Discourse skills of bilingual children: precursors of literacy

Discourse skills of bilingual children: precursors of literacy RICHARD P. DURAN Children's exhibition of discourse skills The ability of non-English-background children to acquire an education in English is not amenable to simple analysis. On commonsense grounds, no one expects children to learn in a classroom if they cannot understand language as it is used in the classroom. The general public and even many educators are under the impression that the educational problems of non-English-background children can be adequately understood in terms of chüdrens' lack of proficiency in English. In the past it has been contended that the solution to children's educational problems involves teaching children standard English and that this problem reduces to teaching children English grammar, punctuation, pronunciation, vocabulary, and other basic English skills. In recent years, such a perspective has begun to be rejected as adequate in and of itself. Research in the areas of sociolinguistics, ethnography of communication, cross-cultural psychology, psycholinguistics, and most recently, cognitive schema theory suggests that we cannot understand children's acquisition of a second language in simple terms related only to the grammar and structure of a second language. Children's ability to acquire a language is not a simple function of their learning structural rules of a language. Acquisition and learning of a new language system takes place in a broader context defined by society and by the realization of sociocultural factors in a community. Norms for social uses of a language are integral with the development of competency in a language. Competency in a language is tied intimately to social interaction and to ways of acting in interaction. Hymes (1972) and Ervin-Tripp (1964), among many others, have pointed out that communicative events can be described in terms of a set of specific factors that influence how language is used in everyday settings. When children learn to become competent speakers of a new language, they are able to accomplish purposes of communication that fulfill the social roles within communicative situations. All elements of language use are possibly affected by the social characteristics of a situation. 0165-2516/85/0053-0099 $2.00 © Mouton Publishers, Amsterdam Intl. J. Soc. Lang. (1985), pp. 99-114 Brought to you by | University of Michigan Authenticated Download Date | 6/17/15 6:14 PM 100 R. R Duran Grammar, word choice, and pronunciation of words can vary under the influence of social and situational variables. In addition, other qualities of language use may vary. For example, rhythm, pace of delivery, use of intonation, and key, i.e. affective connotation of speech, may vary according to the social characteristics of a communicative event and its unfolding. Among the most important factors influencing language use in a communicative event are setting and location, functions or purposes of communication, persons and their social roles and relationships to each other, and the domain of topics that are the object of interaction. Subtle changes in the way the foregoing factors are realized can have a dramatic impapt on how children use language and on how well they can communicate. This paper reviews research findings on the foregoing topics, especially ability to deliver and understand narrative, which are important to understanding factors affecting Hispanic children's literacy development. While Hispanic children are the focal population of concern, a good many of the issues and findings which are reviewed emanate from research on ethnicminority children and children in general. A lot of attention is given to language behavior in classrooms since this domain has been studied most extensively. Narratives Special attention is given to narrative behavior because narratives are one of the earliest and most important kinds of language use that children encounter in their early years. Stories, as a form of narrative, are likely to provide children with their earliest experiences in language as a medium for communication of ideas referring to information beyond their immediate perceptual environment. A narrative discourse describes a history of events that occur over time. Typically, narratives are about people and their activities, as depicted, for example, in a story or recounting of past experiences. Children first encounter narratives when they hear stories told to them by their parents and other caretakers. Narratives have a structure suited to their expository purpose. Stories in our culture, for example, have a structure characterized by a beginning, middle, and end. The beginning of a story sets the stage for the rest of a story; it introduces the major story characters, characteristics of the story setting, and the initial situations which motivate the direction and subsequent scenes of a story. The middle of a story depicts a progression of story events; these events are related thematically to each other. Typically, there are some critical problems or goals which major story characters must contend with. The middle of a story describes this information and goes on to develop the complications which ensue in the lives of Brought to you by | University of Michigan Authenticated Download Date | 6/17/15 6:14 PM Discourse skills of bilingual children 101 story characters as they act to solve problems and reach goals. The end, or final part of a story, centers upon one or more critical problems, crises, or goals, which are resolved in one way or another near the end of a story and which bring a story to its conclusion. Researchers in children's discourse analysis and in adults' language have developed more sophisticated and analytically useful descriptions for structure of story narratives than the sketch that has been given (see Stein and Glenn 1979; Labovl972 for examples). Research has found that speakers and listeners in a story-reading or story-telling event pay close attention to which parts of a story are being referred to by a speaker and to how the parts of a story are integrated together. These findings imply that children develop a lot of knowledge about how to use and understand language that cannot be explained considering only grammar, pronunciation, and other surface features of a language. Another reason why children's mastery of narrative skills is important is because this mastery places sophisticated cognitive demands on children. Children need to develop an understanding of how words and speech can represent imagined situations and events. The world of a story, for example, is not the same as the world in which a story is heard. In becoming acquainted with narratives, children learn that they can use their minds to understand talk or to deliver talk about situations that are not before them. They learn that language is a medium for expression of ideas and that these ideas can describe imaginary or hypothetical events. Narratives thus act as a bridge between children's early language experiences and language experiences that they will encounter in academic settings. In the review which follows, evidence will be brought out which demonstrates how narrative production is related to findings on children's general repertoire of linguistic and sociolinguistic skills, as well as to research findings specifically concerning narratives. The final section of the paper will describe a new avenue for research based on cognitive schema theory which helps tie children's knowledge. about speech events to their communicative performance. The new approach which is outlined is based on recent cognitive science research on memory and discourse behavior. The new approach is also valuable because it draws attention to how children acquire cognitive skills related to their performance of literacy activities. Research on Hispanic and minority children's discourse, cultural influences on classroom communication, and communication in other contexts A recent overview of cultural factors in communication which influence minority children's ability to participate in classroom settings is provided by Brought to you by | University of Michigan Authenticated Download Date | 6/17/15 6:14 PM 102 R. P. Duron Saville-Troike (1980). Some factors mentioned by Saville-Troike which are culturally dependent include voice volume or level; direction of gaze during discourse reception or production; rules for turn taking in discourse; use of formulaic utterances for purposes such as mitigating requests; variety of language permissible in a setting; and norms for responding to questions or commands. As Saville-Troike (1980) indicates, there is no exhaustive catalog of all the cultural factors which might influence communication in classrooms. Indeed, serious pursuit of such a catalog for any one ethnic group might not lead to a historically stable taxonomy since the very phenomena under study are affected by ongoing changes in cultural accommodation to life in the US and social patterns of cultural and language contact among persons of diverse ethnic identity. Perhaps the best known ethnographic study of cultural influences on ethnic-minority children's discourse behavior is by Philips (1972). This researcher found that American Indian children from the Warm Springs Indian reservation in Central Oregon were reluctant to publicly answer teachers' questions in the classroom, but that American Indian children were found to actively interact with others in group classroom activities. Philips's research on the structure and communication patterns of children outside the classroom revealed that American Indian children had norms for communication which made public question answering inappropriate based on rules for question answering learned in home settings. In addition, it was found that American Indian children preferred to interact with other children, rather than with adults who imposed their own rules for communication. Philips introduced the notion of participant structure to refer to the collective ways of communicating shared among persons habitually sharing speech in the same social setting. Research studies of Hispanic children's discourse behavior have yielded many interesting findings and motivated further work contrasting children's language behavior across home and school settings. One of the earliest ethnographic studies of Hispanic children's school discourse was by Brück and Shultz (1977) in collaboration with Flora Rodriguez-Brown. This research collected data on first-grade children's use of Spanish and English in a halfday 'pull-out' transitional bilingual program. Results of the study showed that children's use of English increased over years in school and that the bilingual teachers' language dominance affected language choice by children. The work of Carrasco et al. (1981) and Carrasco (1979) found that in teacher-bilingual student discourse, Hispanic children in some circumstances were perceived by teachers as less articulate and skillful in school subject matter than they really were. When the participant structure for children involved peer tutoring in relatively unsupervised settings, children who were judged as relatively inarticulate by teachers manifested considerable linguistic Brought to you by | University of Michigan Authenticated Download Date | 6/17/15 6:14 PM Discourse skills of bilingual children 103 fluency and detailed knowledge of school subject matter in acting out the role of tutor to other children. Research by Lindholm and Padilla (1981) and Laosa (1982) found that Mexican American children's discourse in home settings is intrinsically related to mothers' teaching of social and cognitive knowledge to preschool children. Careful analysis of topic of discourse or of function of utterances in problem solving in these studies revealed that mothers' interaction with children is a potentially very powerful determinant of cognitive development of children, ultimately affecting children's adaptation to school. The Lindholm and Padilla (1981) research, for example, revealed that over 30% of parents' speech to children in the home context could be identified as 'mastery skill' communication of value to children's learning of critical social and cognitive skills. The findings of studies such as the foregoing lead naturally to concern for the particular discourse strategies which children find necessary or helpful in their communication. An important study by Wong Fillmore (1979) demonstrated that six- and seven-year-old Mexican immigrant children acquired their initial facility in English in classroom settings by following a variety of cognitive and social strategies in interaction with classmates who could only speak English. Consistent with a great deal of recent literature on pragmatics and children's discourse, Wong Fillmore found that the social and pragmatic structure of communicative events was very important in guiding children's initial acquisition of English. Children learned English most quickly when they strove to become participants in communication. Children sometimes followed a strategy of approximating appropriate utterances which they might make in communicating with others. For example, children displayed a strategy of using and refining formulaic utterances (repeated grammatical frames for performing discourse functions) in their interaction with other children, and this helped them participate in communication and learn English more quickly. Wong Fillmore concluded that there was a good deal of variation among children in their propensity to use different strategies. Children who were inclined to participate socially with other children showed more willingness to try communication strategies which were only partially well formed but which had a clear communicative goal. Use of such strategies accelerated children's learning of English. A study by Moll (1978) found that eight- and nine-year-old Mexican American children's success in communicating description of objects to other children was affected by the ethnicity and language background of the other children. Mexican American children were more successful in communicating with children of the same age regardless of the language they used. More ambiguity of meaning resulted when the 'listener' children were younger Anglo children, aged five or six, or when they were older Hispanic Spanishdominant children, aged 11 to 12. Moll interpreted his results in the light of Brought to you by | University of Michigan Authenticated Download Date | 6/17/15 6:14 PM 104 R. P. Duran the social contact patterns of the community he studied and concluded that lack of social contact and familiarity between his subject children and the latter two groups mentioned weakened the shared background necessary for successful communication to occur. Moll concluded that shared language background alone was inadequate to account for relative success in communication among his children. In a later research, Moll et al. (1980) reported results from an ethnographic study of bilingual third-grade students' interaction with English- or Spanish-language reading teachers. They found that students in a Spanish reading class appeared to be learning more sophisticated literacy skills than students in an English-only reading class. Students in the Spanish reading class were given more opportunity by their teacher to engage in high-level tasks such as preparation of book reports or opportunities to answer openended questions about what they read. In contrast with these sorts of tasks, third-grade students who attended the English reading class had less emphasis placed on high-level tasks of the sort mentioned. Instead, English reading class students (regardless of their assessed English reading level) were given more practice in word decoding and pronunciation. Moll et al. (1980) concluded that the organization of the English reading lessons they observed placed too much emphasis on 'accurate' pronunciation of English words even though children had only minor difficulty in decoding of words and in receptive understanding of written sentences. The resulting focus on pronunciation activities in English inhibited students from acquiring, practicing, and demonstrating high-level reading skills. Some recent studies have touched on the importance of the organization of speech events and their influence on Hispanic children's display of discourse skills. Eisenberg (1982), for example, observed the discourse of three two-year-old Mexican American children at home. The three children were observed and tape-recorded carrying out a number of social speech activities in Spanish or English in everyday interactions with a caretaker (usually the mother). Eisenberg found that there was a close coupling between caretakers' control of their own speech and children's responsiveness to adults' speech. Similar to the research findings of Lindholm and Padilla (1981) and Laosa (1982), Eisenberg discovered that communication between children and adults was often oriented toward teaching children social and cognitive skills about language use and behavior. Eisenberg also found that adults were careful to monitor children's ability to maintain a conversation; adults, at home, deliberately tried not to restructure or correct children's delivery of narratives and this helped children maintain their continuity of speech. Some aspects of children's speech in English and Spanish seemed to be less influenced by social interaction features than other aspects. The two-year-old children observed were not capable of telling a coherent story; their early level of Brought to you by | University of Michigan Authenticated Download Date | 6/17/15 6:14 PM Discourse skills of bilingual children 105 cognitive development did not permit them to describe events occurring one after the other, and they showed difficulty in manipulating verb tense and aspect in delivering narratives. Eisenberg found that her children's ability to tell stories coherently depended on their ability to develop conversational skills that were specifically needed in story telling; these skills seemed partially independent of children's grammatical skills. Slaughter and Bennett (1982) also found that Mexican American children's perception of the social structure of interaction affected their fluency in Spanish and English. A sample of grade one through grade five children at different levels of English and Spanish proficiency was studied. Interactions between children and a bilingual Hispanic adult were observed in a school setting. Children engaged in informal conversation about familiar topics, told stories from a wordless picture book, and played a board game in interactions with an adult. Children varied dramatically in their fluency and utilization of discourse skills in activities. The conversational strategies of the adult interviewer had a strong impact on children's discourse in both languages. Some children who were low in proficiency in either language were reluctant to participate in a dialogue with the interviewer. This was especially noticeable when the interviewer fell into a pattern of using questions to prompt children for more information. At these times, children appeared to deliberately respond with one word answers such as 'yes' or <no'» with no subsequent development of a reply; Slaughter and Bennett termed such behavior 'shutdown' strategies. Slaughter and Bennett gave special attention to analyzing children's discourse strategies in delivering oral story narratives. Their analysis focused on ways in which children's language marked relationships between elements of a story. They found five discourse variables that seemed to be most important in this regard. These variables were clause type used to mark what part of the story was being described; verb structures and verb tenses used to mark when story actions took place; interclausal connectors joining descriptions of different story events; paralinguistic cues such as prosody, intonation, and stress used to emphasize new versus given information, shifts in story focus, etc.; and selection of details for presentation in a narrative. Slaughter and Bennett remarked that a simple frequency tally of the occurrences of cues belonging to the foregoing categories was not as informative as studying the coordination and integration of cues into the stream of the narrative event as it unfolded. Each narrative event had to be studied as a whole in order to learn how effectively children used various cues and strategies in developing a story. The manipulation of verb tenses in delivery of narratives was investigated by Lavandera (1981). She investigated the use of the past tense in Spanish among young Mexican American adults telling stories about past experiences Brought to you by | University of Michigan Authenticated Download Date | 6/17/15 6:14 PM 106 R. R Duran to each other. The subjects Lavandera studied were bilingual in English and Spanish, but they judged that they did not have a highly literate or sophisticated knowledge of Spanish. Using a model for analyzing narrative structure developed by Labov (1972), Lavandera found that subjects manipulated their use of the imperfect and preterite past tense according to the part of a narrative that was being told. The imperfect tense was most used in development of a story line, while the preterite was most used to describe events that occurred one after the other in the main body of a story. Differential reliance on the imperfect and preterite verb tenses in different parts of a story revealed a thorough knowledge of the difference between these tenses in Spanish. English has no similar contrast in its past-tense structure, and thus, according to Lavandera, her subjects' preference for Spanish in storytelling among friends may have served two purposes. One purpose was as a signal of social affiliation, while a second purpose was to use the most effective language to communicate nuances in story structure. Spanish was preferred over English because it featured syntactic devices that allowed for more subtle communication about ordering of events in a story than was possible in English. Several previous studies of narrative behavior in children deserve mention; some of these studies have included ethnic-minority children, but none Spanish-English bilinguals. Kernan (1977), in a study of narrative production among 7- to 8-, 10- to 11-, and 13- to 14-year-old black females, found that the older children devoted more attention to setting up the background for narrative accounts. Older children were also found to link description of successive events in a narrative more frequently by use of causative terms, such as 'so', 'and so', 'so then', rather than by simple linkage terms, such as 'and', 'then', 'and then', which were preferred by younger children. Younger children were also found to interject more expressive elements, i.e. verbal and other cues revealing their feelings about the events that were being described. Collins and Michaels (1980), in a contrastive study of narrative behavior among first- and fourth-grade children, found that literacy training in school affected the way children delivered narratives. Younger children relied more on prosodic cues involving rhythm and cadence to signal agent focus, causal connections, and old versus new information in a narrative. In contrast, older children relied more on formal cohesive devices (Halliday and Hasan 1976) involving use of anaphora, relexification, and lexical labeling of causal or other linkages among components of a narrative. Collins and Michaels (1980) pointed out that use of paralinguistic cues by younger children required a match between the structure and style of a narrative, an audience's knowledge of how such cues are used, and knowledge by the child of how to produce such cues. Collins and Michaels (1980) emphasized that shared cultural backgrounds were thus critical to interpretation of paralinguistic Brought to you by | University of Michigan Authenticated Download Date | 6/17/15 6:14 PM Discourse skills of bilingual children 107 cues accompanying speech and that proper delivery and reception of such cues was basic to the understanding of meaning relayed by a narrative discourse. Gumperz and Kaltman (1980) have provided an extensive theoretical discussion of how to go about analyzing intonational and prosodic cues in discourse. In their work the term 'prosody' is used to include intonation, i.e., pitch levels on individual syllables and their combination into contours; changes in loudness; stress, a perceptual feature generally comprising variations in both pitch and loudness; variation in vowel length', phrasing, including utterance chunking and accelerations and decelerations within and across chunks; and shifts in overall pitch register [underlining added for emphasis]. Production and interpretation of intonational and prosodic cues in speech are governed by speaker's and listener's individual conversational inferences that guide the negotiation of communication in a specific context and setting of communication. Gumperz and Kaltman (1980) gave extensive examples of how prosodic cues can affect the semantic interpretation of discourse and how differences in the use of prosodic cues among communicants from different cultural background can affect the meaning and pragmatic function of utterances. In another paper, Gumperz (1981) described prosodic cues used by firstand second-grade black lower-class children and white middle-class children in delivery of narratives. Some results of this work showed that these two groups of children used and interpreted intonational and prosodic cues differently in the classroom and that teachers proved insensitive to usage of such cues not consistent with their own sociocultural background. Cook-Gumperz and Gumperz (1981) present an analysis of how literacy development in children was related to oral speech practices (prosodic and intonational cues included) and cultural background. Mention was made that use of formulas ('a group of words that have a common network of syntactic patterning') accompanied by prosodic and intonational routines helped young children learn about discourse structure and discourse constraints. Children, for example, learned that repeated reliance on a formula such as 'x was y', with appropriate prosody and intonation could be used to create a crude narrativelike description when the slots represented by ÷ and y were replaced by words — as in The dog was happy', The park was trees', 'It was lovely'. With increasing language experience and development children begin to expand their repertoire of formulas and their structure and to refine their usage of formulas to conform to syntactic requirements of a language. The importance of setting and participants in children's discourse is Brought to you by | University of Michigan Authenticated Download Date | 6/17/15 6:14 PM 108 R. R Daran exemplified by the work of Cook-Gumperz and Corsaro (1977) and McDermott et al. (1978). These researchers demonstrated how production and interpretations of children's utterances often make sense only in the environmental context within which speech occurs. In particular, the latter workers gave attention to the communicative significance of nonverbal interactive behavior, such as posture and direction of gaze. Analysis of acts of oral reading by children revealed that extralinguistic cues of the sort mentioned were often important signaling devices about how an event of Oral reading' begins, proceeds, and ends. The McDermott et al. (1978) work demonstrated that extralinguistic cues are generated and interpreted simultaneously by all children present in an oral reading activity and that such cues were used to 'fine tune' how the reading activity was to proceed. Some studies on children's communicative competence have stressed the correspondence between structure of a discourse activity and the proper sequence of turn taking in discourse. The work of Mehan (1979) on children's classroom discourse, for example, found that teachers' and children's utterances of speech acts fit patterns determined by the nature of an interaction and its progression. Garcia and Carrasco (1981) using Mehan's (1979) model of teacher-student recitation interaction found that mothers of bilingual children and their children followed progressions similar to those between teacher and students, when mothers were practicing language arts tasks with children using both Spanish and English. Au (1980) reported a finding of the Kamehameha Early Education Program demonstrating that Hawaiian-background children improved in their English reading because they incorporated elements similar to a Hawaiian cultural routine, known as 'talk story' in their classroom reading lessons. Research showed that children's enthusiasm and participation in classroom reading was enhanced by allowing children in a reading audience to interrupt a reader in order to add comments or to engage in joking conversation with the narrator. An important component of such reading interactions was the functional role of '... already musical contours of Hawaiian English ... (which) create a speech contour which resembles chanting ...' (Watson 1975). The Kamehameha research indicates that minority-background children may benefit from bridging between culturally based ways of delivering narration and majority culture rules for the conduct of reading lessons, with a benefit to reading and literacy development in academic settings. Schema theory approaches The research which has been cited, beyond suggesting the wide repertoire of skills which children master in developing narrative skills, raises fundamental Brought to you by | University of Michigan Authenticated Download Date | 6/17/15 6:14 PM Discourse skills of bilingual children 109 questions about how a culture and society influence the development of literacy competencies in children. Research by Scollon and Scollon (1981) and Heath (1982) points out that conceptions of literacy are closely connected with norms and ways of communication in a community and social group. There need not be an intimate and clear link between children's learning of narrative as a form of discourse and children's development of literacy skills required in academic settings. This is so despite the fact that the cognitive and some linguistic demands of narrative delivery require use of skills of value to acadernic literacy. Edelsky (in press) points out that literacy activities such as classroom reading form wholes unto themselves. Literacy activities are part of the fabric of school experience; they become familiar and are exercised as important ways of acting and behaving in the stream of classroom life. According to Edelsky, reading skills cannot be assessed adequately by traditional reading tests since such tests emphasize assessment of discrete performances that are separated from the purposeful stream of classroom life. In order to understand how literacy skills develop in language-minority children, attention is needed for the whole — i.e. for children's knowledge of the values and functional outcomes of knowing how to read and use language as an integral part of important daily activities. This concern for the whole, in turn, raises questions about connections between children's social and academic development. How do children learn that reading and writing activities are an important part of their self-identity? While educational linguistics and bilingual education as fields may contribute some of the needed answers, there is also value in psychological approaches to the underlying issues. Cognitive schema theory (Norman et al. 1975) provides a useful theoretical perspective for describing how children's knowledge about how to speak and how to write is integrated with their knowledge about the structure of social situations and communicative events. According to schema theory, people's knowledge and information about the world around them is structured in units termed 'schemata' (plural) which are stored in long-term memory. All knowledge is schematic, and this includes not only knowledge in the form of ideas but also knowledge about the structure of everyday experiences, language itself, and norms for how to use language. A schema is a memory unit composed of a set of expectations about how information should be represented when the environment evokes thought and behavior. Schemata may have other schemata embedded within them, and schemata may interact with each other — e.g. knowledge about how to tell a particular story might depend on knowledge about the concepts, events, and ideas that are part of the story, knowledge about how to communicate with the audience at hand, etc. The value of schema theory lies in its openness to social and cultural information as knowledge stored in memory, although at this point only a Brought to you by | University of Michigan Authenticated Download Date | 6/17/15 6:14 PM 110 R.RDuran few cognitive theorists have attempted to describe discourse processes in interaction from a schema theory perspective. Freedle and Durän (1979) suggest that knowledge about how to communicate in situations can be analyzed in terms of 'scripts' — or mental schemata that people have about how to behave in situations. Scripts are schema which describe expectations about how everyday cultural events begin, unfold, and terminate. Examples of scripts include knowledge about events such as eating at a McDonald's Restaurant; doing household shopping at a supermarket; answering teacher's questions at lesson time; reading a storybook to an audience of children, etc. Scripts have a format. Thus, every script has a version — e.g. whether dining at a McDonald's versus a French restaurant, a set of entry conditions, a set of expected or possible scenesr constraints on the ordering of some of the scenes, and a set of actors ana props making up a setting in which a script unfolds. Freedle and Durän (1979) note that there is a direct correspondence between the elements of scripts and most of the parameters of speech events cited by sociolinguists such as Hymes (1972). Indeed, scripts and speech events are about the same kinds of factors affecting human interaction, and it is possible to expand the notion of scripts to include knowledge about norms for communication and language use, in a manner consistent with speech event analysis. Scripts, however, emphasize the importance of ways of behaving in situations. Thus, scripts emphasize the speech event parameters pertaining to 'act sequences' — the ways of interacting and speech acts that make up the sequence of action in a communicative event. The foregoing theoretical perspective is of value when we ask how it is that minority-language children learn about literacy activities as whole units of experience. It may be hypothesized that children learn scripts for how to carry out literacy activities in a way that is sensitive to the social and cultural parameters of a literacy event and setting. For non-English- and bilingualbackground children we might inquire how it is that they adjust their cultural and linguistic reportoire to their performance of literacy scripts. Is there a smooth transition for these children between their ability to perform literacy scripts in their home language versus in English? Might some children's limited performance of literacy scripts in English be related more to an underlying lack of familiarity with a script as a behavioral unit than due to lack of English knowledge? Could it be that contrasts between a home and community culture sometimes do not facilitate children's ability to perform a literacy script in one language versus another? Going beyond these issues we might inquire whether there are special schemata in memory which orient children toward performance of literacy acts based on children's knowledge of their self-identity. As the final portion Brought to you by | University of Michigan Authenticated Download Date | 6/17/15 6:14 PM Discourse skills of bilingual children 111 of this section will indicate, this is indeed a possibility worth examining closely in the context of bilingual-background children. Franklin and Rodriguez Talavera (1982) explored a development of bilingual Hispanic children's literacy skills from a schema theory point of view, drawing on some of the issues suggested above. Franklin and Rodriguez Talavera, drawing on the work of Rosenblatt, mention that every reading event for children is influenced by a context of time and situation in which it occurs and also influenced by the psychological history and current psychological state of the child. They state (1982: 4), It is the role of the reader to organize a framework of meaning. He does so by selectively attending to certain features in the text. The reader's selective attention is guided by his existing schemata for the context of situation. The text, however, because of its particular features and constraints, may modify a reader's active schemata and influence subsequent perceptions during the reading encounter. The reading of a text, therefore, is a transaction between reader and text. It is an on-going, continuous process in which a framework of meaning is gradually built up by the reader as he transacts with a text over time. Franklin and Rodriguez Talavera go on to describe a summer reading program for bilingual Hispanic children which drew on their insights. The program focused on having children read and work with materials stemming from award-winning children's books. Besides reading books or having books read orally to them, children wrote book reviews, composed songs or dances, or told stories about books, to name just some additional activities. The Franklin and Rodriguez Talavera program thus enhanced children's understanding of books in a way that stimulated their development of a range of literacy and language skills centered on meaningful activities. Duran and Guerra (1982) discussed refinements needed in schema theory approaches in order to capture how children incorporate different levels of knowledge in the development of their literacy skills. Based on the research of Schank and Abelson (1977), they suggested that there is a hierarchy of knowledge structures or schemata which encompass literacy development. They indicated that literacy learning cannot be viewed in terms of isolated reading and writing encounters with print. In performing isolated literacy acts, such as reading of stories to an audience at home, bilingual children, and children in general, draw on knowledge of their self-identity and social identity in ways that extend beyond knowledge of the language elements required to read and write. Each time that they enact a literacy event, bilingual children encounter new evidence reaffirming or disconfirming their self-identity as it involves meaningful activities with print. Each time children Brought to you by | University of Michigan Authenticated Download Date | 6/17/15 6:14 PM 112 R. R Daran encounter a new literacy context in which they will have to perform, they draw on a conception of themselves which fits a social situation and the demands of communication in that situation. Over the course of their development, bilingual children's self-identity stabilizes, and the personal importance and significance of reading and writing become more frozen. If bilingual children do not value participation in literacy events — i.e. if their self-identity does not reinforce the importance of participating in such events, then in the future they are less likely to engage in such events and also likely to grow more slowly in their literacy abilities. The foregoing conjecture, however, is not definitive — else how would we expect interventions to help children in developing new skills. The more important issue, indeed, is likely to be how capable children are of changing in directions strengthening literacy development. The Franklin and Rodriguez Talavera program suggests new directions involving interventions. It is much too early to evaluate the utility of schema theory models of discourse skills and literacy development for bilingual education. The advantage of the schema theory views over traditional sociolinguistic and educational linguistic views is that they emphasize the individual and the individual's characteristics without sacrificing the fundamental importance of social and cultural influences on behavior. Schema theory approaches suggest that educational interventions on behalf of language-minority children ultimately have to have an impact on individual children. Given the early state of evolution of schema theory views and their sparse connection to empirical data as of yet, there remains much research to be done before we can appraise the practical values of such views to the education of languageminority children. Educational Testing Service Princeton, New Jersey References Au, K. (1980). On participation structures in reading. Anthropology and Education Quarterly 11 (2), 91-115. Brück, M., and Schultz, T. (1977). An ethnographic analysis of the language use patterns of bilingually schooled children. In Working Papers on Bilingualism 13. Ontario School for Studies in Education. Carrasco, R. L. (1979). Expanded awareness of student performance: a case study in applied ethnographic monitoring in a bilingual classroom. Sociolinguistic Working Paper 60. Cambridge, Mass.: Harvard University. - ,Vera, A., and Cazden, C. B. (1981). Aspects of bilingual students' communicative competence in the classroom: a case study. In Latino Language and Communicative Behavior, R. Duran (ed.). Norwood, N.J.: Ablex. Brought to you by | University of Michigan Authenticated Download Date | 6/17/15 6:14 PM Discourse skills of bilingual children 113 Collins, J., and Michaels, S. (1980). The importance of conversational discourse strategies in the acquisition of literacy. Proceedings of the Sixth Annual Meeting of the Berkeley Linguistics Society 6, 143-156. Cook-Gumperz, J., and Gumperz, J. J. (1981). From oral to written culture: the transition to literacy. In Writings: Volume l, M. Farr Whiteman (ed.). Hillsdale, N.J.: Erlbaum. -, and Corsaro, W. A. (1977). Social-ecological constraints on children's communicative strategies. Sociology 11(3), 1977. Duran, R., and Guerra, E. (1982). Chicano children's literacy learning at home. Paper presented at the American Anthropological Association Annual Meeting, Washington, D.C. Edelsky, C. (i.p.). A critical look at several versions of a popular theory. In Advances in Second Language Literacy, H. Trueba and B. Blair (eds.). San Diego State University. Eisenberg, A. (1982). Language acquisition in cultural perspective: talk in three Mexicano homes. Unpublished Ph.D. dissertation, University of California at Berkeley. Ervin-Tripp, S. (1964). An analysis of the interaction of language, topic and listener. American Anthropologist 66, 86-102. Franklin, E., and Rodriguez Talavera, L. (1982). Schema theory and the development of reading and writing strategies for the bilingual child. Unpublished manuscript, University of Indiana. Freedle, R., and Duran, R. (1979). Sociolinguistic approaches to dialogue with suggested applications to Cognitive Science. In New Directions in Discourse Processing, R. Freedle (ed.). Norwood, N.J.: Ablex. Garcia, E. E., and Carrasco, R. (1981). An Analysis of bilingual mother-child discourse. In Latino Language and Communicative Behavior, R. Duran (ed.). Norwood, N.J.: Ablex. Gumperz, J. J. (1981). Conversational inference and classroom learning. In Ethnography and Language in Educational Settings, J. Green and C. Wallat (eds.). Norwood, N.J.: Ablex. -, and Kaltman, H. (1980). Prosody, linguistic diffusion and conversational inference. Proceedings of the Sixth Annual Meeting of the Berkeley Linguistics Society 6, 44-65. Halliday, M. A. K., and Hasan, R. (1976). Cohesion in English. London: Longman. Heath, S. B. (1982). Ways with Words: Language, Life, and Work in Communities and Classrooms, New York: Cambridge University Press. Hymes, D. (1972). Models of the interaction of language and social life. In Directions in Sociolinguistics, J. Gumperz and D. Hymes (eds.). New York: Holt, Rinehart and Winston. Kernan, K. T. (1977). Semantic and expressive elaboration in children's narratives. In Child Discourse, S. Ervin-Tripp and C. Mitchell-Kernan (eds.). New York: Academic Press. Labov, W. (1972). The transformation of experience in narrative syntax. In Language in the Inner City, by W. Labov. Philadelphia: University of Pennsylvania Press. Laosa, L. (1982). School, occupation, culture, and family: the impact of parental schooling on the parent-child relationship. Journal of Educational Psychology 74(6), 791-827. Lavandera, B. R. (1981). Lo Quebramos, but only in performance. In Latino Language and Communicative Behavior, R. Duran (ed.). Norwood, N.J.: Ablex. Lindholm, K. S., and Padilla, A. M. (1981). Socialization via communication: language interaction patterns used by Hispanic mothers and children in mastery skill com- Brought to you by | University of Michigan Authenticated Download Date | 6/17/15 6:14 PM 114 R.P.Duran munication. In Latino Language and Communicative Behavior, R. Duran (ed.). Norwood, N.J.: Ablex. McDermott, R. P., Gospodinoff, K., and Aron, J. (1978). Criteria for an ethnographically adequate description of concerted activities and their contexts. Semiotica 24, 245275. Mehan, H. (1979). Learning Lessons. Cambridge: Harvard University Press. Moll, L. C. (1978). The importance of the social situation in assessing bilingual communicative performance. The Quarterly Newsletter of the Laboratory of Comparative Human Cognition (University of California at San Diego) 1(1). -, Estrada, E., Diaz, E., and Lopes, L. M. (1980). The organization of bilingual lessons: implications for schooling. The Quarterly Newsletter of the Laboratory of Comparative Human Cognition 2(3). Norman, D. A., Rumelhart, D. E., and the LNR Research Group. (1975). Explorations in Cognition. San Francisco: Freeman. Philips, S. U. (1972). Participant structures and communicative competence: Warm Springs children in community and classroom. In Functions of Languages in the Classroom, C. B. Cazden, U. P. John, and D. Hymes (eds.). New York: Teachers College Press. Saville-Troike, M. (1980). Cross-cultural communication in the classroom. In Proceedings, Thirty-first Georgetown University Roundtable on Linguistics: Current Issues in Bilingual Education. Washington, D.C.: Georgetown University Press. Schank, R., and Abelson, R. (1977). Scripts, Plans, Goals, and Understanding. Hillsdale, N.J.: Erlbaum. Scollon, S., and Scollon, S. (1981). Narrative, Literacy and Face in Interethnic Communication. Norwood, N.J.: Ablex. Slaughter, H., and Bennett, A. (1982). Methods of Analyzing Samples of Elicited Discourse in English and Spanish for Determining Student Language Proficiency. Final Report to the National Institute of Education. Stein, N., and Glenn, C. (1979). An analysis of story comprehension in elementary school children. In New Directions in Discourse Processing, R. Freedle (ed.). Norwood, N.J.: Ablex. Watson, K. (1975). Transferable communicative routines: strategies and group identity in two speech events. Language in Society 4, 53-72. Wong Fillmore, L. (1979). Individual differences in second language acquisition. In Individual Differences in Language Behavior, C. Fillmore, D. Kempler, and H. Wang (eds.). New York: Academic Press. Brought to you by | University of Michigan Authenticated Download Date | 6/17/15 6:14 PM