextual support. Skill in identifying the intentions o f speakers requires understanding the rela­ tion betw een the linguistic features o f utterances and aspects o f p articip an ts’ epistem ic states (know ledge, attitudes, assu m p tions), that is, m etapragm atic com petence. As R eeder show s, the effects o f a com plex set o f developm en tal advances such as increased d ifferen tiatio n s o f the feelings, m otives, intents, agreem ents, and conflicts o f varied participants is reflected in increased n arra­ tive sophistication in both spoken and written m odes. Taking the broader view across all the chapters in this volum e, one them e is strongly evident. To wit, new cognitive, social, and language representations are on ly grad u ally woven into flexible, differentiated narrative skills fitted to m any varied contexts and goals. SOCIAL, E M OT I O NA L , AND C O G N I T I V E PR OC ES S ES IN R E L A T I O N TO D I S C O U R S E D E V E L O P M E N T The studies su m m arized earlier, w h ich explore vario u s aspects o f ch ild ren ’s developing narrative abilities, em phasize bidirectional influences, p articu larly between linguistic and cognitive developm ent. The focus o f research reported in the fo llo w in g three ch apters, on the oth er hand, requ ires, in a d d itio n , a consideration o f the interrelations between social and em otional developm ent, because advances in these dom ains also reciprocally feed into and bring about changes in language and cognition. The inquiries into the expression o f anger and sadness (chap. 3), justification o f opposition (chap. 6), and the learning o f form s o f politeness (chap. 3) all place language developm ent w ithin the broader context o f social-com m unicative developm ent whereby the child learns to take up different positions or roles. In ch apter 3 B am b erg takes a d iscu rsive app ro ach to the d evelo pm en t o f em otion, arguing that language use plays a dynam ic, foundational role in the social construction o f em otions such as anger and sadness. He defines em otions as m eaningful positions taken up by a person to signify self-other relationships, Introduction xv perform ed in the form o f discourse actions. His analysis o f the “ gram m ars” o f anger and sadness (i.e., the linguistic construction types that m ark the d iscu r­ sive position s o f being an gry and being sad) highlights the interface between lin g u istic stru ctu res, d iscou rse genres (p erso n al in volvem en t n a rrative vs. e x p la n a to ry d isco u rse) and social p o sitio n s in the d iscou rse o f ch ild ren betw een ages 4 and 1 1. T h e e m p irical data revealin g these p rocesses are children s explanations, for exam ple, “ w hat it m eans to be sad,” and narratives centered on personal experiences o f anger and sadness. The context o f creating a first-person sadness narrative proved problem atic for 4- to 6 -year-olds, w ho did not clearly differentiate attributin g blam e and eliciting sym pathy in their first-p erso n sadness and anger narratives. H ow ever, in their exp lan atio n s o f an ger and sadness, y o u n g e r as w ell as old er (7 to 11 years) ch ild ren clea rly differentiated sadness and anger. B am berg’s account o f the contrasts between the developm ental patterns in the first-person narratives and explanations rests on the children’s developing ability to use construction types to differen tially m ark discursive stances. Successful narratives require the com plex co o rd in a ­ tion o f such linguistic com ponents with other personal, social, em otional, and contextual factors. E m otion talk functions both to inform the child about what em o tio n s are, and h ow th ey are dealt w ith in the so c ia l-e m o tio n a l realm (Eisenberg, 19 8 6 ; M iller, 19 8 6 ; Schieffelin & Ochs, 19 8 6 ). Thus, the language that un fold s in a clear and ap p ro priate narrative reflects the c h ild ’s sense o f h ow s e lf sh ould in terp lay w ith the current listen er-p articip an ts, em o tion alsocial context, and shared expectations o f com m unicative goals. Veneziano (chap. 6), who utilizes longitudinal data from children between ages 11/ 2 and 2 1/2 to explore early developmental stages o f children’s justification and persuasion, also presents a dynam ic approach. Children first learn to be convinced by their m others’ ju stifications; soon after, they themselves becom e effective at p rod ucing justifications and explanations to persuade their m others to give in w ith in a conflictu al situation. M ap p in g rich ly to the them es o f this volu m e, Veneziano’s data and discussion show that children’s justifications o f oppositions reflect their grow ing pragm atic, conversational, and social abilities (Peterson & M cC abe, 19 9 2 ; Reim ann, 19 9 8 ; Snow, 19 8 9 ; Sorsby & M artlew , 19 9 1; Watson, 19 8 9 ). In m anaging oppositions children learn to take up positions vis-a-vis the other, m aking rightful dem ands or requesting concessions, thereby creating for the other a position to be filled. Insisting on their initial intention in contrast with respo n d in g to the o th er’s op p osition , children m ove between self- and othercentered perspectives and acquire com petence in recognizing the other’s m ental states very early on. Discourse acts and self positions thus develop in interaction, via processes very sim ilar to those observed by B am b erg in the case o f em o tion -talk. These ob servation s are tied to an acquisition m odel in w hich co m m u n icatio n advances rest on ch ildren’s concom itan t developm ents in flexib ility o f cogn itive-lin guistic representations and in representations o f their conversational partners. These tw in d evelopm ents feed into narrative xvi A KS U- K O £ , NELSON, AND JO H N S O N conversational sequences in w hich children have the intent to p ersuade— through specific, explanatory, argum entative expressions— another person seen as h olding a different, conflicting position. D ynam ic negotiation is part o f the discourse-m aking process. As necessary, w hen their view points have not been accom modated in prior conversational moves, children demonstrate persistence, redundancy, and/or reform u lations. Partly because there is a delay between children’s increased response to their m others’ explanations and the children’s ow n en try into effective explan atio n -m ak in g, Veneziano argues for a gradual process o f com plex abstractions. The children learn rich, interrelated patterns from maternal verbal justifications: narrative devices, their pragm atic impact, and the emotional and cognitive states o f the m other relative to those o f the child. The significance o f p arent-child interaction for advances in conversational discourse and pragm atic knowledge is also exem plified in chapter 5. Politeness in Japanese ch ild ren’s discou rse is charted d evelo pm en tally by N akam ura, w ho investigated the acquisition o f the different registers b y 1- to 6 -year-olds. She uncovers the child’s increasing sensitivity to the social-contextual variables that underlie the pragm atics o f polite linguistic form s (Andersen, 19 8 6 ; Clancy, 19 8 6 ; E rvin -T rip p , 19 7 9 ) . In b oth sp ontaneou s speech and role p lay in pretense contexts, N akam u ra observes that Japanese children display a “ know h o w ” o f p olite language use as early as 1 year b y u sing nonverbal m eans and verb al fo rm u las in co n textu ally ap p ro p riate ways. Betw een ages 2 and 5 Japanese children progress to produce form s whose use depends on awareness o f in/out­ grou p m em bership, power, age, sex, social setting, and the nature o f the inform ation com m unicated. Japanese children’s control o f politeness form s and their understanding o f the pragm atic rules that govern shifts in register develops grad ually w ith age and w ith exposure to discourse d em onstrating such use. A significant aspect o f this process is again the gradual unfolding o f foi m :function relations, w hereby different ways o f show ing deference are form ally expressed w ith a range o f linguistic form s and conversational strategies. Like em otions, politeness is a dom ain in w hich language functions to inform the child about social-cognitive distinctions. N akam ura suggests that children gradually learn to com preh en d p sych olo gical feelings such as con sideration for others and discernm ent which underlie the use o f different politeness form s in the process o f socialization to the use o f registers. The com plexity o f social understanding and cognition required to su pp ort the m ore advanced levels o f politeness, relates closely to other discussions in this volum e o f how com m unicative developm ent interfaces with related developm ental changes in other dom ains. HO W AND W H E N N E W R E P R E S E N T A T I O N S ARE C O N S T R U C T E D TO S U P P O R T N E W C O M P L E X I T I E S IN D I S C O U R S E This overview shows that the questions raised in the present volum e are m u lti­ faceted, lead in g to the e x p lo ra tio n o f the in terfaces betw een the cogn itive, Introduction xvii social, and lin g u istic skills that feed into p ragm atic com peten ce. W ith out exception, all the contributions invoke the dynam ic processes o f discourse as the d evelo pm en tal con text for the con co m itan t changes that o ccu r in these dom ains. As argued across different chapters such as Veneziano’s, B am b erg’s, and N akam ura’s, early in developm ent, these processes result in a “ kn ow h ow ” o f the m ind that entails a practical u nderstanding o f interpersonal relations. The next level o f representations that em erge later in developm ent involve an u n d e rsta n d in g o f the “ k n o w th at” type. Th ese “ k n o w th a t” rep resen tatio n s support new com plexities o f discourse, w ith narrative as a case in point. Across the d ifferen t c o n trib u tio n s and, in p articu lar, in B erm a n , R eeder, and B am b erg’s chapters, older children’s grasp o f the pragm atics o f interpersonal com m unication are exam ined and further com plex abstractions are proposed as a foundation for success in narrative perform ance. D espite m an y sim ila ritie s across the au th o rs, how ever, it sh o u ld also be recognized that the particular kinds o f representations children construct often are seen variously. For exam ple, B am berg and Veneziano stress em ergent d y ­ nam ic processes that com prise any conversational and narrative perform ance. In th eir respective fram ew o rk s, th ey present self- and c o g n itive-p ro cesses as d eep ly in te rtw in e d w ith p ragm atic, social, in ten tio n al, and lin g u istic rep resen tatio n s. In con trast, oth er au th o rs exam in e ch ild re n ’s in c rea sin g ly co m p le x d iscou rse and n arrative skills after ages 5 to 6 in term s o f in te r­ coo rd in ation s betw een m etarepresen tational and m etalingu istic op eration s, and between these and social-cognitive knowledge at the “ kn ow that” level. The m etarep resen tatio n al c ap acity to reflect on the oth er in term s o f his or her beliefs and in ten tio n s, and the m etalin gu istic skill o f relatin g aspects o f the code to d ifferen t in te rp e rso n a l fu n ctio n s constitu te the co m p o n en ts o f the higher m etapragm atic capacities that can perhaps best be accounted for within a general “ th e o ry o f m in d ” fram ew o rk (A ksu -K o^ & A lici, 1 9 9 9 ; A stin gto n , 19 9 0 ; Nelson, 19 9 6 ; Perez-Leroux, 2 0 0 0 ) . M ETHODOLOGIC AL AND OTHER ISSUES Finally, the chapters in the present volum e point to som e issues em erging from m ethodological considerations. For exam ple, variations in p erform ance due to differences in the nature o f the task, such as m ethod o f elicitation (spontaneous n arratives vs. those su p p o rted w ith p ictures) or con text o f co m m u n ica tio n (role play in pretense activity vs. experim ental situations), are now treated in their ow n right as sources o f inform ation about processes or determ inants o f d evelo p m en t. B erm a n , fo r exam ple, finds that d evelo p m en ts in settin g in fo rm atio n are first elaborated in m ake-believe narratives, then in p ersonal experience accounts, and fin ally in narratives based on p icture bo ok s. T h u s, genre and elicitation m ethod interact and make a difference to the am oun t o f inform ation produced. xviii AKSU-KOg, NELSON, AND JO H N S O N The same observation, furtherm ore, brings to focus the im portance o f the pretend w orld as an assim ilatory zone or a special operational space where new capacities related to sto ry stru c tu rin g are first tried ou t (G arvey, 19 8 6 ; N ic o lo p o u lo u , 19 9 5 ; P iaget, 19 5 1 ; V ygotsky, 19 6 7 ) . N a k a m u ra s ro le-p la y procedure with toys suggests that sophistication in politeness carried by specific gram m atical and pragm atic devices, em erges earlier than m any experim ents reveal. In e x p erim en tal contexts ro le-p la y tasks often place cogn itive and p ersp ective-takin g d em and s on ch ildren that result in less sop h istication in language use than that observed in n atu ralistic discourse. W hat we obtain in experim ental settings, w hich typically put the child in a reflective-interpretive m od e, app ears to be a d ifferen t capacity, the “ m eta” versio n o f the k in d o f com petence that researchers aim to m easure. If taken to represent “ linguistic” instead o f “ m eta lin gu istic” com petence, such find ings attenuate the ages o f em ergence. T h ese m eth o d o lo gica l issues, w h ich u n d erscore the fact that d ifferen t tasks m ay be tap p in g d ifferen t c o g n itive -lin g u istic capacities at different levels, are likely to be taken up in future research and discussions. The context o f play, opening up the possibility for deferred im itation o f adult speech and behavior, is noted to be im portant for the operation o f processes o f acqu isitio n . Research (C lancy, 19 8 5 ; Day, 2 0 0 0 ) p oints again to the role o f im itatio n , albeit in the fo rm o f rep resen tin g a n o th e r’s u tterance, in the developm ent o f pragm atic knowledge. R ole-play speech often entails repetition o f an utterance by preserving its form al properties to a large extent. As noted by N akam ura, repeating dialogues heard from adults in sim ilar contexts, children grasp the underlying linguistic rules, deduce the relevant interpersonal factors, and generalize to other contexts with increasing productivity and appropriate­ ness. These processes m ay lead to insight into the distinctions that the linguistic form s em body and thus prom ote m etapragm atic awareness. Lastly, as evident in m ost chapters o f the volum e, crosscultural differences em erge as a fascinating source o f diversity that needs to be taken into account in ad dition to cro sslin gu istic variatio n . The B erm an and W igglesw orth and Stavans chapters, however, by taking up cultural differences in their ow n right, shift a once m ethodological concern to the status o f an interesting independent variable, in line with a whole body o f research carried out in the ethnographic tradition (Clancy, 19 8 6 ; H eath, 19 8 3 ; H ou gh -E yam i 8c Crago, 19 9 8 ; Ochs 8c Schieffelin, 19 9 5 ; Schieffelin 8c Ochs, 19 8 6 ). REFERENCES Aksu-Kc>9, A ., & Alici, D. M. ( 19 9 9 , April). M arking o f uncertainty and understanding belief states: Child's theory o f evidentiality. Paper presented at the 30 th Child Language Research Forum , Stanford University. A ndersen, E. S. ( 19 8 6 ). T he acquisition o f register variation by A nglo-A m erican children. In B. Schieffelin & E. Ochs (Eds.), Language socialization across cultures (pp. 1 5 3 - 1 6 1 ) . C am bridge, England: Cam bridge U niversity Press. Introduction xix A stin gton, J. W. ( 1 9 9 0 ) . N arrative and the child ’s th eory o f m ind. In B. B ritton & A. P ellegrini (Eds.), N arrative thought and narrative language (pp. 1 5 1 - 1 7 1 ) . Hillsdale, N J: Lawrence Erlbaum Associates. B erm an , R. A ., & Slobin , D. I. (Eds.). ( 19 9 4 ) . Relating events in narrative: A crosslinguistic developm ental study. Hillsdale, N J: Lawrence Erlbaum Associates. C lancy, P. (19 8 5 ). T he acqu isition o f Japanese. In D. I. Slobin (Ed .), The crosslinguistic study o f language acquisition, Vol. 1. The data (pp. 3 7 3 -5 3 4 ) . Hillsdale, N J: Lawrence Erlbaum Associates. Clancy, P. (19 8 6 ). The acquisition o f com m unicative style in Japanese. In B. Schieffelin & E. Ochs (Eds.), Language socialization across cultures (pp. 2 1 3 - 2 5 0 ) . C am bridge, England: C am bridge University Press. Day, C. (2 0 0 0 ) . A developm ental perspective on m odal verb use by French speaking children. In K. E. Nelson, A. Aksu-Ko^, & C. E. Johnson (Eds.), C hildrens language, Vol. 11: Interactional con­ tributions to language development. Hillsdale, N J: Lawrence Erlbaum Associates. Eisen berg, A. (19 8 6 ) . Teasing: Verbal play in two M exicano hom es. In B. Sch ieffelin & E. Ochs (Eds.), Language socialization across cultures (pp. 18 2 - 1 9 8 ) . Cam bridge, England: C am bridge University Press. E rv in -T rip p , S. ( 19 7 9 ). C h ild ren ’s verbal tu rn -tak in g. In E. O chs & B. Sch ieffelin (Ed s.), D evelopm ental pragmatics (pp. 3 9 1 - 4 14 ) . New York: Academ ic Press. G arvey, C. (19 8 6 ). Peer relations and the growth o f com m unication. In E. M ueller & C. C oop er (Eds.), Process and outcome in peer relationships (pp. 3 2 9 - 3 4 5 ). Orlando, FI: Academ ic Press. H eath, S. B. ( 19 8 3 ) . Ways with words: Language, life and work in com m unities and classroom. Cam bridge, England: Cam bridge Unversity Press. H ickm ann, M. (19 8 2 ).The developm ent o f narrative skills: Pragmatic and metapragmatic aspects o f discourse cohesion. Unpublished doctoral dissertation, University o f Chicago. H ickm ann, M. (19 91). The developm ent o f discourse cohesion: Som e functional and crosslinguistic issues. In G. Pieraut-Le Bonniec & M. D olitsky (Eds.), Language bases . . . discourse bases: Some aspects o f contemporary French-language psycholinguistics research (pp. 15 7 - 1 8 5 ) . Am sterdam : Benjam ins. H ickm ann, M. (19 9 5 ). D iscourse organization and the development o f reference to person, space, and time. In P. Fletcher & B. M acW hinney (Eds.), The handbook o f child language (pp. 19 5 - 2 18 ) . O xford, England: Blackwell. Hicks, D. (19 9 1). K inds o f narrative: Genre skills am ong first graders from two com m unities. In A. M cC abe 8c C. Peterson (E d s.), D eveloping narrative structure (pp. 5 5 - 8 7 ) . H illsd ale, N J: Lawrence Erlbaum Associates. H ou gh -E yam i, W., & C rago, M . ( 19 9 8 ). T hree interactional p ortraits from M ohaw k, In uit and White C anadian cultures. In A. Aksu-Ko^, E. Erguvanli-Taylan, A. S. Ozsoy, & A. K iintay (Eds.), Perspectives on language acquisition: Selected papers from the V llth International Congress fo r the Study o f C hild Language (pp. 1 2 4 - 1 3 9 ) . Istanbul: B ogaziq University Press. K iintay, A. ( 19 9 7 ). E xtended discourse skills o f Turkish preschool children across shifting contexts. U npublished doctoral dissertation, University o f California, Berkeley. M cCabe, A., & Peterson, C. (Eds.). (19 9 1a ). Developing narrative structure. Hillsdale, N J: Lawrence Erlbaum Associates. M cCabe, A ., & Peterson, C. ( 19 9 1b ). Getting the story: A longitudinal study o f parental styles in elicitin g n arratives and d evelop in g n arrative skill. In A. M cC abe & C. Peterson (E d s.), D eveloping narrative structure (pp. 2 17 - 2 5 5 ) . Hillsdale, N J: Lawrence Erlbaum Associates. M iller, P. ( 19 8 6 ) . Teasing as language socialization and verbal p lay in a w hite w ork in g-class com m u n ity. In B. Sch ieffelin & E. O chs (Ed s.), Language socialization across cultures (pp. 1 9 9 - 2 1 2 ) . C am bridge, England: C am bridge University Press. Mayer, M. (19 6 9 ). Frog, where are you? New York: Dial Press. N elson , K. ( 19 9 6 ) . Language in cognitive developm ent: The emergence o f the m ediated m ind. Cam bridge, England: Cam bridge University Press. XX AKSU-KO£, NELSON, AND JO H N S O N N icolopoulou, A. (19 9 5 ). N arrative developm ent in social context. In D. I. Slobin, J. Gerhardt, A. Kyratzis, & J. Guo (Eds.), Social interaction, social context, and language: Essays in honor o f Susan Ervin-Tripp (pp. 2 1 - 3 6 ) . M ahwah, N J: Lawrence Erlbaum Associates. Ninio, A., & Snow, C. (19 9 6 ). Pragmatic development. Boulder, CO: Westview Press. O chs, E., & Sch ieffelin, B. (19 9 5 ). T he im pact o f language socialization on gram m atical d evelopm en t. In P. Fletcher & B. M acW hinn ey (Eds.), The handbook o f child language (pp. 7 3 - 9 4 ) . Oxford, England: Blackwell. Perez-Leroux, A. (2 0 0 0 ). Subjunctive m ood in Spanish child relatives: At the interface o f linguistic and cognitive development. In K. E. Nelson, A. A ksu-Koc, & C. E. Johnson (Eds.), C hildrens lan­ guage, Vol. 22: Interactional contributions to language developm ent. H illsdale, N J: Law rence Erlbaum Associates. Peterson, C ., & M cC abe, A. ( 19 9 2 ) . Parental styles o f narrative elicitation: Effect on children ’s narrative structure and content. First Language,12, 2 9 9 - 3 2 2 . Piaget, J. (19 5 1). The language and thought o f the child. London: Routledge. (Original published in 19 2 3) Reim ann, B. (19 9 8 ). M aternal question responses in early child-m other dialogue. In A. Aksu-Ko<;, E. Ergu van li-T aylan , A. S. O zsoy, & A. K iin tay (Eds.), Perspectives on language acquisition: Selected papers fro m the VHth International Congress fo r the Study o f C hild Language (pp. 1 0 8 - 1 2 3 ) . Istanbul: B ogaziq University Press. Sch ieffelin, B., & O chs, E. ( 19 8 6 ) . Language socialization across cultures. C am b rid ge, En glan d: C am bridge University Press. Scollon, R., & Scollon, S. (19 8 1). Narrative, literacy and face in interethnic communication. N orw ood, N J: Ablex. Slobin, D. I. (19 9 6 ). Two ways to travel: Verbs o f m otion in English and Spanish. In M. Shibatani & S. A. Thom pson (Eds.), Essays in semantics (pp. 19 5 - 2 17 ) . Oxford, England: O xford University Press. Snow, C. E. (19 8 9 ). Understanding social interaction and language acquisition: Sentences are not enough. In M. Bornstein & J. B run er (Eds.), Interaction in human developm ent (pp. 8 3 - 1 0 3 ) . H illsdale, N J: Lawrence Erlbaum Associates. Sorsby, A., & Martlew, M. (19 9 1). Representational dem ands in m others’ talk to preschool children in two contexts: P icture bo ok reading and a m od elin g task. Jo u rn al o f C h ild Language,18, 373-396. Vygotsky, L. S. (19 6 7 ). Play and its role in mental developm ent o f the child. In J. S. Bruner, A. Jolly, & K. Sylva (Eds.), Play: Its role in developm ent and evolution (pp. 5 3 7 - 5 5 4 ) . N ew York: Basic Books. Watson, R. (19 8 9 ). Literate discourse and cognitive organization: Som e relations between parents’ talk and 3 year olds’ thought. A pplied Psycholinguistics, 10, 2 2 1-2 3 6 . Setting the Narrative Scene: How Children Begin to Tell a Story R U T H A. B E R M A N Tel Aviv University This chapter deals w ith one facet o f its am biguous title. It concerns h ow children begin the stories that they produce, rather than h ow they first learn or when they first begin to tell stories. The segm ent o f a narrative text that constitutes its start or opening has been the subject o f considerable interest in literary theory, and the exposition is generally recognized as a critical com ponent o f narrative fiction (e.g., Oz, 19 9 6 ; Said, 19 7 8 ; Sternberg, 19 7 8 ). As a p sych ological cou n terp art to this notion, the setting constitutes an integral part o f narrative structure in cognitively oriented “ story gram m ar” analyses (Rum elhart, 19 7 5 ; Shen, 19 8 8 ). D iscourse linguists, too, have paid attention to elem ents that set the narrative scene, analyzed as orientation in L ab ov’s (19 7 2 ) study o f personal-experien ce narratives, o r as in itia l background inform ation in R e in h a rt’s ( 19 8 4 , 19 9 5 ) discussions o f literary and other texts. In this line, Polanyi (19 8 5 ) referred to contextualizing state clauses in her an alysis o f c o n v ersa tio n ally em b ed d ed narratives. Labov identified the orientation as belonging to the narrative rather than the evaluative elements that constitute a story, whereas R einhart suggested that scen e-settin g elem en ts constitu te p art o f the n arrative b a c k g ro u n d , as d istin gu ish ed fro m its fo re gro u n d . In line w ith R e in h a rt’s p ro p o sa l, I have suggested that scene setting, or b a ck g ro u n d o rie n ta tio n , m ay in clu d e bo th in terp retive evalu ative elem en ts and in fo rm ative d escrip tive elem en ts as p recu rsors to the th ird type o f n arrative elem ent, the sequential or eventive elem ents that make up the story plotline (Berm an, 19 9 7a ). Researchers agree that the opening o f a story typically relates to the state-ofaffairs existing p rio r to the onset o f the plot. As such, it provides a backdrop to the ensuing chain o f events by specifying the who, where, when, and w hy o f the events to be reported. In the present context, story setting is defined as serving several fu n ctio n s, term ed here presentative, in fo rm a tio n a l and m otivating, respectively. The presentative function serves to introduce the characters that w ill be referred to su bsequ en tly as p articipan ts in events. T he inform ational fram in g function provides a spatio-locative and/or tem poral fram ew ork for the events. The m otivating fu n ctio n exp lain s w h at sets the ch ain o f events in m otion and w hy an account o f these events is relevant to the hearer/reader or o f l 2 BERMAN interest to the narrator. These three functions have the shared aim o f orienting the audience toward what is to come. This analysis suggests that “ h ow to start a sto ry” constitutes an im po rtan t feature o f the developm ent o f both narrative know ledge and storytelling p e r­ form ance am ong children (Berm an, 19 9 5 ; Reilly, 19 9 2 ). The ability to provide adequate b ack g ro u n d setting in fo rm a tio n is o f considerable interest fo r re ­ search on narrative developm ent for a num ber o f reasons. First, understanding how the different functions o f narrative setting develop should throw light on im portant cognitive abilities that relate to how children develop “ a representa­ tion o f the listener” (Berm an & Slobin, 19 9 4 , p. 6 0 4 ). This shows that they take into account au d ien ce needs — in the case in p oint, b y p ro v id in g adequate background inform ation to the story they are about to tell (M enig-Peterson & M cCabe, 19 78 ). G iving a suitable setting to the story also means that the n arra­ tor can construct a text autonom ously, by m eans o f a self-sufficient m onologic narrative rath er than th ro u gh in te rlo cu to r qu eries, p ro m p ts, and oth er scaffold ing devices. M oreover, it requires preplan n in g o f the text as a w hole, w hich in the case o f narrative discourse im plies a hierarchical, global view o f the chain o f events that are about to be related. These are com plex cognitive d e­ m ands, which take a long tim e to evolve. Related to the d evelo p m en t o f these ab ilities is co m m an d o f n arrativespecific know ledge. B ein g able to p rovid e adequate setting in fo rm atio n and m otivation w ill depend on m ore overall narrative com petence, in the sense o f global plot-organization or “ action-structure,” as defined by cognitive theories o f narrative discourse (G iora & Shen, 19 9 4 ; Rum elhart, 19 7 5 ; Van D ijk, 19 7 6 ). That is, children m ust have recourse to a narrative schem a, w ith an initiating event or problem , one or m ore episodes directed at solving that problem , and an eventual resolution. This suggests that in order to “ begin a story” adequately, children need to be able to structure the rest o f the text appropriately. The present stu d y was u n d ertaken on the assu m p tio n that, w ith age, the scene-setting elements provided to narrative texts will change along three inter­ related dim ensions: am ount, content, and expression. Thus, young preschool ch ild ren p ro v id e little or no such in fo rm a tio n (Peterson , 1 9 9 0 ; U m ikerSebeok, 19 7 9 ) . A n d w h ile ch ild ren from as yo u n g as age 4 p ro v id e som e o rie n tin g b a ck g ro u n d in fo rm a tio n to the stories they p rod u ce, yo u n g e r ch ildren relate to fewer, and to d ifferent, types o f setting fu n ctio n s than do older storytellers (K ernan, 19 7 7 ; Peterson & M cC abe, 19 8 3 ). One aim o f the presen t study, then, is to go b eyo n d these relatively few studies that have analyzed ch ild ren ’s sto ry b egin n in gs b y extending the analysis to a database co n sistin g o f p ic tu reb o o k based n arratives as w ell as p erso n a l-e x p e rie n c e accounts, co m p a rin g p resch oolers w ith school-age ch ildren as w ell as adult storytellers. A second aim , one to the best o f m y know ledge not addressed in p rio r studies o f scene-setting, considers the linguistic form s used to express this com ponent o f narrative discourse. i Setting the Narrative Scene 3 The present study is thus em bedded in earlier w ork that has focused on the relation between linguistic form s and narrative functions across developm ent (Berm an, 1 9 9 6 , 19 9 7 c ; Berm an 8c Slobin, 19 9 4 ; Slobin, 19 9 3 ). A nd it considers three in terrelated d evelo p m en tal p red ictio n s. First, w h at we term ed scenesetting elements (B erm an 8c Slobin, 19 9 4 , p. 8 6 ) w ill change w ith age both in quality and quantity. In quality, preschool children w ill m ention different types o f setting elem ents than older speakers; for exam ple, they m ay m ake explicit reference to p articip an ts bu t not to m otivation s ( who vs. why) and they w ill m ention place m ore than tim e (where vs. when). A n d in quantity, settings w ill constitu te a larger p ro p o rtio n o f texts p rod u ced b y o ld er speakers. Second, elicitation context or genre w ill affect the nature o f setting elem ents provided, but this effect w ill be less m arked w ith age. Third, with age the linguistic form s used fo r scen e-settin g w ill b ecom e less fo rm u la ic , m ore exp licit, and m ore varied, in m eeting different narrative functions. Findings are based on m onologic narrative texts produced by children and adults in d ifferen t e licitatio n settings. N arratives elicited w ith the p ictu red storybook, Frog, where are you? b y M ercer M ayer (19 6 9 ) , constitute the “ frog stories” and accounts o f a personal experience elicited by asking subjects, “ Have you ever had a fight or quarreled w ith som eone? Tell me abou t it,” constitute the “ fight stories.” Supplem entary data com e from an additional set o f oral and w ritten “ fight stories” elicited from older school children and adults, from other personal-experience accounts, where children were asked to tell a sto ry about som ething that had h appened to them , and from texts based on pictures and picture-series. The bulk o f these analyses are from texts produced by speakers o f Israeli Hebrew, on the assum ption that the language in w hich they are co n ­ structed has little effect on the quality and narrative fun ctions o f setting ele­ m ents, when speakers share sim ilar literate, western-type cultural backgrounds o f the kind considered here. O ur findings for H ebrew -specific linguistic form s used for the narrative functions o f m arking story openings, the transition from scene-setting to narrative events, and the distinction between background set­ ting elem ents and narrative events can and should, however, be extended for com parison with other languages. N A T U R E OF S E T T I N G E L E M E N T S IN “ F R O G ” A N D “ F I G H T ” S T O R I E S E arlier research has show n that children favor som e types o f setting elem ents over others. Specifically, they tend to provide m ore “ fram ing” inform ation at an earlier age about the place rather than the tim e at w hich events took place, and they give relatively little in fo rm atio n abou t p articipants and even less abou t background m otivations or reasons for the events (Peterson, 19 9 0 ; Peterson & M cC ab e, 19 8 3 ) . Such studies have ty p ic a lly related to p erso n a l-e x p e rie n c e narratives, whereas the present analysis starts by considering the different kinds o f setting elem ents provided on the basis o f a picturebook (the “ frog story” ). 4 BERMAN T A B L E 1.1 Percentage of Subjects M entioning Different Setting Elements in the Hebrew and English Frog Stories, by Age (N = 24 per agegroup) 3yrs 5 yrs 9 yrs Adults “w ho” [by noun]= boy 58 75 83 92 “where” “when” = jar = at night __ 25 33 8 79 — “why” = feeling — 12 12 63 Setting Elem ent 17 33 Distribution of Setting Elements in Englishand Hebrew-Language “Frog Stories” Th e p ic tu reb o o k Frog, where are you? (M ayer, 19 6 9 ) has been the basis fo r num erous studies o f narrative developm en t in different languages and from different perspectives over and above the database relied on here (B erm an 8c Slobin, 19 9 4 ). These include both published works (e.g., Bam berg 8c D am radFrye, 19 9 1; Bazanella 8c C alled, 19 9 1; Berm an, 19 9 3 ; Kail 8c H ickm ann, 19 9 2 ) and doctoral dissertations (H erm an, 19 9 6 ; Kern, 19 9 7 ; W igglesworth, 19 9 2 ). Yet to the best o f m y knowledge, the nature o f the background scene-setting to this story has not been the subject o f separate study apart from considerations o f referent in tro d u ctio n (K ail 8c San ch ez-L op ez, 19 9 7 ) and a b r ie f note on form ulaic openers (Berm an 8c Slobin, 19 9 4 , p. 86). The 24-pictu re booklet in question tells a story about a boy and his dog w ho go out in search o f their pet frog, w h ich escapes from its ja r d u rin g the night, w hen the b o y and d og are sleeping. The setting scene is provided by the first picture in the booklet, which stands alone, to the righ t o f the title page. It show s the b o y and his d og at the fo o t o f th eir bed, lo o k in g at the frog in its glass jar. Table 1.1 gives the breakdow n o f setting elem ents m entioned in the frog story texts produced by speakers o f (Am erican) English and (Israeli) Hebrew. The figures in Table 1.1 show a marked increase with age in elements counted as belonging to the setting, together with a change in the type o f such elemen ts m entioned by different age groups. O nly h alf the youngest children, aged 3 to 4, introduce the m ain protagonist, the boy, by an appropriately explicit noun phrase, not just as “ he,” com pared with nearly all the older subjects. Few children provide the relevant spatial setting for the frog, as being inside a jar from which it subse­ quently escapes; even fewer subjects m ention that the events took place at night (as shown by the m oon shining in at the window). These findings seem to clearly confirm the first prediction, particularly because very few children provide evalu­ ative or m otivational elem ents to set the backgroun d for how the b o y cam e to have a frog, and w hy he m ight w ant to go out lookin g for it. E xam ples o f such m otivational elements taken from the English-language texts are given in (1). l (1) Setting the Narrative S cene 5 Motivational Elements in English Language Frog Story Texts a. Well— there was a little boy, he liked his— pet frog and his his pet dog— very much [boy, aged 552] b. Um — once there was like— a dog— w ho liked a frog, but the frog didn’t like him, so he decided . . . [girl, aged 957] c. This is a story about a boy— a dog— and his frog. Right now, in the begin­ ning o f the story— he’s— the boy and his dog are just basically admiring his frog, looking at the frog in the jar. The frog looks kind o f happy— obviously he’s not real satisfied with his existence, because w h e n . . . [female adult] The figures in Table 1.1 indicate a clear age-related developm ent in the ability to begin a sto ry w ith scene-setting in form ation . B ut these figures need to be evaluated qualitatively as well. First, in relation to character introduction , the task at hand, where both child and investigator have the p ictu rebook open in front o f them , m ay bias the cognitive issue o f shared know ledge and required level o f inform ativeness in referring to the m ain protagonist. Indeed, this has been shown to be the case in a series o f studies on ch ildrens ability to introduce story characters using this same picturebook, but com paring ou r design w ith a situ atio n w h ere there is no m u tu al kn ow led ge shared betw een the su b jectnarrator and the investigator-audience (Flickm ann, Kail, & R oland, 19 9 5 ; Kail 8c H ickm ann, 19 9 2 ; Kail 8c Sanchez-Lopez, 19 9 7 ). Second, over h a lf the 9 -yearold and adult subjects do in fact explicitly m ention £at night, nighttim e’ as the tem poral setting for their narratives. However, they do so not on ly at the outset in talking about the first picture but subsequently, to introduce the event w hich in itiates the p lo t, fo r exam ple: “A t night, w hen the b o y and his d o g w ere sleeping, the frog ju m p ed out o f the jar and escaped.” This is in m arked contrast to the preschoolers, on ly one o f w hom started to describe the event b y saying “ Then one day at night [szc].” M oreover, the exam ples in (1) are not typical o f the frog -sto ry sam ple: T h eir settings are m ore elaborate than m ost, including the adults’ texts. A gain, this could be task related, inasm uch as b oth n arrator and interlocutor have the same pictures in front o f them. The figures in Table 1.1 thus need to be som ew hat hedged. M ethodologically, p icture-based elicitations create problem s for character introd uction and the need fo r p ro v id in g detailed b ackg rou n d in fo rm a tio n (see the C o m p a riso n s Across Elicitation Settings section). This analysis also suggests that, in principle, story-setting elem ents cannot be defined by a prescribed list o f categories such as who, w here, w h en , and why. R ather, sto ry settings need to be an alyzed in relation to the particular story that w ill unfold. In the case o f the frog story, this m eans taking into account the relationship betw een the b o y and the frog, as m otivating the events to com e. This was m in im ally achieved b y subjects w ho started out b y saying that the boy has or keeps a frog, that he thinks the frog is cute, m ore elaborately b y those who described how the frog cam e to be in the b o y ’s p ossession (he fo u n d it, got it as a present). A gain , alm ost none o f the 6 BERMAN 3- to 4 -year-olds refer to m otivating circum stances, about one third at age 5 to 6 (37% ), and twice as m any 9 -year-olds and adults. Nature of Setting Elements in Hebrew-Language “Fight” Stories The data set for this analysis also includes 12 H ebrew -speaking children at ages 3 , 5 , 7 , and 9, com pared with a group o f 12 adults.1 To elicit a “ fight” story, su b ­ jects were asked if they knew what a fight or quarrel was, and to tell about one they had been involved in .2 Scene-setting elem ents were defined as all m aterial that preceded the event that initated the quarrel or fight, that is, any verbal ref­ erence to w h en , w here, why, o r u n d er w hat circu m stan ces, as w ell as w ith w hom , the altercation took place. These elem ents were analyzed into the fo l­ low ing categories, ranging from least to m ost explicit, from less to m ore elabo­ rated, from juvenile to m ature, as in (2). (2) Ranking o f background-setting elements in personal-experience narratives: 1. No background 1.1 Background element provided by adult input 1.2 Initial Event 1.3 Initial Event plus formulaic opener, for example, pa am ‘once’, yom exad ‘one day’ 2. Minim al inform ational or framing background 2.1 Name o f antagonist plus specifying sex or relationship to narrator = protagonist 2.2 Mention o f place or time o f initial event 3. Specific framing inform ation 3.3 Specification o f a particular time and/or place 3.4 Temporal distancing to specify circumstances surrounding initial event 4. Motivational background, scene-setting orientation 4.1 Temporal distancing to set events o ff in past time 4.2 Metacognitive orientation to the act o f storytelling and/or the nature o f quarreling 1 The texts used in the Hebrew-language data-base were elicited from different groups o f subjects in the various studies referred to in these sections. However, it seems legitim ate to com pare results across these populations, since the subjects all shared the following background: they are children o f educated, m iddle-class speakers o f Hebrew as a first language (like the adult subjects who serve as controls in each study); the preschoolers attend Hebrew nurseryschool or daycare from the age o f 2 and enter kindergarten at age 5 to 6; and the schoolchildren are in gradeschool from 6 to 11 or 12 years o f age, they enter ju n ior high in Grade 7, and complete high school at the end o f Grade 12. 2 Instructions were worded as follows: at a yodea ma ze la-riv, meriva? hayita p d a m bi-meriva? saper li al ze, saper al p d a m fie ravt(a). ‘Do you know what it is to-quarrel, have a-quarrel? Have you ever been ina-quarrel? Tell me about it, tell about a time when you-quarreled’. If the subject hesitated, a prompt was provided: saper li sipur al riv fie hayita bo. ‘Tell me a story about a quarrel/fight you were = took part in’. l Setting the Narrative Sce n e 7 This ranking, as suggested, corresponds to a clear developm ental pattern in am ount and type o f backgroun d setting inform ation provided by respondents. Thus, as illustrated by the excerpts in (3), the youngest children in ou r sam ple, 3- to 4-year-old nurseryschoolers (m ean age 356), typically gave no background at all, but plunged straight into a report o f the events. The on ly exceptions to this were when children used a form ulaic opener such as words m eaning once, one day, as sem an tically non specific m arkers o f d iscourse in itiatio n (see the S to ry O peners section ) o r gave the nam e o f the antagonist, w ith o u t fu rth e r id en tifyin g com m ent. O lder presch oolers, in this sam ple 5 -year-old kin d ergartners (m ean age 554), in som e cases did the sam e, but m an y o f them also added a locative fram e, sp ecifyin g the place w here the in ciden t occu rred , as illustrated in (4). (3) No background scene setting a. ravti im EVad ve baxiti ‘I-quarreled with Elad and I-cried’ [Adi, girl, 355] b. p d a m Orly hi natna li beita al ha-rosh im sirgadol kaze ‘Once O rly she gave me a kick on the-head with (a) big kinda pot’ [Yafit, girl, 3UO] (4) M inim al background scene setting a. ba-gan yalda axat daxfa oti me-ha-nadneda ‘At kindergarten a girl pushed me off-the-swing’ [Efrat, girl, 554] b. etmol ba-gan ravti im xavera sheli Roni ‘Yesterday at-kindergarten I-quarreled with my (girl)friend Roni’ [Meital, girl, 554] In contrast, som e preschoolers, nearly all the 7-year-o ld (G rad e 2) and all the 9 -year-old (G rade 4) school children, provided additional fram in g in fo r­ m ation, often in the form o f som e su rrou n d in g circum stances or event, as in (5 a), or b y being highly specific about the exact place or tim e as in (5b) and (5 c), respectively. O lder children also quite often gave two or m ore different types o f fram ing inform ation (5d, 5e). (5) M ore specific framing by circumstances, time, and/or place a. yom exad sixakti xevel ba-xacer ‘One day I-played [= was playing] jum prope in-the-yard’ [Galit, girl, 551] b. p d a m hayinu ba-kantry ba-brexa, ve haya li mishkefet kazot. . . ‘Once we were at-the-sports-center in -the-pool, and I had kinda goggles’ [Liron, boy, 5 ;6 ] c. hayom seyaradnu ba-m aalit mi-safta . . . ‘Today when we took the-elevator down from Granny . . . ’ [Amit, boy, y ;6 ] d. p a ’am haya li vikuax im ima sheli, ze haya ba-telefon, se ani cilcalti laavoda sh ela. . . 8 BERMAN ‘Once I had (an) argum ent with my M om , it was on-the-phone, when I called her at w ork’ [Dafna, girl, 9 ;2] e. ani ve xaver sheli Ronen halaxnu la-xanut matanot liknot matana la- yomuledet shel Dan . . . ‘Me and my friend Ronen went to-the gift store to buy (a) gift for Dan’s birth­ day’ [Tal, boy, 955] Interestingly, these m ore elaborated settings illustrated in (5) were also quite generally set o ff explicitly from the onset element, or plot initiation, by an overt segm entation m arker, ty pically in the form o f expressions such as ve-az ‘and then or p ifo m ‘suddenly’ (see the Transition M arker section). In addition, proficient narrators, m ainly adults but also som e 9 -year-olds as shown in (sd ) and (se), typically provided som e background m otivation for the quarrel. T h ey did this b y talking abou t the relations between the antagonists b eyo n d this specific in ciden t, b y d istan cin g them selves fro m the events, b y going back in tim e and setting the reported incident in a m ore general fram e o f m em o ries, and also b y v o lu n te e rin g “ m etaco gn itiv e” or “ m eta te x tu a l” com m ents relating to fight scripts in general, or to the storytellin g situation. These strategies are illustrated in (6). (6) M aturely elaborated scene settings a. nizkarti. ani ravti im baxur se haya iti baxeder bakibuc, kse hayiti gar bak- ibuc. al zese hu haya maklit lo flirim al ha-kaseta seli, hu haya ( h)ores li et hakasetot. . . ‘I remember quarrelling with a fellow that was my roommate on the kibbutz, when I was living on the kibbutz, he used to record songs o ff my tape-cassettes, ruining them’ [Yuval, male, 22] b. be-bet sefer yesodi haya p d a m yeled se hecik li. k a ru lo Zohar S. hu nahag laruc axaray ve lehatrid o ti. . . ‘In grade school there once was a kid that gave me trouble. He was called Zohar S. He used to run after me and bother me’ [Yair, male, 28] c. tov, ani xoshevet le-saper al mashehu me-ha-gan. hayta li xavera nora tova, Enav, ba-gan. ve kol hazman hayinu ravot al miney shtuyot. . . ‘Okay, I think I ’ll tell you about som ething from kindergarten. I had a very close friend, Enav, in kindergarten. And all the time we used to quarrel about all kinds o f stupid things’ [Havatselet, female, 22] d. racitse ani asaper lax sipur al riv, aval loh ravti im afexad. ani loh ish se rav im axerim. ve afpa 'am loh ravti im afexad. Ka'asti po ve sham, ka'asu alaypo ve sham, betor yeled xatafti m akot. . . ‘You wanted me to tell you (a) story about a quarrel, but I never quarreled with anybody. I’m not a person that quarrels with others. And I’ve never quarreled w ith anyone. I ’ve gotten mad here and there, people have gotten mad at me here and there, as a kid, I got beaten up’ [Eran, male, 24] l Setting the Narrative Scene 9 The picture that emerges from these excerpts, one that is reinforced b y fin d ­ ings from other personal-experien ce accounts in H ebrew as in English (e.g., Peterson 8c M cCabe, 19 8 3 ) is o f clearly age-related patterns in the ability to take the listener into account in p roviding adequate background inform ation and scene-setting orientation. These can be sum m ed up in term s o f fou r develop­ m ental phases: juvenile, transitional, structured, and proficient: (a) Im m ature, ju v e n ile n arrators p rovid e no setting elem en ts at all, or else o n ly fo rm u laic starters, as in the exam ple in (3). This is consistent with other studies that have noted that yo u n g p resch o o l ch ild ren tend to give little o r no b a c k g ro u n d setting, but instead start their stories with “ im m ediate action,” whether they are m ak in g up fictive stories (Prad l, 19 7 9 ) , em b ed d in g them in con versatio n al in teractio n (M in a m i, 19 9 6 ) , or b asin g them on fam ilia r scrip ts (S eid m an , Nelson, 8c G ruendel, 19 8 6 ). This is followed by (b) a transitional phase, when m inim al inform ation is provided to identify the relevant participants (in the case o f the “ figh tsto ry” sam ple, the antagonist) or the location o f the event. N ext, older, school age children typically provid e (c) structured scene-setting fram es by specific identification o f the place and/or tim e o f the events, co m ­ bined with som e tem poral distancing, and with sequential events being clearly set o ff b y overt m arking o f the transition from background to foreground, plotinitiating events. Finally, (d) m aturely proficient narrations are not o n ly fu lly stru ctu red and te m p o ra lly d istan ced b y m eans o f in itiatin g elem en ts that distinguish the events to be reported from the time and place o f their reporting, they often contain personalized or other evaluative com m en tary concerning the relationship between the participants, the narrator’s attitude to the events reported and to others like them , and/or to the act o f storytelling and reporting on these events. C O M P A R IS O N S ACROSS ELICITATION SETTINGS The two sets o f analyses presented here, on the “ frog stories” and “ fight stories,” respectively, d iffer alo n g a n u m b er o f d im en sio n s, alth ou gh b o th deal w ith n arrative texts p ro d u ced b y sim ilar grou p s o f su bjects (see note 1). T h ese differen ces suggest that an alysis o f scene-setting elem ents m u st take account o f the particular kind o f story being told. In the present case, for exam ­ ple, the frog story is based on the script o f an adventure story, and it is in the genre o f children’s storybooks. The fight stories, in contrast, are based on the script o f a conflict situation and belon g to the genre o f perso n al-experien ce accounts. In fact, the type o f task and the con text o f te x t-e licita tio n tu rn s ou t to have an im pact on the am ount as well as the natu re o f the setting elem ents p ro v id ed across d ifferen t ages. T h is was revealed b y an alysis o f the o verall BERMAN am o u n t o f scene-setting, analyzed as the p ro p o rtio n o f clauses servin g this function across the two types o f texts.3 As shown in Table 1.2 , the prediction o f developm ent w ith age in am oun t o f scene setting was not confirm ed for the frog story sample. The first line o f Table 1.2 shows a consistently low p ro p o r­ tion o f clauses dedicated to scene setting (the first picture ou t o f a total 24 pictures in the book), between 4 % to 7 % o f all clauses across age-groups. In m arked contrast, as show n in the second line o f Table 1.2 , the “ fight sto ry ” reveals a clear and consistent rise in m ean prop ortion o f setting clauses with age, up to 10 % am ong preschoolers, around one quarter at school age, and over one third am ong adults. These find ings are robust, as they tally w ith findings fo r sim ilar types o f elicitatio n across oth er p op u latio n s. A n alyses o f frog sto ry texts in other languages from the Berm an and Slobin (19 9 4 ) study, together w ith French data elicited by the same m ethodology by Kern (19 9 7 ), reveal sim ilar trends. The first picture, providing the background antecedent to the plot-initiating event o f the frog escaping from its jar, yields the same low figures for adult narrators in English (m ean o f 5.5% o f all clauses in the sam ple), Spanish (m ean 7.6 % ), and French (m ean 7.3% ). In m arked contrast to these low figures, analysis o f the setting elem ent in a range o f other “ fight stories” elicited from other H ebrew -speaking school children and adults reveal a closely parallel trend to the origin al “ fight story” sam ple in Table 1.2. This additional database consisted o f other Hebrewlanguage fight stories elicited in m uch the same way as the original set from eight second graders (aged 7 to 8), 12 fourth graders (aged 9 to 10 ), and 12 students (aged 17 to 18 ) and adults, each o f w h om prod u ced two narratives abou t tw o separate experiences w ith a quarrel or fight, one in speech and one in w ritin g (balanced for order o f m odality). In these fight stories, sim ilar to the figures in the second line o f Table 1.2 , the second and fourth graders produced an average o f 2 0 % to 2 5 % backgroun d setting clauses out o f the total clauses in their n arra­ tives, whereas the adults devoted as m any as one third (33% ) to one h alf (4 9 % ) o f their narratives to background setting clauses. Further evidence for these general trends for “ personal experience” stories is that there was no notable difference between the figures for the narratives produced in w riting com pared with speech.4 3 The clause, defined in Berm an and Slobin (19 9 4 ) as “ any unit that contains a unified p red i­ cate . . . that expresses a single situation (activity, event, state)” constitutes a unit o f analysis highly relevant to the characterization o f narrative texts in both form and content (p. 6 6 0 ). 4 Written narratives reveal another, unique feature m arking background scene-setting w hich also has developm ental consequences: M ore than h a lf o f the adults but alm ost none o f the children m arked o ff their story-setting from the initial episode or enabling event graphically, by a separate paragraph. In fact, one adult, a young com puter scientist called Itay, set o ff his first paragraph by a heading in the m argin with the word reka ‘background’ and started his second paragraph (both in ­ dented, with a two-line space between them ) with the one-word heading in the m argin ha-m aase ‘the incident’. I am grateful to N urit Assayag o f the Tel Aviv University project on developing liter­ acy for bringing this subject to my attention. i Setting the Narrative Scene n T A B L E 1.2 Mean Percentage of “Setting” Clauses out ofTotal Clauses in Hebrew Frog Stories Com pared With Hebrew Fight Stories, by Age (N = 12 per age group) Age Group Story type 3 yrs Frog story 6.6 5-2 6.4 9.8 Fight story 5 yrs 7 yrs 9 yrs Adults 4-5 5-5 7.2 20.6 27.7 36.4 These figures appear, m oreover, h igh ly consistent w ith oth er fin d ings for am o un t o f backgrou n d setting m aterial com pared w ith overall text length as defined b y num ber o f clauses across a range o f other m aterials, in English and in Hebrew. The analyses in the rest o f this section derive from less detailed but sim ilarly m otivated exam inations o f background setting elem ents in narrative texts produced by three distinct m ethods o f elicitation: narration o f content o f p ictu re series, re ca p itu la tio n o f p erso n a l-e x p e rie n c e acco u n ts, and m akebelieve, fictive stories. T h e database fo r pictu re series consists o f H eb rew language m aterials based on three sets o f four pictures each told by preschool children aged 4, 5, and 6, com pared with 1 0 -year-olds and adults (B erm an & Katzenberger, 19 9 8 ; Katzenberger, 19 9 4 ), on a H ebrew replication o f the “ cat and h o rse ” series used b y H ick m an n and h er associates (H ick m a n n , 1 9 9 1 ; H ick m a n n , H en d rik s, R o lan d , & L ian g, 19 9 6 ) , w h ich elicited texts fro m 15 H ebrew speakers aged 5, 7, and 11 years com pared w ith 10 adults (Kahanow itz, !995)> and on oral stories based on a series o f pictures about a visit to the zoo by seven H ebrew -speaking preschoolers, and w ritten versions from 14 adults (B erm an , in press). T h e database fo r person al-experience accounts consists o f a range o f H eb rew -lan gu age m aterials on v a rio u s su bjects elicited fro m children aged 3 to 12 and adults (described in Berm an, 1 9 9 5 , 19 9 7 a ). These are supplem ented b y w ritten and spoken versions o f two different fight stories by 48 H ebrew -speaking grade-school and high-school students and 16 adults, and co m b in ed w ith the E n glish -lan g u ag e texts in the a p p e n d ix to Peterson and M cC abe (19 8 3 ). A n d the m aterials surveyed for m ake-believe stories are based, again , on a ran ge o f H eb rew -lan gu age texts cu lled fro m v a rio u s sou rces, com bined w ith the large num ber o f stories collected by Pitcher and Prelinger (19 6 3 ), as published in Sutton-Sm ith (19 8 1). T h is su rve y o f a w id e range o f m aterials p ro vid es stron g su p p o rt fo r m y earlier claim regarding “ intertask differences in children’s narratives” (Berm an, * 9 9 5 >PP- 2 9 5 - 3 0 2 ) . There, the issue at hand was how the type and context o f elicitation affects children’s ability to give expression to principles o f narrative discourse organ ization . In developm en tal term s, this ranges from im m atu re exp ressio n o f iso lated events, via en co d in g o f te m p o ra l sequence and local 12 BERMAN relations o f causality, on to a global, hierarchically organized action structure. C om parison across different narrative genres and varied m ethods o f elicitation (recoun tin g o f fam iliar script, o f a p erson al experience, o f the contents o f a p ictu re series, a p ictu red sto ry b o o k , and a film w ith o u t w o rd s) y ield ed the conclusion that “ narrative abilities . . . do not develop along a un iform ly linear curve . . . [since] d ivergen t results em erge in d ifferen t settings and across different tasks” (Berm an, 19 9 5 , p. 29 8 ). In the present context, I wish to make an analogous claim for the developm ent o f childrens ability to start a story by providing relevant, and adequate, background, scene-setting inform ation. That is, here as in other dom ains o f developm ent, task effect needs to be taken into account. Children proved able to dem arcate setting elem ents better and earlier in p erso n al exp erien ce accoun ts than in n arratives based on a p ic tu reb o o k story. They did so as young as age 3 when they were free to tell about anything that had h app en ed to them , but o n ly fro m aro u n d age 5 w hen asked to tell specifically about a fight they had experienced. In general, personal experience accounts appear to p rovide m ore authentic contexts for elaborating on scene setting than picture-based elicitations. The exam ples in (7) are based on a series o f six pictures depicting what happens to two children visiting a zoo, when the m o n k ey snatches the ice cream from the y o u n ger child (B erm an , in press). T h ey suggest that young children m ay not bother to provide setting elements at all in picture-series narrations. (7) Opening clauses o f three children’s picture-series based “ zoo stories” 1. pa am hayu shney yeladim ve hem halxu im ima shelahem le-gan xayot ‘Once there were two kids and they went with their mom to the zoo . . . ’ [Doron, boy, 6;8] 2. po ani ro e yeled ve yalda ve xayot ve az hem . . . ‘Here I see a boy and a girl and animals and then they . . . ’ [Tibi, boy, 558] 3. po hem be-gan xayot ve po hem mistaklim al arye’im. ‘Here they’re at the zoo and here they are looking at lions . . . ’ [Batya, girl, 4;8] A lth o u gh they all clearly recognized the “ zoo” script situation in the p ic ­ tures, on ly from school age were children able to provide any kind o f scene set­ ting as a fram e fo r their texts, and o n ly children aged 6 and old er p rod u ced stories organized around an acceptable action structure. This finding is consis­ tent w ith results for another set o f picture series as analyzed in Katzenberger (i9 9 4 )- In contrast to these p ictu re-series elicitations, n o n pictu re-b ased fic ­ tional accounts, where children are asked to create im aginary stories, appear to provide a particularly rich context for expressing early scene-setting abilities (as can be in ferred from w hat k in d ergarten children do in a “ p re te n d -re a d in g” task, as in Segal, 19 9 6 ) . A fter all, in prose literature, b ackgrou n d exp o sitio n plays a crucial role in construction o f narrative texts. A gainst this backgroun d, I suggest that som e types o f narrative-elicitation l Setting the Narrative S cene 13 tasks and certain com m unicative situations w ill prom ote earlier, and richer, ex­ pression o f backgroun d settings than w ill others. These can be ranked as in (8), from m ost to least likely to elicit appropriate backgroun d m aterial from rela­ tively early on. (8) Ranking o f scene-setting evocation, by narrative genre, elicitation method, and communicative context 1. Fictive Make-believe fantasy a. based on own imagination b. based on known film or book 2. Veridical Personal-experience account a. outside investigator b. familiar interlocutor 3. Fictive Picture-storybook based a. no mutual knowledge b. mutual knowledge 4. Fictive Picture-series based a. no mutual knowledge b. mutual knowledge We can thus explain the disparity in Table 1.2 between the am ount o f textual m aterial given over to background setting elements in the “ frog story” database (type 8.3b) com pared w ith the “ fight story” m aterials (type 8.2a) as a function o f the differences in narrative genre and in the elicitation context. I suggest that, in general, as indicated b y its ranking in (8.1), m ake-believe fantasy w ill be the first type o f n a irative in w hich children w ill provide som e scene-setting in fo rm a ­ tion. In fact, it is in the context o f fam iliar m aterials that yo u n g p resch oolers have seen, heard, or had read to them (as listed in 8.1b), that they w ill first a c ­ quire the conventional m arkings o f story openers such as ‘once upon a tim e’ or its B iblical style classical H ebrew counterpart hayo haya p a am ‘be was once = once there w as’ (see Story O peners section). Because the interlocutor needs to be introduced to the fantasy w orld being created or recreated in the narrative text, adequate b a ck g ro u n d in fo rm a tio n is essential fo r o rie n tin g the au d ien ce to w hat is about to be told. A nd this is m ore critical in the case o f a story that is u n ­ fam iliar to both narrator and audience, as in (8.1a), than to one they have shared knowledge o f (e.g., a favorite fairytale or w ell-know n fable) as in (8.1b). N ext in rank as “ setting-evocative” are personal experience accounts. Here, the factor o f m utual knowledge is critical. As shown by the excerpts from the 3 -year-olds in (3) com p ared w ith those from the 5 -year-old s sh ow n in (4), y o u n g ch ild ren often fail to p ro v id e the m in im a l referen tial in fo rm a tio n n e cessa ry fo r an unfam iliar investigator, as in (8.2a), in situations o f “ no m utual knowledge,” to identify the participants in the event. On the other hand, in fight stories elicited in H ebrew in a situation where school children recounted a personal experience to a frie n d or classm ate, often on e w h o had been p resen t at the even t, (e.g., situ atio n 8 .2b ), even teen agers felt no need to sp e cify d etails o f the oth er participants’ identity, beyond their nam es, and they tended to provide m inim al locative fram ing if the event took place at school. But these high-school students, 14 BERMAN like the adults telling a fight story to a friend, all gave overt expression to other setting elements, b y providing suitably detailed and distanced tem poral fram ing. They also often gave extensive m otivational background in term s o f the general relation sh ip betw een the n arrato r and the other p articip an t(s) in the events, personal predilections o f the narrator and/or other participants, and so on. In contrast to such rich settin g-evo catio n in self-co n stru cted n arratives, w h ether fictive or veridical, p ictu re-based elicitation procedures o f pictu red sto ryb o ok s (8.3) or p ictu re-series (8.4 ), app ear far less accessible to explicit v e rb alizatio n o f setting elem ents. T h is is sh ow n, as noted, fo r the relatively sm all am o u n t o f backgrou n d in form ation p rovid ed in a w ide range o f frogsto ry texts in d ifferen t lan gu ages, even am o n g adults. A n d K a tz en b e rg er’s (19 9 4 ) large sam ple o f H ebrew -language texts elicited on the basis o f several different picture series from 4 -, 5-, and 6 -year-olds com pared w ith 1 0 -yearolds and adults, reveal that from age 5, ch ild ren often give som e stan d ard “ op en er” such as the w ord for ‘on ce’ (S to ry O peners section) plus a m inim al referent introduction w ithout any additional background inform ation p rio r to the initiating event. The single exception was one 1 0 -year-old, n early h a lf o f whose clauses about a series o f four pictures show ing a w om an in a hat store buying a new hat were given over to “ distanced” and m otivational background settin g.5 True, situations o f no m utu al know ledge, that is, (8.3a) and (8.4a), enrich the am ount and form o f referent introduction, as noted for frog storybased studies o f this kind (Setting Elem ents section), and as such they are better suited to m eeting the “ presentative” function o f establishing story background. B ut they, too, fail to stim ulate m uch in the w ay o f the other two functions o f sto ry settings: locative and p articu la rly , te m p o ra l fra m in g and evalu ative m otivation. This is noteworthy, because picture-based elicitations have yielded p a rtic u la rly rich analyses o f ch ild re n ’s n arrative ab ilities across d ifferen t languages. These findings highlight a general point o f both principle and m ethodology: D ifferent types o f elicitation procedures and com m unicative contexts prom ote, or at least allow expression to, som e types o f abilities earlier or m ore than others (Berm an, in press). True, in developm ental term s, once both narrative com pe­ tence and sto ry tellin g p erfo rm a n c e are w ell establish ed , o ld er and m ore p roficien t n arrators w ill prove less su sceptible to effects o f task and context w ith respect to n arrative sto ry -se ttin g as in oth er d om ain s. N on eth eless, the ra n k in g ten tatively p ro p o se d in (8) is w o rth e x am in in g u n d er su itab ly c o n tro lled c o n d itio n s, u sin g co m p arab le m aterials (e.g., b ased on the sam e them e, to p ic, or scrip t) across differen t types o f tasks and d ifferen t developm ental phases, to further test the prediction that setting evocation w ill be strongest at level (8.1a), weakest at (8.4b). 5 I am grateful to Dr. Katzenberger for m aking her sum m ary and illustrations o f these data available to me. i Setting the Narrative Scene 15 E X P R E S S I O N O F L I N G U I S T I C F O R M / F U N CTI O N R E L A T I O N S This final set o f analyses focuses on the linguistic form s used for three related narrative function s: to m ark n arrative openings, to dem arcate the tran sition from scene-setting to narrative events or episode inception, and to distinguish between background setting elements and narrative events. Story Openers An early developm ent in m arking linguistic form :fu nction relations is use o f a tem poral term identifying the text as in the narrative m ode. Thus Prad l’s (19 7 9 ) study based on the large Pitcher and Prelinger ( 19 6 3 ) corpus o f fictive stories children were asked to tell, noted that only 2 0 % o f the 2 -year-olds began their stories w ith a form al op enin g device, w hereas this increased to n early threequarters (7 3% ) am ong the 5 -year-olds. This is consistent with findings from the frog-story sam ple in English and Hebrew, in which nearly all the preschoolers plunged directly into picture description or narration, whereas 23 out o f the 24 fo u rth grad ers (9 - to 1 0 -year-o ld s) p ro v id ed som e in tro d u ctio n , and tw o-thirds o f them used what we term ed “ form ulaic opening expressions” such as ‘once, once upon a time, one day’ or their H ebrew equivalents, as described next (Berm an & Slobin, 19 9 4 , pp. 7 4 - 7 5 ) .6 A term such as ‘on ce’ (H ebrew p a am) specifies that som eth ing h appened, and it h app ened in the past. A qu arter o f the 3 -year-old s and n early h a lf the 5 -year-olds opened their fight stories with this w ord, or w ith a sim ilar expres­ sion, ‘one tim e’ (H ebrew pa'am axat), ‘one day’ (yom exact), and an occasional ‘yesterday’ (etmol). So did the 7- and 9 -year-old school children, but their use o f these term s d iffered im p o rta n tly from that o f yo u n ger ch ildren. A m o n g preschoolers, these tem poral openers were invariably text initial, whereas am ong older children they could be text internal. For exam ple, 7-year-old D avid (756) started o ff with halaxti p a am la-ken im S a a r ‘I went once to the clubhouse with Saar’, and 9 -year-old Etti (917) started her story with ani asaper lax riv se haya li im axoti ha-gdola. p a am a x a t. . . ‘I’ll tell you about a quarrel I had w ith m y older sister. One tim e .. . . ’ M oreover, occasionally am ong 5-year-olds and invariably w ith o ld er ch ild ren, these openers were acco m pan ied b y a dd itio n al setting in form ation about tim e and/or place, in line w ith the general d evelopm ental trends noted earlier. In contrast, early use o f these narrative openers was very restricted in function: The children started o ff a story as h aving som e generic 6 One o f these fourth graders but only two o f the Hebrew- and English-speaking adults started out with the highly narratively oriented com m ent “ It’s a story a b o u t . . . ” This m ight reflect cu l­ tural conventions, since in Kern’s (19 9 7 ) French sam ple, which used the sam e elicitation p roce­ dures as the Berm an and Slobin ( 19 9 4 ) study, over h alf the adults although only one 11-year-old and none o f the younger children started their narratives with a com m ent along the lines o f C est une histoire de . . .” l6 BERMAN “ p rio r-to -th e -p re sen t” tem p o ral location , but failed to an ch o r it w ith in any specific tim e fram e. Use o f these openers is thus essentially form ulaic, rather than well m otivated in term s o f the sem antics o f tem porality or the discourse function o f tem poral fram ing. But they do show that children are fam iliar with conventional narrative devices m arking story beginnings in their culture. In the picture-based frog-story corpus, as noted, use o f sim ilar expressions, such as English ‘once upon a tim e’, occurred at the begin n in g o f m any o f the texts p rod uced b y 9 -year-old s but in alm ost none o f the yo u n ger ch ild ren ’s. The m ost typ ical such op en er was in the form p a 'am/pa am axat haya yeled ‘once/one tim e (there) was a-boy’ (used by no fewer than 6 out o f 12 H ebrew ­ sp eakin g 9 -year-o ld s and 3 ou t o f 12 o ld er ch ild ren , aged 1 1 to 12 ). These exp ression s serve a rather d ifferen t, th o u gh no less stereotyp ic, c u ltu ra lly conventional function than the term s pa'am /pa'am axat ‘once/one tim e’ in the personal-experience fight stories o f the younger, preschool children; they m ark the start o f a children’s fairytale or fictional storybook account. M oreover, four o f the 11- to 12-year-o ld sixth-grader frog stories, but none am ong the 9 -yearold fourth graders, used an archaically flavored literary type opener in Biblicalstyle H ebrew, in the form o f hayo haya p a'am ‘be was once = there once w as’. T h is is the classic o p en er fo r H eb rew ch ild ren ’s literatu re in fairytales and fables, so that this finding ties in with w hat was noted in the previous section about the im portant effect o f genre on narrative setting or expositions. These conclusions are supported by analysis o f make-believe stories written b y third-grade 8- to 9 -year-olds com pared with sixth-grade 11- to 12-year-olds asked to m ake up a sto ry ab o u t a ch ild w ho m eets a stran ge creature on a jo u rn ey (Argem an, 19 9 6 ). O f the younger children, two started out with yeled exad ‘a b o y = ch ild ’, tw o w ith yom exad ‘one d a y ’, m ost (7 ou t o f 12 ) w ith pa'am /p a a m axat ‘once/one tim e’, and one w ith the fable-m arking hayo haya ‘there was o n ce’. A n alm ost id en tical b reakd o w n m arked the open ers o f the m ake-believe stories w ritten by the older group o f 11-year-old s, except that two o f them started w ith the classically flavored hayo haya opener. T h is suggests that m ode o f elicitation (w ritten versus spoken) and narrative genre (m akebelieve fable versus p ictu red ad ven ture sto ry) w ill evoke earlier, m ore w id esp read use o f stric tly con ven tio n alized sto ry op eners. T h at is, these children were m anifesting knowlege not only o f the narrative as a type o f text, but o f literacy-related awareness o f subgenres o f narrative as well. M oreo ver, the d ifferen t lin g u istic an d textu al o r situ atio n al contexts in w hich a single term such as pa'am ‘once’ is used as a story opener supports ear­ lier findings with regard to form :fu nction relations in language acquisition in general, and in the d evelo p m en t o f n arrative abilities sp ecifically (B erm an , 1 9 9 6 ,1 9 9 8 ; B erm an 8c Slobin , 19 9 4 ; Slobin, 19 9 3 ). Early uses o f a given lin ­ guistic form w ill serve a restricted range o f fun ctions, w hereas the sam e su ­ perficially identical form s m ay serve different functions across developm ent. With age, the range o f form s used to indicate that a story is about to begin not l Setting the Narrative Scene 17 on ly becom es m ore varied and m ore personalized or less stereotypic, the form s also tend to be m ore explicit as m arkers o f a particular narrative genre and/or function. Thus, for exam ple, the young m an referred to in note 4 as m arking o ff the backgrou n d section not o n ly by a separate p aragraph but also by explicit m ention o f the w ord reka ‘backgrou n d ’ in the m argin o f his w ritten story in ­ troduced an oral narrative by m eans o f this same word when telling a friend o f his about a quarrel he had been involved in at work. In the opposite direction, that o f the sam e fo rm servin g different fu nctions w ith age, none o f the adult n arratives, w h eth er fro g -b o o k based or p erso n a l-e x p e rie n ce fig h t-sto ry a c ­ counts, started out w ith tem poral m arkers like pa'am (axat) ‘once, one tim e’ or yom exad ‘one d ay’. As noted earlier, adults start ou t th eir stories w ith m ore specific te m p o ra l an d /o r locative fra m in g , e.g., etm ol h alaxti lem ale delek ‘Yesterday I w ent to take gas’ [H anan, m an, 25], lifnev xam esh shanim horay, baali ve ani yacanu le -tiy u l. . . ‘Five years ago m y-parents, husband, and I went on a trip . . .’ [Sara, w om an, 4 0 ], be-bet sefer yesodi haya p a'am yeled ‘At grade school there once was a b o y’ [Yair, m an, 28]; or else they make som e m etacognitive com m ent on their recall or reconstruction o f the event, for exam ple, ani rotse le-saper al mikre riv ‘I want to tell about a quarrel incident’ [Shlom o, m an, 32], toy, ani xoshevet le-saper al mashehu me ha-gan ‘Okay, I think I’ll tell about som ething from kindergarten’ [Havatselet, w om an, 22]. Tem poral adverbs such as yom exady p a 'a m ‘one day, o n ce’ do occu r in the adult n arratives, b ut in a different place, not at the outset, and for a different purpose — to m ark b ack ­ ground setting o ff from plot initiation, as next discussed. Transition Markers Consider, next, h ow narrators m ark o ff or otherwise indicate the b o u n d ary b e ­ tw een scene setting and p lo t in ceptio n . In w ritin g , this m ay be g ra p h ic a lly m arked by m eans o f paragraphin g (see note 4), but in spoken texts, som e overt lin g u istic fo rm is n eeded to p erfo rm this k in d o f seg m en tatio n . Y ou n g p re sch o o l ch ild ren use overt, co n ven tio n al lin g u istic m eans to m ark sto ry openings even p rio r to the developm ent o f a w ell-structured narrative schem a, but in contrast, the transition from setting to the events w hich start the sto ry p er se is often blurred and not clearly m arked in their texts. Table 1.3 shows the expressions used to m ark the transition from introd uctory setting to the plotline chain o f events in the H ebrew frog-sto ry texts, w here the frog is depicted escaping from its jar while the boy and dog are asleep in the bed nearby. T here is an alm ost com plem en tary d istribu tio n betw een the yo u n ger and older speakers in m arking the transition, show n by the different clustering o f the figures in each colum n o f Table 1.3, and b y the figures in bold, w hich stand for the favored m eans at each age group. T hree-year-olds favor zero m arking, 5 -yea r-o ld s p re fe r ‘a n d ’, and 9 -year-o ld s rely on an e x p lic itly te m p o ra l expression. The fact that youn g children generally provide no overt m arker o f l8 BERMAN TABLE 1 .3 Markers of Transition to Plotline Chain of Events in Hebrew Frog Story Texts, by Age (N = 12 per age group) 3 yrs D evice 5 yrs 9 yrs Adults 2 6 1 — ve ‘and’ 5 6 — 1 (ve) hine and here(on)’ 1 — 1 Zero marking az ‘then, so’ — 2 — — p it’om ‘suddenly’ 2 — axarkax ‘afterwards’ — — 1 2 — — — yom exad ‘one day’ — — 1 1 benatayim ‘meanwhile’ — — 2 — balayla ‘at night’ — — 7 2 balayla kse at night when’ — — — 2 bizman se while (that)’ — — — 3 narrative-event inception indicates failure to distinguish between background setting and fo re g ro u n d p lo tlin e. I f 3 -year-o ld s do m ark su bseq u en t events versus p rio r states, they use the vague, general connector ve ‘and’, in a way as yet lacking in conventional syntax or sem antic content, and m eeting no norm ative narrative function (Berm an, 19 9 6 ; Peterson 8c M cCabe, 19 9 1). O lder speakers alm ost always m ark the b o u n d ary explicitly, either b y the general episode m arker ‘one day’, or b y m ore specific term s for points in tim e (‘at night’) or duration (‘while the boy slept’). A dults use a w ider range o f form s than other age groups, and they avoid sequential expressions like ‘suddenly, and then, after that’, w hich are favored by school-age children. Besides, where the yo u n gest and old est grou p s share su rface fo rm s, these serve quite d istin ct fun ctio n s. Z ero m arkin g in the case o f the 3 -year-old s is indicative o f their picture-by-picture description o f isolated scenes and events. Adults w ho fail to use an overt m arker o f plot inception rely on other devices to m ark the transi­ tion from background setting to foreground plot— a switch in verb tense or a shift from stative to dynam ic predicates.7 The transition m arkers used by the 5- and 9 -year-olds, again, illustrate m ore general trends in the developm ent o f n arrative fo rm -lin g u istic fu n ctio n relatio n s. F ive-y e ar-o ld ch ild ren reveal 7 T he locative hine ‘and here’ also functions differently in the younger and older texts: For the little children, it has a deictic spatial function, corresponding to the use o f the tem poral deictic now, like French void', for adults, H ebrew hine is anaphoric, it m arks o ff a given point in the chain o f events u nder discussion. A sim ilar switch from a deictic to an an ap h oric, d iscourse-m otivated function occurs with the time word axshav, much like its English counterpart ‘now ’ (e.g., The boy is in danyer now that the owl has been disturbed). i Setting the Narrative Scene 19 com m and o f narrative sequentiality through m arkers o f linear clause-chaining: m ultifunctional ‘a n d ’, along with sequential term s like Then, suddenly, after­ w ards’ . N ine-year-olds are m ore like adults in segm enting background setting from plot initiation through use o f specifically tem poral term s like ‘m eanw hile, that night’ . In the fight-story sam ple, too, 3 -year-olds on ly occasionally m arked o ff the initial chain o f events b y a sequential term such as az ‘then, so’, axarkax ‘after­ w ard s’, or p ifo m ‘su d d en ly’. These expressions serve this tra n sitio n -m a rk in g fu n ctio n in the b u lk o f the c h ild ren ’s fight stories fro m age 5 years up, for exam ple, az hitxilu c a a k o G so/then (there) started sh ou tin g’ [Shay, boy, 550], p ifo m yeled exad shovav kafats ‘ Suddenly a n augh ty kid ju m p ed (o u t)’ [Galit, girl, 5;i], az axarey ze hu om er ‘A nd so after that he said” [Tomer, boy, 755]. The adults rarely used the term p ifo m , which serves as a typical m arker o f episode in itiatio n in ch ild ren ’s sto ryb o oks. A n d if they did, it was the m ore literary, h ig h -re g iste r e q u ivalen t le-feta ‘a ll-o f a su d d en ’ (e.g., le-feta xash A v i be-ra ‘of-a-sudden, Avi was taken bad = ill’ [Sara, w om an, 4 0 ]), in line with what was fo u n d fo r the H eb rew fro g -sto ry sam ple as w ell (B erm a n & S lo b in , 1 9 9 4 , p. 3 0 1). In contrast to the children, adults m ainly used the punctual term (az) yom exad ‘(then) one d ay’ as tran sition m arkers; fo r exam ple, an ad u lt fight sto ry o f 38 clauses lon g abou t how in ju n io r high they used to th row things dow n on people in the street below started w ith 10 backgroun d in tro d u ctory clauses, then switched to the initial event as follows: az yom exad lakaxti tapuax se heveti me ha-bayit ‘So one day I-took an apple that I-brought from hom e . . .’ [Udi, m an, 23]. In general, tem poral m arkers o f transition from background to plot onset used b y old er speakers are m ore specific, fo r exam ple, boker exad ‘one m o rn in g ’ [Sarit, w o m an , 2 1], co rresp o n d in g to the tra n sitio n -m a rk in g ba-layla ‘at n igh t’ o f the frog story. T h ey tend, also, to be m ore detailed, and often introd uce an em bedded tem poral clause, for exam ple, ba-yom bo hexel ha-kurs ‘on-the-day on-w hich started the-course’ [Idan, m an, 22], yom exad se tiyalti ito ‘One day when I was out w ith-him = the dog’ [Shlom o, m an, 3 1], yom exad, kse hu hecik li ‘One day, when he bothered m e’ [Yair, m an, 28]. A sim ila r p referen ce fo r a p a rtic u la r fo rm to m ark the sh ift fro m scene setting to the start o f the action am ong older H ebrew speakers was even m ore m arked in anoth er narrative task. T hree- and 4-year-old presch ool children, 11-yea r-o ld sixth graders and adults were asked to make up a story based on a large picture show ing an old m an carrying a sack o f fish w alking toward a house where a w om an and children stand w aiting on the porch (B en-H aviv, 19 9 6 ) . M ore than h a lf the yo u n ger children started their texts w ith the expressions p a am ‘once’ or yom exad ‘one day’, show ing that they knew they were supposed to “ tell a story.” The school children and adults w ith on ly one exception used sim ilar exp ressio n s, fo r exam ple, p a yam axat ‘one tim e ’, yom exad ‘on e d a y ’, boker exad ‘one m orn in g’ at a point three to four clauses into their narrations. T h ese tem p o ral ad verbs served to ind icate a sw itch fro m scen e-settin g 20 BERMAN background description to reporting the narrative chain o f events. In m arking this transition, too, the identical linguistic form s serve different narrative fu n c­ tions at different phases in the developm ent o f storytelling abilities. Tense/Aspect Shifts Th is h ead in g con cerns the ab ility to encode rh eto rical altern ation s betw een backgroun d setting and foreground plot elements. I exam ined use o f tense/as­ pect shifting to distinguish story introductions from the chain o f plotline events, since gram m atical aspect is recognized as a key m eans for distinguishing fo re­ ground and background elem ents in narrative (e.g., B erm an 8c Slobin, 19 9 4 , pp. 6 -9 ; C hvany, 19 8 5 ; H op per, 1 9 8 2 ; Labov, 19 7 2 ; R ein h art, 19 8 4 ) . M o d ern Hebrew, unlike the classical Biblical language, does not m ark aspectual distinc­ tions gram m atically by inflections on the verb, so that today’s H ebrew speakers need on ly m ark the inflectional distinction between finite verbs m arked for past com pared w ith present and future form s (Berm an 8c D rom i, 19 8 4 ; Ravid, 19 9 5 , pp. 4 2 - 4 5 ) . Two relevant findings em erged from ou r large-scale crosslinguistic stu d y in this respect. First, we fo u n d alm o st no evid en ce fo r linguistic com pensation, d efined as exp ressin g b y lexical m eans n o tio n s th at are not m o rp h o lo gica lly gram m aticized in the language. M ore specifically, we noted that “ w ith regard to verbal aspect, we found on ly rare instances o f attem pts in G erm an and H ebrew to ad d d istinctions o f pu n ctuality or d urativity that are n ot m arked g ram m atic a lly in the lan g u a g e ” (B erm an 8c S lo b in , 19 9 4 , pp. 6 2 1 - 6 2 2 ) . Second, in the H ebrew fro g -sto ry corpus, narrators “ use tense shifting as a means o f global discourse organization,” one o f whose functions is “ to set o ff background settings . . . from the central body o f the plot” (Berm an 8c Slobin, 19 9 4 , p. 295). But narratively m otivated deploym ent o f tense shifting is restricted in several ways. First, tense shifts are o n ly from past to p resent or present to past, depending on w hich tense the text was anchored in in general. Second, on ly som e out o f the 16 adults in the H ebrew sam ple (Berm an, 19 8 8 ) sh ifted tenses to a no ticeable extent; and o n ly som e o f these did so fo r the purpose o f distinguishing background settings from the m ain plotline. Third, in developm ental perspective, none o f the H ebrew -speaking children used tense shifting as a device for global narrative organization in the frog-sto ry sam ple; for exam ple, school-age children used it as a local device to express the tem poral relatio n s betw een co m p lem en t clauses and th eir m a trix p redicates (Shen 8c Berm an, 19 9 7 ). G lobal text-level tense shifting was a peculiarly adult device, not em p loyed even b y 9- to 1 0 -y ea r-o ld fo u rth grad ers w ith fu ll co m m an d o f com plex syntax and global narrative action structure. For purposes o f the present study, I exam ined a range o f other narrative texts produced by H ebrew-speaking children and adults to see whether tense shifting w o u ld serve to d istin gu ish b ackg rou n d setting or o rien tatio n from the foreground narrative events. M y hypothesis was that in nonpicture-based n arra­ i Setting the Narrative Scene 21 tives, present-tense benoni ‘interm ediate’ participial form s w ould function to set o ff background situations from foreground events. A nd indeed, the findings for the H ebrew frog-story corpus were strongly confirm ed in these other sam ples, too: O nly older speakers, and only som e o f them, used tense shifting to serve the narrative function o f m arking o ff story beginnings from their continuation. This confirm s the prediction that, with age, the particular elicitation setting exerts less effect on narrative production than am ong younger subjects. M oreover, in one sam ple, in w hich texts were elicited from older, teenage children, they behaved m ore like the adults in this connection than did younger grade schoolers. The im ­ pact o f increased exposure to different types o f narrative and other text types and well-established literacy evidently makes 1 2 -year-olds m ore fam iliar with a range o f cultural conventions and rhetorical options o f the narrative genre. A second find in g also went beyond w hat em erged from the H ebrew frogsto ry sam ple. N arrato rs used a range o f other fo rm al op tion s in ad d itio n to present/past-tense shifts to distinguish setting from story, and they did this sim ­ ilarly in quite different contexts. These included (a) an interview -type situation, in which Israeli adults were asked to tell about their experiences in high school and the arm y; (b) an elicitation setting, in which children and adults were asked to pretend th ey w ere telling a sto ry based on a large colored p ictu re to their friends or their pupils at nursery school; and (c) personal-experience accounts as represented by the fight-story sample. In all these settings, narrators w ho are characterizable as “ p roficien t” used a range o f devices to d istinguish between predications in the backgroun d settings and those describing narrative plotline events.This distinction is achieved b y shifting between the sm all repertoire o f relevan t tense/aspect m ark in g fo rm s in the language: p resen t tense benoni ‘ interm ed iate’ form s, w hich are also p articipial in fun ction; past tense form s inflected for person as well as num ber and gender, which cover the w hole range o f English past-tense form s — progressive and perfect as well as sim ple; and the com plex form o f haya + benoni ‘was/were + p articiple’, equivalent to E nglish ‘w o u ld do, used to d o ’. T h e excerpts in (9 ) and ( 10 ) illu strate tense/aspect sw itches used to d istin gu ish setting fro m n a rrative-even t p redicates in o u r H eb rew n a rrative sam p le: fro m p resen t-ten se (p artic ip ia l) fo rm s fo r b a c k ­ ground setting to the past tense in event recounting (9.1); shifting from past to p resent tense in a h isto rical o r narrative p resent (9 .2 ); sh iftin g betw een the co m p lex fo rm o f h ab itu al past follow ed by the u n m arked sim ple past tense (10 .1); and sim ple past follow ed by com plex habitual past, as in (10 .2 ). (9) Switches between Present [Participial] and Past Tense 1. Present ~ Past Tense [Present = Participial] a. misphaxat Yisraeli hi omnam misphaxa ktana. yesh ba rak saba, savta., shney yeladim ve zug horim. az zo hi mixphaxa me’usheret ve smexa. hasaba ve ha-savta garim ba-kfar ve le-yadam shoxenet brexat dagim. Yom exad ha-saba hexlit. . . 22 BERMAN ‘ The Yisraeli fam ily is actually just a sm all fam ily. They have only a grandpa, grandm a, two kids and parents. So this is one happy and con­ tented family. The grandparents live in a village, with a fishpond that lies nearby. One day, the grandpa decided . . . ’ [Pnina, girl,12 , 7th grade, pic­ ture-based fiction— continues all in past tense] b. tov. ani sonet se n ogim li ba-dvarim , se m ifaskim li im hadvarim baxeder. yom exad axoti Yael hexlita . . . ‘Okay, I hate it when (people) touch my things, mess around w ith the things in my room. One day my sister Yael decided . . . ’ [Hila, girl, 13, 7th grade, fight story] c. lifney beerex shvuayim noda li, se ani nosa'at le-xul b e-ta a rix se mitnagesh im mivxan be-anglit sheli, az nigashti la-mora le-anglit sh eli .. . ‘About two weeks ago I learned that I go = am going abroad on a date that conflicts w ith my English test, so [= then] I-w ent-up to my English teacher . . .’ [Merav, girl, 1 6 , 11th grade] 2. Past ~ Present [Present = Historical, Narrative Present] etmol halaxti le-male delek ba-oto ha-tsahov ve ksehigati lesham, bederex klal ani memale ki ha-ovdim be-taxanot ha-delek m ifaclim laasot et avo datam. azvacati me haoto ve lakaxti et ekdax ha-delek ve samti oto betox hamexonit, hitxalti le-male delek ve hu mistakel alay kaxa, amarti lo se ani roce lemale shemen, azar li le-male shemen. axarey ze ani ba le-shalem lo im ha-viza, kmo se ani meshalem bederex klal, ve bederex klal ani tamid sam lev . bekicur, hu omer li se ha-m exir ha-kolel hu . .. ‘Yesterday I went to fill up the yellow car with gas, and when I got there, I usually do it because the guys working there are too lazy to do it properly. So I got out o f my car and took the hose and inserted it, I started to fill up, and he looks at me in a weird kind o f way, and I told him I need oil as well, (he) helped me with that. A fter that, I come to pay him, with my credit card, like I always pay, I usually watch what he does, to cut a long story short, he tells me the price is . . . ’ [Hanan, man, 25, fight story, continues in present tense fo r rest o f story, until last 8 clauses out o f 80 = the coda, also in­ troduced by “ to cut a long story short'*] It is not b y chance that there are three exam ples in (9 .1) o f sh iftin g from present to past, from older school children, but on ly one exam ple o f shifting from past to present, from an adult. In general, across our database, there were far m ore exam ples o f the first than o f the latter shift betw een the tw o tense form s. T h is m igh t seem su rp risin g , becau se past tense (b a sica lly p erfective though also possible w ith durative predications in Hebrew) m ight seem better suited to the anterior nature o f background, scene-setting situations. However, as noted, in H ebrew present-tense form s also function as nontensed participials in com plem ent and adverbial clauses expressing attendant circum stances. Thus, from late schoolage, but not before then, narrators showed the im pact o f l Setting the Narrative Scene 23 B ib lica l and oth er lite ra ry fictio n fo r exp ressin g n a rrative te m p o ra lity and b a c k g ro u n d -fo regro u n d distinctions by m eans o f the p articip ial-(gen eric or durative) present for setting versus use o f the m ore com pletive, sequential pasttense form s for narrative events. The reverse exam ple in (9.2) is less typical and reflects a highly individual “ D am on Runyonish” type o f style, which on ly a few adults and none o f the younger subjects adopted. As noted, sh ifting betw een present to past and past to present to m ark o ff sto ry scene-setting from sto ry plo tlin e is o n ly one device used b y p ro ficien t H ebrew -speaking narrators. A nother is by contrasting the sim ple or perfective past w ith the com plex habitual, durative past form . This is overtly m arked by com bining the past-tense form o f the verb haya ‘be’ (or any o f its alternants in 1st and 2nd person, singular vs. plural) as an auxiliary w ith the participial, p re­ sent tense form o f the m ain verb (which agrees with the subject in num ber and gender). T h is quite co m m o n verb form never once occu rred in the H eb rew fro g -sto ry sam p le, and it is extrem ely rare in the co n v ersa tio n al u sage o f p reschool children through age 5 (Berm an & D ro m i, 19 8 4 ). B ut it does serve proficient narrators as a rhetorical option for form ally m arking o ff background settings from the foregrou n d narrative chain o f events, or vice versa, as illu s­ trated in (10 ). (10 ) Shifting between Simple (Perfective) and Com plex (Habitual) Past Tense: 1. Habitual Past [= haya was/were + Benoni Participle’] ~ Simple Past a. ze haya lifney shana, ve yeladim ba-kita sheli havu osim shtuyot, mitkashrim habayta ve ze. az yeled exad xashav se ani hitkasharti elav . . . ‘It was last year, and the kids in my class were doing crazy-things, calling people at home and so on. So [= then] one kid thought that I (had) called him . . . ’ [Tal, boy, 1255, 7th grade] b. havta lixavera axatse hi havta mexatetet b a a f ve ani haviti koseset cipornayim, havinu voshvot axat ley ad ha-shniya, ve haviti rava ita, hayiti omeret la, hayiti tso’eket aleha, ve hi havta omeret l i . . . . a z p a ’am halaxti ita makot bemizderon bet ha-sefer ‘I had a friend in first grade, that was picking [= used to pick] her nose, and I was biting my nails, we were sitting next to each other, and I was arguing with her, I was saying to her, I was shouting at her, and she was saying to me . . . . So once I sim ply got into a fight with her in the school corridor. [Shani, wom an, 23, continues all the rest in simple past tense] 2. Simple Past ~ Habitual Past loh haya lahem musag ex le-tapel be-tinok, az hem masru ota le-imuts, aval ze sipur axer legamrey, ex ve lama kara seasu kax. ha-saba sheli axarey kama shanim nisa le-baxura tse’ira bat 17, ve hava lahem od shney yeladim beyaxad. ha-aba sheli gadal be-mosdot, haya lo aba ve ima aval hu xav kmo ba-sipurim im ima xoreget rashaitse havta me'ira otam be-arba ba-boker ve 24 BERMAN m e'if a otam me ha-mita, kulam hayu kamim be-arba ba-boker lenakot et ha-bayit, hayta meshuga'at le-nikayon. be-shlav mesuvam ha-aba sheli ayar le-kibuts, hu bila sham shanim , pagash sham et ha-ima sh eli . . . ‘They didn’t have a clue how to take care o f a baby, so they gave her up for adoption, but that is a whole other story, how and why it happened that they did so. M y grandfather after a few years m arried a young girl o f 17, and they had another two kids together. My father grew up in institutions, he had a m other and father but he lived like in the storybooks w ith a wicked stepmother who was waking [= used to wake] them up at four in the m orning and (was) throwing them out o f bed, all o f them were getting up at four a.m. to clean the house, she was compulsive about cleanliness .. .. At some point my dad moved to a kibbutz, he spent years there, met my mother there’ [Chaya, woman, 33, telling lifestory to a friend] A lo n g w ith use o f tense/aspect sh iftin g to d istin gu ish b a ck g ro u n d circum stances from foregrounded events, m ore m ature or proficient H ebrew narrators alternate predicates in two additional ways. First, they rely heavily on the verb haya to indicate both copula ‘be’ and possessive ‘have’ in background clauses, in con trast to the le xic ally specific verbs that th ey p refer in the sequential part o f the narrative. Second, narrators use stative-durative verbs as background predicates, and activity or event verbs elsewhere; that is, they make use o f inherent aspect or Aktionsarten distinctions to set o ff backgroun d from foregrounded events. These findings for how H ebrew speakers alternate across predicate types in order to m ark o ff different com ponents o f their stories illustrate several m ore general them es. In c ro sslin gu istic term s, speakers w ill rely on the fo rm a l op tion s m ade available to them by the ty p o lo g ica l stru ctu re o f th eir native language, rather than seeking to use “ co m p en sato ry” p eriph rastic m eans for m arking distinctions not m ade in their gram m ar. On the other hand, proficient speakers, and they alone, resort to a full range o f textual devices for m arking relevant distinctions, across a range o f form s which is not im m ediately obvious from gram m atical or even lexical analysis at the level o f the single sentence. F u rth erm o re, p ro ficien t n a rrato rs d ep loy these devices in a w ay that is not accessible to younger, less proficient speakers in constructing narrative texts. Besides, even am ong fully proficient narrators, use o f these devices is optional rather than obligatory. N arrative texts in H ebrew sound perfectly well form ed if they are constructed entirely in past or in present tense, or w ithout any surface m arking o f habitual past aspect contrasting with sim ple past tense. H owever, the ability to exploit such rhetorical options gives the narratives constructed by skilled narrators a textual flavor, a richness and variety w hich are the hallm ark o f “ good ” storytellers and storytelling. l Setting the Narrative Scene 25 CONCLUSIONS This study has confirm ed findings o f prior research on narrative developm ent to the effect that young preschool children do not appear to recognize the need to provide their audience with relevant backgroun d inform ation. Subsequently, at a m ore structured, m iddle-level phase o f developm ent, narrators provide at least the m inim al background inform ation needed to fram e events in place and tim e, and th ey o c c a sio n ally add m o tivatio n fo r the events th at w ill ensue. H ow ever, m etacognitive com m ents on the task itself and /or on its them atic content or on the script itself (say, o f an adventure story or a personal exp eri­ ence with a conflict situation) are given on ly by m ature narrators, reflecting a quite d ifferent type and level o f com m u nicative com petence. M oreover, as I have noted elsewhere (Berm an, 19 8 8 ,19 9 5 ) , the greatest individual variation is found at the two extrem es, am ong the youngest children and the adults. Som e adults tell stories as straightforw ardly inform ative and well structured as school ch ild ren ’s, w hile oth er adults devote as m uch as 5 0 % o f th eir texts to b a c k ­ ground before proceeding to the onset o f the action. T he question o f w hat ch ild ren ’s n arrative abilities can tell us abou t their knowledge o f language is not a sim ple one, because narrative construction is a d om ain in w hich linguistic structure interacts in com plex ways w ith general cognitive faculties (Berm an St Katzenberger, 19 9 8 ; Shatz, 19 8 4 ). These include the ability to give expression to an internalized narrative schem a in the form o f an action structure organized around a goal or problem , attem pts to m eet this goal, and a resolution. A lso dependent on cognitive underpinnings is the ability to provide adequate and appropriate background inform ation to set the scene fo r the sto ry that is abou t to u n fold . N onetheless, certain c o m m o n them es em erge to illum inate how children develop the ability to use linguistic form s for m eeting such narrative functions. These themes are shared by the findings o f the large-scale crosslinguistic “ frog-story” study o f Dan Slobin and ou r col­ laborators (Berm an St Slobin, 19 9 4 ); by the analysis o f the expression o f tem ­ p orality and connectivity in five different contexts used for narrative elicitation am ong H ebrew -speaking subjects (Berm an, 19 9 5); and by the m ore specialized study o f story-beginnings presented here. First, from the p oin t o f view o f forrm function relations, the sam e surface form s (e.g., the H ebrew counterparts o f once, one day’) fulfil different n arra­ tive functions w ith age. M oreover, som e form s initially serve in only restricted contexts, but w ith tim e com e to m eet a w id er range o f n arrative fu n ctio n s. Thus, yo u n g children use stereotypic lexical item s to introduce their stories, whereas m ature narrators rely on less conventional rhetorical devices to set o ff background orienting elem ents from the m ain storyline. A m on g the youngest narrators, the distinction between backgroun d and foregroun ded elem ents is often unm arked or initially blurred, whereas subsequently it is m arked by rela­ tively n o n ex p licit additive or tem p o ral expressions like those m ean in g ‘and 26 BERMAN (then), after that’. O nly later in developm ent is the transition from scene setting to plot onset clearly m arked by explicit lexical as well as gram m atical devices, including tense/aspect shifting in som e cases. Second, and relatedly, m ost o f the relevant lin gu istic form s are available from early on, for exam ple use o f past tense m arking o f verbs or lexical m arkers o f tem poral sequence like one time, afterwards. Yet even where children do have com m and o f the relevant linguistic form s at the level o f the sim ple clause and, later on, for relating adjacent clauses, it takes them a long tim e to learn how to deploy these form s both flexibly and appropriately in the context o f extended discourse. In the present context, they need to know which linguistic form s to use in ord er to d istin gu ish b a ck g ro u n d scen e-settin g elem en ts from the foreground chain o f narrative events. A nd they m ust do so by using ap p ro p ri­ ate lexical m arkers o f the transition and by flexible shifting between predicate sem antics, tense, and aspect in b a ck g ro u n d versu s fo re g ro u n d elem ents. Fu rth e rm o re , som e fo rm s do not ap p ear to be used at all u n til quite late. Exam ples include use o f the past perfect in English and Spanish (Kupersm idt, 19 9 6 ; Sebastian & Slobin, 19 9 4 ), use o f syntactic passives in Flebrew (Berm an, 19 9 7 b ) and, as shown here, use o f H ebrew habitual past aspect m arking. These findings provide strong m otivation for further exam ination o f the m ore general issue o f “ late acquisitions” and the need to account for the delay in em ergence o f som e form s com pared with others (Berm an, 19 9 8 ; Ravid & Avidor, 19 9 8 ). These findings point to the im portance o f including adult subjects as a basis fo r co m p a riso n and fo r evalu atin g the range o f op tion s used b y p ro ficien t speakers in d ifferen t types o f n arratives. The present stu d y show s that we should include teenage narrators, too, as was done to such fine effect by Labov ( 19 7 2 ) . A d olescen ts in general, and h igh sch o ol students in p articu lar, can illum inate in im portant ways how developing narrative abilities and linguistic fo rm m arra tive fu n ctio n relatio n s are affected b y sch o o l-b ased literacy and increased exposure to and awareness o f different types o f narrative genres and the rhetorical options suited to each one. It seems to take through to adulthood until this knowledge is further incorporated into a personal style and the n a r­ rative stance that each individual selects to deploy in any given context. Next, as in other d om ains o f developm ent, task effect is relevant here, too. Children proved able to m ark o ff setting elements better and earlier in personal experience accounts than in narratives based on a picturebook story. A nd they did so as y o u n g as age 3 w hen th ey were free to tell abou t an yth in g that had happened to them , but only from around age 5 when asked to tell specifically about a fight they had experienced. In general, personal experience accounts appear to provide m ore authentic contexts for elaborating on scene setting than do picture-based tasks. These prelim in ary findings indicate that, as noted, the m ethodological and developm ental issue o f task effect could be illum inated by in-depth, suitably controlled studies o f how and when each setting elem ent is expressed across different narrative genres and in different elicitation settings. l Setting the Narrative Scene 27 A dditional avenues for further research that em erge from this study are indepth exam ination o f crosslinguistic and crosscultural differences that m ight affect h ow scene-setting circum stances as distinguished from plotline events are expressed across developm ent, for exam ple, in languages w ith rich tense/as­ pect d istinctions or in cultures with h igh ly conventionalized form ats for this purpose. Finally, as a possible source o f new insights in the dom ain o f general as well as developm entally m otivated narrative research, it would seem o f inter­ est to com pare such analyses o f scene-setting elem ents w ith the extent and w ay in which children and adults give expression to the coda in different types and contexts o f narrative production. A C K N O W L E D G M ENTS Research on which this study is based was funded by grants from the U .S.-Israel B in a tio n a l Science F o u n d atio n to R. A. B erm an and D. I. S lo b in o f the University o f C alifornia, Berkeley, from the Linguistics Program o f the N ational Science Foundation to D. I. Slobin, and from the Israel Science Foundation to R. A. Berm an and D. D. R avid o f Tel A viv University. The author is indebted to students who participated in her sem inar on narrative developm ent at Tel Aviv University in the years 1 9 9 5 - 1 9 9 7 for their assistance in p roviding data, to Dr. Irit Katzenberger for her invaluable help in data collection and analysis, and to Iris Levin and Yeshayahu Shen o f Tel Aviv U niversity as well as to the editors o f this volum e for their valuable com m ents on earlier drafts. R esponsibility for re­ m aining inadequacies rests w ith the author alone. REFERENCES Argem an, K. (19 9 6 ). Sipurey bdaya shel talmidim b e-a lp e u bixtav [Schoolchildrens’ m ake-believe stories in speech and writing]. Tel Aviv University sem inar paper. Bam berg, M ., & D am rad-Frye, R. (19 91). On the ability to provide evaluative com ments. Journal o f C hild Language, 18, 6 8 9 - 7 10 . Bazanella, C., & C a lle r i, D. (19 9 1). Tense coherence and grounding in children’s narratives. Text, 11, 175 - 18 7 Ben -H aviv, O. ( 19 9 6 ) . H ashva’a beyn rexiv h a - ” reka” be-sipureyhem shel yeladim bney 3, 7, ve 12 shanim ve shel mevugarim [A com parison o f the “ setting” com ponent in the narratives o f chil­ dren aged 3, 7, and 12 years and adults]. Tel Aviv U niversity sem inar paper. Berm an, R. A. (19 8 8 ). On the ability to relate events in narrative. Discourse Processes, 11, 4 6 9 - 4 9 7 . Berm an, R. A. (19 9 3 ). The developm ent o f language use: Expressing perspectives on a scene. In E. D rom i (Ed.), Language and cognition: A developm ental perspective (pp. 1 7 2 - 2 0 1 ) . N orw ood, NJ: Ablex. Berm an, R. A. (19 9 5). Narrative competence and storytelling perform ance: how children tell stories in different contexts. Journal o f Narrative and Life History, 5, 2 8 5 -3 14 . Berm an, R. A. (19 9 6 ). Form and function in developing narrative skills. In D. I. Slobin, J. Gerhardt, A. K yratzis, & J. G u (E d s.), Social interaction, social context, and language (pp. 2 4 3 - 2 6 8 ) . M alwah, N J: Lawrence Erlbaum Associates. Berm an, R. A. (19 9 7 a). N arrative theory and narrative development: The Labovian impact. Journal o f Narrative and Life History, 7, 2 3 5 - 2 4 4 . 28 BERMAN Berm an, R. A. (19 9 7b ). Preliterate knowledge o f language. In C. Portecovo (Ed.), Writing develop­ ment: An interdisciplinary view (pp. 6 1 - 7 6 ) . Am sterdam : John Benjam ins. Berm an, R. A. (19 9 7 c ). D eveloping form /function relations in narrative text production. Lenguas Modernas, 24, 4 5 - 6 0 . Berm an, R. A. (19 9 8 , January). Emergence and mastery in language development. Talk presented to Linguistics C olloqium , Tel Aviv University. Berm an, R. A., (19 9 9 ). Bilingual proficiency/proficient bilingualism : Insights from Hebrew-English narratives. In G. Extra 8c L. Verhoeven (Eds.), Bilingualism and migration (pp. 18 7 - 2 0 8 ). Berlin: M outon de-Gruyter. Berm an, R. A. (in press). The role o f context in developing narrative abilities: The frogstory findings in light o f other narrative genres and elicitation settings. In S. Strom qvist 8c L. Verhoeven (Eds.), Relating events in narrative: Variation across languages, cultures, and genres. M ahw ah, N J: Lawrence Erlbaum Associates. Berm an, R. A., 8c D rom i, E. (19 8 4 ). On m arking time w ithout aspect in child language. Papers and Reports on Child Language Development, 23, 2 1 - 3 2 . B erm an, R. A., 8c Katzenberger, I. (19 9 8 ). Cognitive and linguistic factors in developm ent o f p ic­ ture-series narration. In A. G. Ram at 8c M. Chini (Eds.), O rganization o f learners’ texts, Studia italiani i linguistica theoratica e applicata (special issue), 27, 2 1 - 4 7 . B erm an, R. A., 8c Slobin, D. I. (19 9 4 ). Relating events in narrative: A crosslinguistic developm ental study. Hillsdale, N J: Lawrence Erlbaum Associates. Chvany, C. V. (19 8 5 ). Backgrounded properties and plotline im perfectives: Towards a theory o f g rou n d in g. In M .S. Flier 8c A. T im b erlake (Eds.), The scope o f Slavic aspect (pp. 2 4 7 - 2 7 3 ) . C olum bus, Ohio: Slavica. Giora, R., 8c Shen, Y. (19 9 4 ). Degrees o f narrativity and strategies o f semantic reduction. Poetics, 22, 447-458. H erm an , J. ( 19 9 6 ) . (<G renouille, where are yo u ?” Cross-linguistic transfer in bilingual kindergartners learning to read. U n publish ed d octoral d issertation , H arvard G rad u ate Sch ool o f Education. H ickm ann, M. (19 9 1). The development o f discourse cohesion: Som e functional and crosslinguis­ tic issues. In G. Pieraut -Le-B on niec 8c M. D olitsky (Eds.), Language bases ... discourse bases: some aspects o f contem porary French language psycholinguistic research (pp. 1 5 7 - 1 8 5 ) . Amsterdam : John Benjam ins. H ickm ann, M . , H endriks, H., Roland, F., 8c Liang, J. (19 9 6 ). The m arking o f new inform ation in children’s narratives: A com parison o f English, French, Germ an, M andarin Chinese. Journal o f C hild Language, 23, 5 9 1-6 2 0 . H ickm ann, M ., Kail, M ., 8c Roland, F. (19 9 5 ). Cohesive and anaphoric relations in French children’s narratives as a function o f m utual knowledge. First Language, 15, 2 7 7 -3 0 0 . Hopper, P. S. (19 8 2). Aspect and foregrounding in discourse. In T. Givon (Ed.), Discourse and syn­ tax: Syntax and semantics, Vol. 12 (pp. 2 1 3 - 2 4 1) . New York: Academ ic Press. Kahanowitz, S. (1995)- D arxey ha-rifrur be-sipurey yeladim u mevugarim JExpression o f reference in the stories o f children and adults]. Tel Aviv University sem inar paper. Kail, M., 8c Hickm ann, M. (19 9 2 ). French children’s ability to introduce referents in narratives as a function o f m utual knowledge. First Language, 12, 7 3 - 9 4 . Kail, M ., 8c Sanchez-Lopez, I. (19 9 7 ). Referent introductions in Spanish narratives as a function o f contextual constraints: A crosslinguistic perspective. First Language, i j , 1 0 3 - 1 3 0 . K atzenberger, I. ( 19 9 4 ) . H ayexolet lesaper sipur al p iy sidrat tmunot: hebetim kognitiviyim , leshoniyim, ve hitpatxutiyim [Cognitive, linguistic, and developmental factors in the narration o f picture series]. Tel Aviv University unpublished doctoral dissertation. Kern, S. (19 9 7 ). Comm ent les enfrants jonglent avec les contraintes communicationnelles, discursives et linguistiques dans la production d'u n e narration [H ow children juggle com m u n ication , d is­ course, and linguistic constraints in the production o f a narrative]. Unpublished doctoral dis­ l Setting the Narrative S cene 29 sertation, Universite Lum iere-Lyon 2. Kernan, K. T. (19 77 ). Semantic and expressive elaboration in children’s narratives. In S. Ervin-Tripp & C. M itchell-Kernan (Eds.), C hild discourse (pp. 9 1 - 1 0 4 ) . New York: Academ ic Press. K u persm itd , J. ( 19 9 6 ) . Tense-aspect m arking in Spanish-H ebrew b ilingual narratives. Tel A viv University sem inar paper. Labov, W. ( 19 7 2 ) . T he tran sfo rm atio n o f experience in n arrative syntax. In W. Labov (E d .), Language in the inner city (pp. 3 5 5 - 3 9 6 ) . Philadelphia: U niversity o f Pennsylvania Press. Mayer, M. (19 6 9 ). Frog, where are you? New York: Dial Press. M enig-Peterson, P., & M cCabe, A. (19 78 ). Children’s orientation o f a listener to the context o f their narratives. D evelopm ental Psychology, 14, 58 2-9 2. M in am i, M. ( 19 9 6 ) . Japanese p resch ool children ’s n arrative d evelopm ent. First Language, 16, 339-364Oz, A. (19 9 6 ). M atxilim sipur [The story begins]. Jerusalem: Keter Publishers. Peterson, C. (19 9 0 ). The who, when, and where o f early narratives. Journal o f C hild Language, 17, 433-456. Peterson, C., & M cCabe, A. (19 8 3 ). Developm ental psycholinguistics: Three ways o f looking at a child’s narrative. New York: Plenum. Peterson, C ., & M cCabe, A. ( 19 9 1) . Linking connective use to connective m acrostructure. In A. M cC abe & C. Peterson (Ed s.), D eveloping narrative structure (pp. 2 9 - 5 4 ) . H illsd ale, N J: Lawrence Erlbaum Associates. Pitcher, E. G ., & Prelinger, E. ( 19 6 3 ) . C hildren tell stories: An analysis o f fantasy. N ew York: International Universities Press. Pradl, G. M. (19 7 9 ). Learning how to begin and end a story. Language Arts, 56, 2 1 - 2 5 . Polanyi, L. (19 8 5). Telling the American story. Norw ood, N J: Ablex. Ravid, D. (19 9 5 ). Language change in child and adult Hebrew: Psycholinguistic perspectives. O xford, England: O xford University Press. Ravid, D., & Avidor, 8 .(19 9 8 ). Acquisition o f derived nom inals in Hebrew: Developm ental and lin ­ guistic principles. Journal o f C hild Language, 25, 2 2 9 - 2 6 6 . Reilly, J. S. (19 9 2 ). H ow to tell a good story: The intersection o f language and affect in children’s narratives. Journal o f Narrative and Life History, 2, 3 5 5 -3 7 7 . Reinhart, T. (19 8 4 ). Principles o f gestalt perception in the tem poral organization o f narrative texts. Linguistics, 22, 7 7 9 -8 0 9 . Reinhart, T. (19 9 5 ). M i-tekst le-m ashm aut: Em ca’ey ha’araxa [From text to meaning: Strategies o f evalu ation ]. In Y. Shen (Ed .), Cognitive aspects o f narrative structure (pp. 4 - 3 7 ) . Tel A viv University: The Porter Institute [in Hebrew]. R u m elh art, D. E. (19 7 5 ) . N otes on a schem a for stories. In D. G. B o b ro w & A. C o llin s (Ed s.), Representation an d understanding: Studies in cognitive science (pp. 2 1 1 - 2 3 6 ) . N ew York: Academ ic Press. Said, E. A. (19 78 ). Beginnings: Intention and method. Baltim ore: Johns Hopkins University Press. Sebastian, E., & Slobin, D. I. (19 9 4 ). Developm ent o f linguistic form s: Spanish. In R. A. Berm an & D. I. Slobin (Ed s.), R elatin g events in narrative: A crosslinguistic developm ental study (pp. 2 3 9 - 2 8 4 ) . Hillsdale, N J: Lawrence Erlbaum Associates. Segal, R. ( 19 9 6 ) . K r i’at sefer “ keilu” bekerev yald ey gan b i-tn a ’ey d ivu v shonim [K in d ergartn ers, “ pretend-reading” in two different conditions]. Tel Aviv University sem inar paper. Seidm an, S., Nelson, K., & Gruendel, J. (19 8 6 ). Make believe scripts: The transform ation o f ERS in fantasy. In K. N elson (Ed .), E vent knowledge: Structure and fun ction in developm ent (pp. 1 6 1 - 1 8 8 ) . Hillsdale, N J: Lawrence Erlbaum Associates. Shatz, M. (19 8 4 ). A song w ithout m usic and other stories: H ow cognitive process constraints affect children’s oral and written narratives. In D. Schiffrin (Ed.), M eaning, form , and use in context: Linguistic applications (pp. 3 1 3 - 3 2 4 ) . Washington, D C : Georgetown University Press. Shen, Y. (19 8 8 ). The X -B ar gram m ar for stories: Story gram m ar revisited. Text, 9, 4 15 -4 6 7 . 30 BERMAN Shen, Y., & Berm an, R. (19 9 7 ). From isolated event to global action-structure. In J.Shim ron (Ed.), Psycholinguistic studies in Israel: Language acquisition, reading and w ritin g (pp. 1 1 9 - 1 4 5 ) . Jerusalem: The M agnes Press, the Hebrew University [in Hebrew]. Slobin, D. I. (19 9 3). Passives and alternatives in children’s narratives in English, Spanish, Germ an, and T urkish. In B. Fox & P. J. H op p er (Ed s.), Voice: Form and fun ction (pp. 3 4 1 - 3 6 4 ) . Am sterdam : Benjam ins. Sternberg, M. (19 78 ). Expositional modes and temporal ordering in fiction. Baltim ore: John Hopkins University Press. Sutton-Sm ith, B. (19 8 1). The folk stories o f children. Philadelphia: University o f Pennsylvania Press. U m iker-Sebeok, D. J. (19 7 9 ). Preschool children’s intraconversational narratives. Journal o f C hild Language, 6, 9 1 - 1 0 9 . Van D ijk, T. (19 7 6 ). Philosophy o f action and theory o f narratives. Poetics, 5, 2 8 7 -3 3 8 . W igglesworth, G. (19 9 2 ). Investigating children s cognitive and linguistic developm ent through narra­ tive. Unpublished doctoral dissertation, La Trobe University, Australia.