Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The acquisition of a written language by E.S.L. children during the kindergarten and grade one years Chow, Mayling 1990-10-22

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata


831-UBC_1990_A8 C46.pdf [ 6.66MB ]
JSON: 831-1.0078345.json
JSON-LD: 831-1.0078345-ld.json
RDF/XML (Pretty): 831-1.0078345-rdf.xml
RDF/JSON: 831-1.0078345-rdf.json
Turtle: 831-1.0078345-turtle.txt
N-Triples: 831-1.0078345-rdf-ntriples.txt
Original Record: 831-1.0078345-source.json
Full Text

Full Text

THE ACQUISITION OF A WRITTEN LANGUAGE BY E.S.L. CHILDREN DURING THE KINDERGARTEN AND GRADE ONE YEARS by MAYLING CHOW B.Ed., The University of British Columbia, 1974 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS in THE FACULTY OF GRADUATE STUDIES Department of English Education We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA July 1990 © Mayling Chow, 1990 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive • copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of ^DUO&T/DAI The University of British Columbia Vancouver, Canada DE-6 (2/88) ABSTRACT This research investigates the development of writing in children who are learning English as a second language (ESL). Its underlying hypotheses are that: 1) ESL children will learn to write independently when placed in a social and psychological setting that facilitates language learning; and 2) they will use the same strategies and follow the same general patterns of development as those reported for English-speaking children. Current research on emergent literacy provided the theoretical framework for this study. This investigation followed eleven ESL children from the beginning of Kindergarten to the end of Grade One. The children's writing samples were collected daily and were analyzed and classified within Gentry's (1982) stages of writing development. The data were examined for implied strategies, knowledge and understandings. Observational notes on the children when writing revealed characteristics and behaviours found at each level of writing development. The results point to the similarities between how ESL children and English-speaking children learn to write when challenged to discover the English writing system for themselves. The theoretical perspective of writing as a developmental process was evident throughout the study. Additional findings highlighted the significant role of literature in ESL learning and the importance of a learner-centred approach to literacy instruction. The implications of the research findings for ESL methodology is discussed together with an account of the children's development in writing. ii TABLE OF CONTENTS Abstract 11 Table of Contents "i List of Tables • • v List of Figures vl Acknowledgement lx Chapter One: Introduction 1 Early Literacy Research 2 Issues and Concerns • 5 A Perspective on Literacy Learning - Whole Language 7 Preliminary Investigations 8 Chapter Two: Review of Literature 11 Oral Language DevelopmentWriting Development 14 The Language Learning Environment 20 Shared Book Readings 22 The Role of Literature in ESL Learning 5 Conclusion 26 Chapter Three: Method 7 The Study 30 The PrinciplesParticipants, Programs and Classrooms 31 Instruments 5 ChecklistsMaterials 3ProceduresThe Group Work 7 The Writing Sessions 38 Interactions 41 Analysis 2 Chapter Four: Findings 6 IntroductionLevel One - Precommunicative 47 The Form 4iii The Content 51 The ChildrenLevel Two - Semiphonetic 58 The FormThe Content 66 The Children 71 Level Three - Phonetic 5 The FormThe Content 80 The Children 3 Level Four - Transitional 86 The FormThe Content 94 The Children 8 Summary 101 Chapter Five: Discussion 103 The Kindergarten Year 4 The Grade One Year 8 Level Four - Transitional Spelling 115 Level Four - Transitional Writing 7 The Impact of Literature on ESL Learning 120 Limitations and Implications 124 Conclusion 12References 13Appendix A: The Progress of ESL Writers through Developmental Stages During the Kindergarten and Grade One Years 141 Appendix B: Levels of Writing Development 142 Appendix C: Record of Writing Growth - 1 143 Appendix D: Record of Writing Growth - 2 144 iv LIST OF TABLES Table 1 Demonstrations of Letter-name Strategies 61 Table 2 Letter-name Matches for Consonants 62 Table 3 Representational Consonants 4 Table 4 Pairing of Lax (short) Vowels with Letter-names 76 Table 5 Generalizing Concepts of Word Separation 78 Table 6 Digraph Consonants 79 Table 7 Vowel Markers 8Table 8 Visual Spellings 90 v LIST OF FIGURES Figure 1 Non-alphabetic - scribble writing 47 Figure 2 Mock alphabetic writingFigure 3 Alphabetic writing - letters and numbers 48 Figure 4 English and Chinese writing 4Figure 5 Vladimir (repetition of letters in name) 50 Figure 6 The lower case "i" 5Figure 7 Labelling 2 Figure 8 Retelling of The Great Big Enormous Turnip 54 Figure 9 Picture-writing 55 Figure 10 Nana (peer assistance) 6 Figure 11 Semiphonetic writing 58 Figure 12 Early semiphonetic writing 9 Figure 13 Letter-word correspondence 61 Figure 14 B/P substitutions 3 Figure 15 Word separations and sight words 65 vi Figure 16 Retention of uppercase B,D,P 65 Figure 17 Reduction of written messages 7 Figure 18 Familiar refrains 68 Figure 19 Video gamesFigure 20 Sight words 69 Figure 21 Unconventional use of sight words, #1 70 Figure 22 Unconventional use of sight words, #2 70 Figure 23 Representations of complete sentences 72 Figure 24 Early concept of words 7Figure 25 Familiar songs 4 Figure 26 Letter place-holders 7Figure 27 Words as separate units 7 Figure 28 Spacing 7Figure 29 Omission of nasals 78 Figure 30 Language patterns 81 Figure 31 Story conventions 2 vii Figure 32 Monologue oz Figure 33 Expository writing 0J Names as a constant 84 Figure 34 s Figure 35 "I.M." sentence pattern 85 Figure 36 The Three Little Pigs 8Figure 37 Transitional spelling 50 Figure 38 Phonetic spelling 8^ Figure 39 Recall of sight words 88 Figure 40 Punctuation 91 Figure 41 Nasals 2 Figure 42 Transitional writing 95 Figure 43 Nana's writing, a comparison 96 viii ACKNOWLEDGEMENT The author would like to express genuine appreciation to her advisors, Dr. Wendy Sutton and Prof. Mary Ashworth, for their inspiration and support. Special thanks to Dr. Ken Reeder for his insightful comments and interest, and to Dr. Glen Dixon for his helpful suggestions in the final writing of the thesis. Grateful acknowledgement must be made to Dr. Lee Dobson, Marietta Hurst, Joy Nucich and Shirley Brunke. Their contributions and support have been invaluable. And finally, thank you, Bob and Justin. ix 1 CHAPTER ONE INTRODUCTION In his book, Insult to Intelligence, Frank Smith compares learning with joining a club where the members demonstrate in the normal course of everyday activities what the newcomer is to learn. Newcomers are engaged in club activities from the very beginning. There are no entry fees, no "readiness" requirements. Membership in a spoken language club, a literacy club or any other club adds to the individual's sense of personal identity. "From the beginning, the child is a reader or a writer, a member of the guild, who takes learning for granted and who will learn. These are the children who will attempt to read and write beyond their level of ability, who will challenge themselves constantly in written language activities for the simple reason that they see themselves as competent readers and writers." (Smith, 1986: p.38) What is becoming evident in the research on early literacy is that 1) all children learn constantly, without the need for special incentives or reinforcement, 2) children learn what is done by the people around them, and 3) children learn what makes sense to them (Smith, 1986). Thus, children learn about using written language by observing how the people around them use written language, and by participating in the activities themselves. 2 The present research reflects this view of literacy learning. It regards learning as the natural function of the brain and learning a written language a normal outcome of a literate society (Clay, 1982; Ferreiro & Teberosky, 1982; Teale, 1982; Bissex, 1980; Y. Goodman, 1980). Specifically, this research investigates the progressions made by children for whom English is a second language (ESL) as they discover the English writing system for themselves. It looks at the processes and development reflected in these ESL children's writing, and evaluates these findings against the stages of writing development identified for first-language learners (LI). Its underlying hypothesis is that second-language learners (L2) use the same strategies and follow the same general patterns of development in their acquisition of written language as their LI counterparts. This present research is a natural extension of the direction taken by researchers of early literacy. Early Literacy Research Teale & Sulzby (1986) presented a historical overview of how research in literacy learning has developed and changed over the years. They discussed the fundamental difference between the more traditional reading readiness paradigm and the emergent literacy paradigm. The former grew in the 20's and still exists today in varying forms, and the latter grew in the 60's as a reaction to the traditional approaches and in recognition of the need to reconceptualize "readiness" in accordance with current research. 3 The assumptions underlying the readiness paradigm are: 1) learning to read follows the mastery of a prerequisite set of basic skills, 2) learning to read precedes learning to write, 3) reading must be formally taught by the careful introduction of sequenced sub-skills, logically ordered from an adult's perspective, 4) previous experiences with print are not considered (functional uses of reading are generally ignored), 5) all children go through a scope and sequence of readiness and reading skills and are carefully monitored by formal testing. In comparison, the emergent literacy paradigm acknowledges that: 1) literacy development begins long before children enter school, 2) children develop as writers and readers concurrently and interrelatedly (they are able to read what they write before they are able to read conventional text), 3) the functions and forms of literacy are apparent in all literacy events, 4) the foundations of literacy development begin at birth, the critical cognitive years being 0-6 years, 5) children learn written language by actively participating in and observing literacy events within an interactive, social context, and 6) while there may be generalized stages of development in literacy learning, the rate at which children progress through these stages and the variety of strategies employed within these stages are individual (Teale & Sulzby, 1986). Mason & Allen (1986), in their review of emergent literacy research, also discussed how in the late 60's the trend in research began to move away from focusing on formal instruction in the first grade to looking at literacy learning before young children enter school. They spoke of the social context in which literacy develops and the literacy concepts which are acquired through adult-child interactions. The assumption that most young children are unable to read or understand what it 4 means to read or write until they are formally taught in the first grade was clearly being challenged. The roots of this challenge can be traced to the interdisciplinary influence between cognitive psychology, linguistics and early childhood research. A renewed interest in the infancy years as the critical period for language development and learning in general had established a new discipline of psycholinguistics which spanned the traditional boundaries between psychology and linguistics (Smith, 1986). Children came to be regarded as learning-hypothesis generators and problem-solvers rather than as passive recipients of information (Hall, 1987; Teale & Sulzby, 1986). The language acquisition research studies which identified and described the strategies that children used when learning and using a language established the notion of the child as a "constructor of language" (Hall, 1987). Findings from this body of research were used by researchers who hypothesized that oral language and written language proficiency develop in parallel ways, and that literacy learning as with speech begins within the first few years of a child's life. Subsequent research focused on children in the process of becoming literate and thus the term "emergent" was used. The word "emergent" indicates development - the process of becoming - and the word "literacy" indicates the interrelatedness of writing and reading in young children's development (Mason & Allen, 1986). Researchers of emergent literacy view literacy learning as conceptual and developmental. Growth in writing and reading is seen as coming from within the child, and as the result of the stimulation and interactions which occur within a literate environment. Definitions of reading and writing have been broadened to include pretend- or picture-reading and scribble-writing which are 5 seen to parallel the beginnings of speech (Doake, 1979; Gibson & Levin, 1978; Clay, 1975). Writing appears to be easier for children to learn than reading (Meek, 1982). It is also seen to precede reading (C. Chomsky, 1971). Issues and Concerns An outgrowth of this new perspective on literacy development is the issue of why some children are having difficulties in learning to write and read. If literacy learning is a normal consequence of a literate society, why are so many children failing in school? Dyson (1982) states that schools have failed to "connect" children's natural acquisition of literacy with formal literacy instruction. The cognitive and social processes involved in learning to write, as highlighted by recent research, are not operating in many school settings. Harste & Burke (1980) indicate that many school activities given to students force students to operate within the teacher's assumptive bounds and do not allow students to demonstrate what they know and are capable of doing as language users. Children come to school with different sets of knowledge, social values and varying proficiency in language. Too often, the children most needing to see the sensibleness of literacy are given the most senseless instruction (Dyson, 1984). As Sims (1982) points out, when we take away the meaningful context for writing and reading, "the results have been inane, programmed, drill-the-skill packages - divorced from the language and experience the children bring to school and force fed to them bit by boring bit" (p.227). Smith (1986) refers to the meaningless tasks and demeaning tests which these packages impose on students as "insults" to the intelligence of students and teachers alike. 6 Another concern raised in many school districts is that growing numbers of ESL students pose new questions regarding methodology and curriculum development. A recent 1988 survey conducted by the Vancouver School Board shows a total of 23,732 Vancouver pupils designated as ESL. This represents 46.9% of the District's total enrolment and is the highest percentage in the history of the survey. The survey reported that only one-half of the elementary ESL pupils could match their age peers in three of the four language-related activities (understanding, speaking and reading); and only 42% of the students could match their age peers in writing (Reid, 1988). Historically, ESL curriculum has been determined by whichever textbook was available at the time (Ashworth, 1988). In the early years, language texts for adults were adapted for young children; in recent years, basal readers from traditional, mainstream classrooms determined the language arts programs (Gunderson, 1985). Neither addressed the individual needs, language competencies nor the different interests of ESL children. Both assumed that the development of reading precedes the development of writing. Current teaching practices in ESL reflect the instructional assumptions underlying traditional mainstream programs (Benna, 1985; Zamel, 1985). These practices stress the acquisition of oral language first before the formal teaching of reading and writing begins (Piper, 1987; Chow, Early, Lowe & Yeung, 1978). Implicit in this is the assumption that oral language and written language are two separate processes which require different methodology and learning environments. However, it has become increasingly apparent that current teaching methodology is failing to reach a large portion of L2 learners. ESL children are falling further behind in their 7 acquisition of written language. It seems that they are becoming passive, dependent learners in an environment which is too directive and which fails to acknowledge the skills and information which individual children bring to the learning task (Zamel, 1985; Cummins, 1984). A Perspective on Literacy Learning - Whole Language The present study reflects a perspective on teaching which is more compatible with those mainstream classrooms reflecting the findings of emergent literacy research. K. Goodman (1986) talked about the grassroots resistance to the sub-skills approach in language education. He spoke encouragingly of the growing movement towards and the acceptance of a holistic approach to literacy learning and of the Whole Language philosophy now being put into practice in school curriculum. This holistic or Whole Language philosophy-based approach is a child-centred approach to language instruction. It recognizes that language and literacy development are best facilitated when reading, writing and talk are interrelated. The key theoretical premise for Whole language is that children acquire a written language in much the same way they acquire an oral language, and that a rich infusion of the best in children's literature plays a large part in literacy learning. The major assumption underlying Whole language is that the model of acquisition through real use is the best model for thinking about and helping with the learning of reading and writing, and learning in general (Altwerger, Edelsky & Flores, 1987). Reading and writing are learned in a meaningful context. Whole language rejects the underlying assumption of traditional literacy approaches which view language as 8 something which can be taught in "bits and pieces" to be practised in isolation from a whole text (K. Goodman, 1986). Rich (1985) explained how Whole language is not a formula for teaching but a shift in the way teachers should think about their art and practise it. Children in Whole language classrooms read and write daily from the very first day of school. Their teachers accept that definitions of reading and writing have been broadened to include play reading and play writing. Talk is important and the children have many opportunities to discuss their writing and reading with each other. Children are immersed in literacy and actively exploring the power of print. The functions of literacy are as integral a part of learning about written language as are the forms of literacy. Preliminary Investigations In 1981, Donald Graves called for the kind of research which can shed more light on the strategies children use as they learn to write within the context of classroom. To understand the role of the teacher and the processes of learning within this context, more time must be spent by researchers in the only true laboratory, the school. Graves (1981a) therefore challenged teachers to become researchers in their classrooms. This is particularly important in light of current researchers' views of children as thinking individuals who both affect and are affected by their environment (Calkins, 1986; Clay, 1984; Harste, Burke & Woodward, 1982; Hickman, 1982). Dobson (1983), in response to this challenge, documented the progress of early writers as they discover written language for themselves. She reasoned that when 9 children are expected to write daily on topics of their own choice, they would respond by using their increasing knowledge about letters, words, spacing and other conventions of written language to put their meanings into print. Their strategies and solutions could be expected to reveal the path of their development as well as giving the teacher valuable diagnostic information. Dobson's investigation showed that children can discover the writing system for themselves and, in so doing, produce writing that is fluent and comprehensible. Her findings reinforced the theory that language learning, whether oral or written, is a hypothesis-testing situation in which the learner forms hypotheses, tests them out, and confirms or disconfirms them. Language acquisition is similar to other learning tasks in that it requires the learner to be actively participating in a way that is meaningful to him (Smith, 1986). Past and current research on spelling development in young children support Dobson's findings (Harste, Woodward & Burke, 1984; Gentry, 1982; Temple, Nathan & Burris, 1982; Henderson & Beers, 1980; Forester, 1980; Read, 1975). In 1983, this researcher applied for and received a grant to investigate the results of a two-year project involving ESL children from the beginning of Kindergarten to the end of Grade One (Chow, 1986). The findings from this study provided the database for the present research. The author reasoned that, like Dobson's subjects, ESL children can also make sense of the writing system for themselves when optimal conditions for language learning are in effect. Current research on emergent literacy provided the theoretical framework for this study. Given what is known about how children learn and the conditions under 10 which they learn best, it was hypothesized that L2 children will learn to write independently within an interactive and supportive environment rich in language. The two questions posed now are: 1) What do ESL children actually do when they are left to discover the English writing system for themselves?; and 2) Is the writing development of L2 children similar to the writing development of LI children when the same optimal conditions for language learning are in place? In an attempt to address these questions, this research study will took at how ESL children acquire a written language. It will: a) investigate the processes and developments in ESL children's writing during the Kindergarten and Grade One years; and b) evaluate the resulting data of these L2 learners against the stages reported for LI learners. In conclusion, this research study represents an investigation into how ESL children learn a written language. The evidence presented in this study will have some practical implications for the teaching of literacy and for ESL methodology in general. This chapter discusses the rationale, the questions and approach behind the investigation. Chapter Two surveys the literature leading up to and supporting the present study. Chapter Three considers the appropriateness of the approach and provides a detailed outline of the method. Chapters Four and Five examine the children's writing over the two years of the study and offer some tentative answers to the two questions under consideration for this study. 11 CHAPTER TWO REVIEW OF LITERATURE As early as 1908, E. B. Huey urged educators to connect the home experience in the early years with literacy learning in the schools. In 1932, Legrun noted the writing development in children from the ages of three to six which occurred without direct instruction. Y. Goodman & Altwerger (1981) reported on Legrun's (1932) five stages of early writing development which ranged from unorganized scribbles to a differentiation of forms with linear arrangements and occasional interpositions of true letters and figures. Hildreth (1936) identified the developmental sequence evident in the ways which pre-schoolers spontaneously learned to write their proper names. The findings of these researchers had no discernible effect on the language arts curriculum in their eras. The concept of writing as a developmental process in preschool age children was relatively new then and remains comparatively new today (Teale & Sulzby, 1986; Mason & Allen, 1986). It was not until the late 60's that the developmental aspects of literacy learning were seriously considered and became more widely researched. ORAL LANGUAGE DEVELOPMENT The developmental perspective on literacy learning was founded on the previous research in the infancy years and in language acquisition. Piaget's (1977) observations indicated that from the moment an infant is born, the learning process begins. Bloom 12 (1964) concluded from his analysis of many longitudinal studies of development that the majority of human intellectual development takes place before the age of five, with 50% of the intelligence measured at age 17 being developed by age four. The infancy research demonstrated that preschoolers knew more than they had generally been given credit for and that during the early years, children could be learning many skills (Teale & Sulzby, 1986). Donaldson (1978) stressed that adults' interpretation of how children learn must include an understanding of a child's changing view of the world and what makes "human sense" to a child. Children constandy revise or modify their knowledge or perception of the world around them as new information is being assimilated (Piaget & Inhelder, 1969). This view of the child as an active "hypothesis-tester" was further reinforced by research in the area of oral language acquisition (Brown, 1973; Cazden, 1972). In 1957, Noam Chomsky argued against the then conventional view of language as a "habit-learned" behaviour like all other human behaviours. Chomsky (1968) believed that human beings have an inborn ability to learn language, and that children learn language by inducing the rules of language through the process of forming and testing hypotheses about language. Chomsky and his collaborators laid the groundwork for the later establishment of the new discipline of psycholinguistics which focused on infancy as the critical period in the development of both language and cognition (Smith, 1986; Genishi & Dyson, 1984). Bruner (1975) studied the connection between infants' actions and the beginnings of language. He viewed the young child as inherently sociable and it is this sociableness which provides the motivation for attempts to communicate long 13 before children say their first words. Slobin (1979) talked about the cognitive prerequisites to learning a language. He believed that 1) the child must first be able to perceive, analyze and store verbal messages; 2) the child must see his world as constant even though events, people and things may change; and 3) he must interact socially with the people in his world to achieve various personal and interpersonal goals before he can begin to use speech for communication. Halliday (1975) also studied the social and interactive functions of language. Like Bruner and Slobin, he stressed the precursors of language. Children must first understand that it is by learning a language that they become functioning members of a group, whether it is their families, communities, or cultures. In Halliday's terms, children first learn to "mean", not to speak as they are socialized. The functions of language and the range of possible meanings exist well before the first words are uttered. Children learn an oral language through interactions with parents, siblings and other significant care-givers. Young children learn to communicate because they want to master or make sense of their environment (Piaget, 1974), and because they need to "belong" (Smith, 1986). They are able to learn a language with relative ease because their parents and other care-givers are sensitive to their specific levels of communicative abilities, their immediate interests, and to their intended meanings (Wells, 1985). Parents adjust their speech when interacting with young children. Bruner (1983) talked about "scaffolding" as the linguistic behaviour between parent and child in which the adult supports the child's attempts to communicate by expanding on what he says and by negotiating the intended meaning. In addition to scaffolding behaviours, parents and other care-givers demonstrate what talk can do and what talk 14 is used for (Cazden, 1983). The force of these demonstrations increases as children experience their parents coming into contact with a wide range of siblings, friends, shopkeepers, and so forth. Regardless of their style or rate of language acquisition, children are credited with two kinds of competence. They develop linguistic competence, the unconscious knowledge of phonological, syntactic, and semantic rules; and communicative competence, the pragmatic knowledge that language is used differendy in different situations (Hymes, 1972). Children gradually learn to use varied styles of speaking -formal, informal, colloquial - and they learn to adjust what they say to suit their listeners and the occasion. In other words, they learn rules for differentiating among social situations. The bilingual child who has learned two languages in the preschool years acquires additional rules as to when he speaks one language or the other (Lindholm, 1980). Like monolingual speakers, bilingual children make inferences about social and linguistic appropriateness, based on their on-going interactions with members of different social settings. For all children, knowledge of the forms of language develops simultaneously with knowledge of uses and functions. WRITING DEVELOPMENT The major concepts applied to oral language development appear to apply to written language development as well. Children actively develop their own models of how written language works by purposefully interacting with people and objects in their environment (Genishi & Dyson, 1984). Literacy, like oral language, exists so that meanings can be created and so that communication can take place between people. As 15 with oral communications, literacy events at home are almost consistently embedded within the routine social interactions of adults and children (Anderson & Stokes, 1984; Schieffelin & Cochran-Smith, 1984). Young children from a literate society are constantly exposed to the symbols and products of their print-oriented environment. It is not surprising that long before children enter school, they have begun to discover how print is organized and how it used by members of their society. These children try to make sense of the literate forms in their environment in much the same way that they try to make sense of the rest of their environment In responding to, interacting with, and organizing the written language in their daily world, they begin to understand: 1) the significance of written language; 2) the oral labels used when referring to written language; 3) the purposes written language serve for different people; and 4) the variety of forms used to construct meanings communicated by written language (Y. Goodman, 1980). It is in these interactions between the learner and his world that the origins of literacy can be found. Studies on young children's development of writing have documented the similarities between oral language development and the later acquisition of written language, and have focused attention on the cognitive and social processes involved in learning to write (Graves, 1983; Ferreiro & Teberosky, 1982; Y. Goodman, 1982; Teale, 1982; Bissex, 1980). More importantly, such studies have served to point out how schools have not "connected" children's natural acquisition of literacy with formal literacy instruction (Dyson, 1982; Harste & Burke, 1980). 16 Many products of early writing appear to be more like cursive writing than print (Ferreiro & Teberosky, 1982; Clay, 1975). Gibson and Levin (1978) pointed out the possible parallels between babbling as the beginnings of speech development and scribbling as the beginnings of writing development. Somewhere between three to five years, most children become aware that people make marks on paper purposefully (Clay, 1975). Scribble writing and mock alphabet letters are signs that a child is beginning to know what writing does. Children shift to new hypotheses as their current ones conflict with the written structures they encounter. This experimentation and hypothesis-testing of newly developed concepts are important in learning to express meaning through written words (Deford, 1980). Teachers must provide ample opportunities for early writers to show what they already know about written language's processes and functions (Jagger & Smith-Burke, 1985). Vygotsky (1978) maintained that however complex the process of development of written language may be, there is a historical line in a child's progression from scribbling, to drawing objects, to representing speech. Children who represent their own speech (the basis for understanding written language) by a system of signs are using second-order symbols. Transition from first-order symbolism (scribbling) to second-order symbolism (letters) should be facilitated by early writing instruction. Vygotsky (1978) stated that the secret to teaching a written language is in preparing and organizing for this transition, and that "as soon as the child has mastered the principle of written language ... (he needs) only to perfect this method" (p. 116). Language development, whether written or oral, is an active, not a passive process. From birth, children are engaged in a developmental learning task as they test 17 hypotheses about the structure and use of language (Brown, 1973). Menyuk (1971) suggested that what may appear to be errors (that is, differences between the adult's speech and child's speech) represent the outcomes of stages in the child's development of knowledge of the language. Clay (1975) referred to children's writing errors as "signs of progress" reflecting the use of rules or ambitious attempts to communicate meanings. Thus, unconventional patterns in writing and misspellings can be viewed as representing stages in a child's development of written language as he strives to approximate adult writing (Clay, 1982; Gentry, 1982; Graves, 1982; Bissex, 1981; Henderson & Beers, 1980; Chomsky, 1979). Read (1971) explored children's generalizations of phonological rules by examining their spontaneous writing. His research strongly suggested that young children categorize speech sounds and invent their own spellings. He further discovered that these young children can independently come to understand how written language works, and that the quality of adults responses to spelling attempts was of paramount importance in encouraging invented spelling. When adults do not interfere with children's attempts to write by expecting them to correctly spell, punctuate, and capitalize letters from the outset, the children will arrive at roughly the same system of representation and follow the same general route to closer approximations to conventional English (Harste, Woodward & Burke, 1984; Dobson, 1983; Temple, Nathan & Bums, 1982; Henderson & Beers, 1980; Forester, 1980; Read, 1975). Early (1975) replicated Read's (1971) research using young subjects who were learning English as a second language. She discovered that L2 children (children learning a second language), like LI children, can organize their perceptions of the 18 phonetic features of English in a way which is consistent and systematic, and reflect this in their spelling. Her study supported the notion that there is a high degree of similarity between LI and L2 children in their oral and written language development. Furthermore, it suggested that ESL children's spelling errors were more developmental in nature than a result of first-language interference on second-language learning (Early, 1975). Henderson & Beers (1980), in applying a language acquisition model to children's writing, found that children advance in their knowledge of words through discernible conceptual stages, and that these stages are stable across dialects and across different languages. They contended that it is the teacher's understanding of children's concepts of written language that leads to effective instruction in literacy. Gentry (1982) integrated his own observations with the research of Read (1971) and Henderson & Beers (1980), and came up with a particularly useful model which delineates five major stages in spelling development of young children. He applied this developmental spelling classification system to the Bissex case study (Bissex, 1980) to provide more fully defined stages as guidelines for teachers to assess and to foster writing development in children. The first stage (Precommunicative) represents children's beginning attempts to write with letters, numbers, or approximations of both. Children demonstrate no knowledge of letter-sound correspondence at this level. The second stage (Semiphonetic) shows children beginning to form concepts of the alphabetic principle as they match letters to sounds. The third stage (Phonetic) allows for fuller representations of words as children's writing begins to include letter matches for vowel 19 sounds. The fourth stage (Transitional) indicates a growing awareness of common vowel patterns and inflectional endings in English orthography. In the fifth stage (Correct), children recognize and recall correct lexical representation. The fact that these stages can be delineated suggests "that learning to spell is not simply a matter of memorizing words but in large measure a consequence of developing cognitive strategies for dealing with English orthography" (Read & Hodges, 1972: p. 1762). Dobson (1983) observed and analyzed the written work of Grade One children over an eight-month period. She reported her findings in terms of Gentry's (1982) developmental stages. Dobson's (1983) and other similar investigations in first-grade classrooms (Calkins, 1983; Giacobbe, 1981; Milz, 1980) support a problem-solving, hypothesis-testing view of language learning and a conceptualization of writing as a developmental process. Written and oral language development follow normal learning processes (Graves, 1983; Teale, 1982; Bissex, 1981; Y. Goodman, 1980). Children learn to talk by interacting with an environment that provides rich information about language. They learn to speak by speaking. Children also grow as writers by interacting with an environment that is rich in literacy. They learn to write by writing. Experiences with print which induce natural literacy fall within every child's social context (Teale & Sulzby, 1986; Teale, 1982). Teachers who wish their students to grow as writers must regard all pieces of writing as growing things to be nurtured rather than as objects to be repaired or fixed (Bissex, 1981). Understanding what children do allows teachers to make "connections" between teaching and learning to write (Dyson, 1982). 20 The Language Learning Environment Children learn their oral language in a social environment which is supportive and purposeful. In this environment, children experiment freely in their use of language, and receive immediate feedback for their communicative attempts (Lindfors, 1985; Wells, 1981; Halliday, 1975; Cazden, 1972). Studies of young readers and writers suggest that growth in written language is fostered by similar environmental conditions (Wells, 1986; Teale, 1984; Doake, 1979; Clark, 1976). They emphasize the significant influence of adult-child interactions and the role of the literate adult in the development of literacy (Schieffelin & Cochran-Smith, 1984; Holdaway, 1979; Ninio & Bruner, 1978). There is an increasing number of research studies which are looking at the conditions under which children learn about literacy. It appears that these conditions are very similar to those for learning about oral language. Cambourne (1984) listed seven conditions under which children learn both oral and written language. The first condition is total immersion in a language environment ~ speech or print. The second condition is demonstration of language use. Children become aware of how language is used and what it can do for people by observing the people around them (Smith, 1981). The third is expectation. Expectations are very subtle forms of communications to which learners respond. Adults "give off expectations that babies will learn to walk and talk, or that young children will read and write. Responsibility is the fourth condition. Children must take responsibility for their own learning. They must set their own pace, discover their own route to learning and develop their own strategies. The fifth is approximation. Children learn language through a series of approximations 21 until they reach conventional forms. The sixth and seventh conditions are employment and feedback, respectively. Children must be given many opportunities to use language meaningfully and with purpose. Positive feedback which acknowledges and confirms their growing understanding of language gives them the confidence to continue as language learners. Hurst (1982) surveyed the literature on language acquisition and literacy learning to determine the factors in the environment which seemed to foster their growth. She summarized her findings into nine principles which will be discussed further in Chapter Three. Hurst's principles are similar to Cambourne's (1984) conditions for language learning. Researchers and educators have described classroom environments and/or programs based upon similar principles (Dobson, 1987; Chow, 1986; Calkins, 1983; Deford & Harste, 1983; Hurst et al, 1983; Clay, 1982; Haley-James, 1982; Bissex, 1981; Giacobbe, 1981; Birbaum, 1980; Milz, 1980). Teachers in such environments give literacy a high profile and status (Hall, 1987). They provide valid demonstrations of literacy and many opportunities for children to engage in purposeful literacy acts. They expect learners to actively engage in the use of language, taking what they can from the adult model and establishing control of their own learning (Harste, Woodward & Burke, 1984). Jagger & Smith-Burke (1985) suggest that the teacher's task may not be so much to teach language as to create an environment which enables language learning to occur. 22 SHARED BOOK READINGS In oral language acquisition, the role of the parent or care-giver is extremely important, not as an instructor but as a facilitator, through discussion, play and demonstration (Bruner, 1983; Wells, 1981; Halliday, 1973; Cazden, 1972). Researchers in the field of emergent literacy have found this same facilitative and collaborative behaviour to be most evident in book-reading episodes (Hall, 1987; Wells, 1985; Snow, 1983). Snow (1983) discussed three characteristics of adult-child language behaviour surrounding literacy events - semantic contingency, scaffolding and accountability procedures. When using "semantic contingency," adults continue topics previously introduced by children. An example of this would be the type of conversations adults have with children about the pictures or text in books. In "scaffolding," the adult supports the child in a task by providing needed information or by constraining the task to allow the child greater success. In "accountability procedures," the adult encourages the completion of the task so that the child may demonstrate the knowledge he is known to possess. These three characteristics are seen as typical of both oral language interactions and of interactions which surround literacy events (Wells, 1985; Snow, 1983; Heath, 1982). Book-reading is a frequent and powerful source of learning about language (Hall, 1987). Snow & Ninio (1986) reported that in addition to the obvious effects on children's developing communicative skills, on vocabulary and on linguistic forms, book-reading provides the opportunity for children to learn a great deal about the skills subsumed under literacy. These skills, amongst many, include letter recognition, book 23 handling behaviours, learning story grammars, and understanding that print represents spoken words (Snow, 1983). However, as Teale (1986) and Hall (1987) were quick to point out, book-reading in itself cannot explain why children whose parents read to them are generally more successful in school. The establishing of routines (or formats) which become highly predictable events in children's lives and in the context of books also contributes to literacy acquisition (Snow, 1983; Ninio & Bruner, 1978). Children who expect reading and writing to be part of everyday life come to view themselves as readers and writers. They expect stories to be meaningful and acquire an implicit knowledge of book procedures and story conventions (Applebee, 1978). The powerful effects of parent-child interactions cannot be overlooked. Snow, Perlmann & Nathan (1984) noted that maternal speech is much more focused and complex when topics are determined by books and in highly routine activities such as book-reading. Re-reading favourite books provides a child with exposure to more complex, more elaborate and more decontexualized language than almost any other kind of interaction (Snow & Ninio, 1986). Schieffelin & Cochran-Smith (1984) reported that books are rarely simply read to children. Instead, the reader and listeners work co-operatively to jointly make sense of the text. Through verbal interactions between the child listeners and adult readers, the decontextualized language of storybook texts are given a context (Scheiffelin & Cochran-Smith, 1984; Snow, 1983). Heath (1980) discussed how children with book-reading experiences at home arrive at school already socialized into the school-preferred approach to teaching 24 literacy. Schools are able to capitalize on what the children already know about print and its functions and meaning. Wells (1986) argues that in listening to stories, children are not only familiarizing themselves with the language of books but, more importantly, are pushing towards the use of decontextualized language. However, the stories in themselves are not as important as the verbal interactions which surround them (Wells, 1985; Snow, 1983). These interactions help children make sense of the decontextualized language of books and learn some of the essential characteristics of written language. The higher-order thinking and the language strategies involved in interpreting written text equip children with the necessary skills for later success in school (Bruner, 1984; Schieffelin & Cochran-Smith, 1984). Schooling offers children the opportunity to learn to use language in a decontextualized way (Donaldson, 1978). However, as many researchers have pointed out, many home experiences surrounding book-reading and other literacy events already move children towards the use of decontextualized language before schooling begins (Teale, 1986; Heath, 1983; Snow, 1983; Scollen & Scollen, 1982). Snow (1983) concluded that children need both literacy and decontextualized language skills in order to succeed in school. The role of the school is to facilitate children's growth towards the use of decontextualized language. It is this activity which calls for the high levels of cognitive thinking which are necessary for school learning (Wells, 1986). Just as with younger children acquiring an oral language, the nature of the adult interaction seems to be crucial in influencing children's learning and literacy 25 development. Teachers need to create a classroom environment full of interesting things, books, and activities which will foster learning through language use and through reading, writing, listening and talking (Smith-Burke, 1985). Opportunities for children to interact with each other and with the teacher are essential. THE ROLE OF LITERATURE IN ESL LEARNING In the ESL field, a renewed interest in literature has surfaced (Spack, 1985; McKay, 1982). The prominent role of literature had lessened over the years as linguistics became the focal point of many beginning language programs (Widdowson, 1982). However, Povey (1967) had contended that the linguistic difficulty of literature has been overstated, and that readers do not need to experience total comprehension to gain something from a text. Widdowson (1982) suggested that instead of limiting the focus of literary study to either language usage or cultural content, teachers should view literature as discourse - as another way in which language can be used to express meanings and realities. McConochie (1982) emphasized that ESL students need to discover that English can be a beautiful language as well as a practical and utilitarian one. While these observations were made for college-level ESL students, they are nevertheless applicable to younger ESL subjects, too. This is especially true in light of the current views on book-reading and on adult-child interactions as critical elements in language learning and in learning, in general. 26 CONCLUSION The principles once thought to be unique to oral language development have been shown to be major factors in the emergence of literacy (Hall, 1987; Cambourne, 1984; Hurst, 1982). Young children are making sense of their literate environment in very much the same way they make sense of their speech environment. The question now becomes: How do ESL children learn about literacy? More specifically, the first question for this study is: What do ESL children do when they learn to write independently? Teachers need to capitalize on young ESL students' previous learning experiences, and their knowledge and understanding of what language is and can do. Research in second-language acquisition suggests that strategy use is a stable phenomenon which is not tied to specific language features, but can be applied to new language-learning situations (Block, 1986). Higher level strategies developed when learning a first language can be transferred to the learning of a second language (Cummins, 1984). This suggests some parallels between learning a first and second language. The second question for this study is now posed: Is the writing development of L2 children similar to the writing development of LI children when the same optimal conditions for language learning are in place? The present study is designed to explore these two questions more thoroughly. 27 CHAPTER THREE METHOD "Research about writing must be suspect when it ignores context or process" (Graves,1980: p.917) Researchers who are investigating how children learn by looking at what children are actually doing while engaged in a task often choose designs that are descriptive rather than experimental (Mason & Allen, 1986). This kind of research permits the flexibility and the opportunity to observe a number of variables in interaction with each other. Descriptive techniques are also chosen to trace influences and reveal development over time. Context needs to be explained. Research procedures must describe the full context of human behaviour and environment before research findings can be translated into Actual classroom practice (Graves, 1980). Writing is not done in a vacuum. A young writer is part of a social context in which children, teachers, administrators, parents and a community demonstrate their values about writing. These values and practices affect what the young writer does when he writes. When research looks only at the product and its fragmented parts, it seriously distorts the processes, tasks, values, interactions and realities (K. Goodman & Y. Goodman, 1977). Kantor, Kirby & Goetz (1981) argued for a descriptive, qualitative, naturalistic and holistic approach to English education research. They supported the research strategies which work more with the "wholes" than the parts, which describe the processes leading to the product, and which use the language of the classroom teacher 28 rather than the discourse of the laboratory researcher. This approach to research is particularly appropriate for those researchers who are questioning traditional assumptions about the acquisition of writing (Humes, 1983; Harste & Burke, 1980; Murray, 1977). Bogdan & Biklen (1982) emphasized the relationship between learning and change. They pointed out that the qualitative orientation allows researchers to deal with participants during change. It directs them to see behaviour in context. An understanding of a young writer's changing concepts of the writing process is ultimately dependent on data from the child functioning within the writing process itself as well as from extensive analysis of the writing product (Graves, 1981b). Cazden (1982) claimed that there are two contexts of learning in the schools -in the mind and in the classroom - and that there is a need for a better understanding of how one affects the other. The language of teacher-student and student-student interaction provides a context that influences the process of becoming literate (Bloome & Green, 1984). Bruner (1966) advanced the concept of a hypothetical mode of teaching,one in which teachers become partners in the learning process, proposing and examining tentative theories along with their students. They monitor their options until the weight of evidence indicates particular directions(Kantor, Kirby & Goetz, 1981). Boomer (1987) argued that good teaching is research. He referred to the process of a deliberate, personally conducted, solution-oriented investigation into how children learn as "action research." This intention to investigate, to study change, and to subsequently document the investigative process is what produces teacher-researchers (Allen, Combs, Hendricks, Nash & Wilson, 1988; Bissex, 1986, Graves, 1981a, 1981b). 29 For years, researchers in the field of oral language acquisition have studied language in natural, social settings (Wells, 1986; de Villiers & de Villiers, 1982; Tough, 1977; Cazden, 1972; Labov, 1969). More recently, this approach has been used to study writing development (Calkins, 1986; Baghban, 1984; Graves, 1983; Harste, Burke & Woodward, 1982; Bissex, 1980). These investigations took place within the subjects' normal setting of home or classroom and took into account all the various interactions which may affect performance. In many studies, the researchers have acted as participants-observers, interacting with the children whose responses they wish to record (Calkins, 1986; Dobson & Hurst, 1986; Clay, 1984; Cochran-Smith, 1984; Hickman, 1982). It is crucial in these circumstances to analyze the researcher's participation and consider its influence. To gain a better understanding of children's strategies and perceptions of a task, the researcher must draw out the children's own responses to open-ended tasks rather than look for correct answers (Hickman, 1982). Weinstein (1984), in reviewing the studies on literacy and second-language acquisition , concluded that current trends in literacy research have a great deal to offer those researchers working with ESL subjects. She further proposed that a descriptive approach to literacy and language research is what enables the investigator to observe and document changes, both in the participants' communicative skills and in the contexts in which these skills are developing. 30 The Study The present research study followed a group of ESL children from the beginning of Kindergarten to the end of Grade One. It has looked at how young L2 children grew as writers when placed in a social and psychological setting that facilitated language learning. This investigation provides an longitudinal view of ESL children in the process of writing, and examines the context in which these children's written language develop. The hypotheses underlying this research are: 1) Children who are learning English as a second language can, at the same time, discover the English writing system for themselves; and 2) ESL children use the same strategies and follow the same general patterns of development in their acquisition of written language as their English-speaking counterparts. The Principles A basic premise of this investigation was that children will reveal the depth and breadth of their understanding if they are immersed in a social and psychological setting which is conducive to language learning. Consequently, procedures were chosen to adhere to the following list of principles (Hurst et al., 1983; Hurst, 1982). 1. Provide a warm supportive setting rich in interactions and demonstrations of functional oral language and literacy. 2. Emphasize the process of writing rather than the product, allowing generous periods 31 of time to explore and experience the process. 3. Respond to the intended meaning of the children's writing first. Some attention can be paid to form, but only when the children indicate they are ready to use it in their work. 4. Present the writing task as a problem-solving enterprise in which the children learn to write by writing using their initiative and all the resources at their disposal to discover the meaning and to solve problems of form. 5. Expect children to come up with their own topics to ensure that the writing is meaningful and purposeful from their point of view. 6. Encourage the children to use their own illustrations as a source and support for their writing. 7. Accept the children's own representations of written communication as a legitimate indication of their conceptualization of the writing task. 8. Expect a developmental progression in their writing efforts over time, having confidence that errors are natural rather than habit-forming and that the children will self-correct errors in direct relation to their stage of development. 9. Evaluate individually both in terms of the developmental progression and in terms of their oral language. The Method Participants, Programs and Classrooms Twelve children who were identified by their classroom teacher as having little or no English at the beginning of Kindergarten were referred to the English Language Centre (ELC) for extra help in learning English. The languages represented in this 32 group were Cantonese, Vietnamese, Korean, Spanish, Tagalog (Filipino) and Yugoslavian. Of these children, one Cantonese-speaking child transferred out by the end of the Kindergarten year. Two more Cantonese-speaking children joined the study group in the ELC at the beginning of the Grade One year. Both were recent arrivals to Canada; however, one was from China and had no previous exposure to English and the other was from Hong Kong and had received English language lessons there. Both children were reluctant to speak up in any language for the first month or two. The three children who were not with the research group for the full two years are not included in the data analysis but their work will be discussed in the final chapter. Two of these children were of particular interest as they were completely non-functional in English when they entered school. The remaining eleven children made up the target group and their writing provided the database for this thesis. The children in the study were typical of an ELC group. The group was composed of six boys and five girls. There were no available general intelligence scores or first-language assessments on the children. However, the children appeared to be an average group. A few children appeared to be above average in intelligence, most seemed average and a couple of children experienced more learning difficulties in comparison to their peers. The children's ages ranged from 5-years and 6-months to 4-years and 8-months as of the first month of school. The mean age at the entry level of the this research study was 5-years and 1-month. Again, this is fairly typical of any Kindergarten group in September. The children attended an inner-city, elementary school of about 200 students. At the time of this investigation, the ESL population in the school was reported to be 33 approximately 65% of the total population. While no detailed information of parental background was available, it became apparent in casual conversations with the parents that they varied in educational backgrounds in their first language. None of the parents, however, spoke English fluently although some were more functional in English than others. All except one, required assistance in reading and writing English. All the parents were anxious that their children did well in school; however, they supported their children in different ways. The children's book experiences at home ranged from none to a generous supply of mostly picture vocabulary books. A few of the children were read to in both their first and second language. All the parents, however, attempted to teach English vocabulary to their children either by labelling concrete objects or with picture books. Without exception, the parents of the children in this study expressed a strong desire to maintain their first language in the home environment. The community in which the children lived is regarded as a low socio-economic urban area. All the children except for two were living with both parents at home. At the time of the investigation, all the children's parents were employed with a couple of parents holding down two jobs. For the most part, the parents in the community were very supportive of the school. Every attempt was made to explain the writing program to the parents of the children in the study. The writing program on the whole was well-received by the parents. All the children in the study group had the same Kindergarten teacher although they were not all in the same class. The Kindergarten teacher read daily to the children and acted as a scribe for their written messages. The children were in a print-filled environment as they shared the same open area space 34 with the Grade One class. The children were taught the alphabet (upper- and lower-case) and how to print their names. During the Kindergarten year, time was not allotted for independent writing when the children would have been expected to do their own writing in any way they can. Therefore, in its first year, the research group was writing independendy only in the ELC. In Grade One, the children were in two classes again but were taught by different teachers. Both Grade One teachers used basal reading programs where the children were introduced to a sight reading vocabulary in the context of stories. They printed words on individual cards and the children used the words to build sentences. The teachers taught phonics lessons in a direct manner - one teacher spent 20 minutes per day on a lesson and drill while the other spent 10 minutes per day. Both classroom teachers allotted time for independent writing and followed similar procedures to those described for the ELC writing sessions. The investigation took place in the English Language Centre. The children in the study were in regular classroom settings and went out to the ELC for a specified amount of time on a daily basis. The focus of the ELC program was on language and literacy learning. The teaching approach within the ELC embodied the belief that language - oral and written - is an integral part of the normal functioning of everyday life. Language is used to make meaning and to accomplish purposes. Therefore, the ELC program relied heavily on literature, on reading print for appropriate purposes (e.g. signs, messages), and on writing for a variety of purposes. The children were expected to use language - oral and written - in a meaningful way, using whatever language, knowledge and understanding of language they had. 35 Instruments  Checklists Dobson (1983) referred to Gentry's (1982) stages of spelling development in her analysis of Grade One children's writing. Since the intent was to follow Dobson's study as closely as possible, this reseacher also classified and analyzed her writing data within Gentry's classification system (see Appendix B). As a guideline and to provide a focus for observation and analysis, the researcher used a checklist compiled by Dobson & Hurst (1986) to record her findings (see Appendix C & D). Materials 1) Story books, song books, nursery rhymes, and poems were used in the ELC to stimulate oral language development. The criteria for the choice of literary material was: a) their use of children's natural language; b) their use of repetition and patterns to increase predictability of language; and c) their use of recognized story conventions. 2) Wall pocket chart, picture and word/sentence cards. 3) Half-lined and half-plain exercise books, coloured pens and pencils. Procedures By the second week of school, the children who made up the target group to be observed began attending the ELC. The children were in two separate groups; one in the morning and one in the afternoon. These groups were determined by the Kindergarten class in which they were placed. The two groups appeared fairly equal in ability and in personality. The children came regularly at the same time on a daily basis. The 40-minute sessions began with free reading followed by shared book-reading session and then by independent writing. 36 The free reading time was originally established to resolve the logistic problems caused by overlapping groups in the ELC. However, the inherent value of this time quickly became evident. The children were able to browse through books, to talk about the books with others, and later, to "read" a favourite book on their own or with a friend. The shared book activity involved only "the best" of children's picture books and was central to the children's oral language program. As the books were read and reread, active choral participation of the children was encouraged. To assist the development of the children's attention to print and their growing control of directional and matching skills, the teacher-researcher held the books up and pointed to the print as it was being read, being careful not to destroy the cohesion of the story (Holdaway, 1979). Follow-up activities involving pocket wall charts and picture and word cards provided more opportunities for rereading (McCracken & McCracken, 1985). The group book-reading sessions facilitated the children's development of reading-like behaviour and their learning-to-read strategies (Doake, 1985). It must be remembered that the children's English language ability ranged from none to just functional, although this improved vastly over the two years. In the beginning, the teacher-researcher relied heavily on body language, role-playing and highly predictable pattern books to get the intent of her questions across to the children. Originally, many of the questions, by necessity, were focused on the pictures. Eventually, the routine of the book-reading sessions became established, and the pattern and format of the questions became predictable. 37 While the researcher held an active participatory role in the shared book activities, she took on a much more passive role during the independent writing time. Her main intention was to observe how ESL children learn to write when they are left to discover the English writing system for themselves. The procedures for both the group language work and the independent writing sessions will be described. The Group Work The following procedure took up to four 20-minute sessions depending on how appropriate the wall chart activities were to the storybooks. It varied with each new material introduced. 1. Read a book to children, allowing ample time for discussion before, during, and following the reading. However, first readings of a new book generally were carried out with fewer and shorter stops to allow the children to simply enjoy the story and the natural language of the book. Questions to stimulate discussions included: "What do you think this story is about?" (looking at the book cover); "What do you think will happen next?" - "Why...?"; "What's happening in this picture?"; "Why do you think (character) is behaving this way?"; "How do you think (character) is feeling about ...?"; "What would you do?"; "What makes you think that?" (checking illustrations and story convention); "Is there a problem needing to be solved? How...?"; 38 "How do you feel about this?"(e.g. after Goldilock's break-and-entry act); "Is there a lesson to be learned from this?". 2. Invite children to predict words or phrases and to join in on the second, third and fourth readings. 3. Review the story using picture cards to identify key characters or events. 4. Have children participate in re-telling the story. 5. Expect the children to join in, once the patterns or repetitions within the story become obvious (e.g. "Who has been sitting in my chair?"). 6. Match children's verbalizations of story patterns through the use of word cards or by acting as a scribe. 7. Vary wall chart activities by inviting children to locate, sequence or substitute cards. 8. Expect all children to participate orally in all activities while accepting that each child will do so at his own language level. The Writing Sessions During the writing sessions, the children were seated at a round table with the teacher-researcher. Initially, she told the children to draw a picture and to write about it. This procedure quickly became routine. The onus was on the children to choose their own topics and to write in any way they can. Typical remarks reinforcing this expectation were along the lines of "You can draw anything you want - you choose," "Write it the way you think it looks like," "You may use pre tend-writing." When 39 asked for help, the researcher replied "It's your writing, do it your way." She did not make suggestions or offer information, but encouraged any efforts to self-correct and to add information. The researcher observed the children while they wrote. She recorded behaviours such as: how they began and ended; where the drawing and writing were placed on the page; the directionality of the "writing"; any evidence of pre-planning and in what form; child-child or adult-child interactions; any vocalization while writing; overt help-finding strategies such as copying or asking someone for specific information; and any indications of their incorporating teacher-initiated literacy activities into their writing (see Appendix C and D). Upon the completion of their "writing," the children were expected to read their work to the researcher who, in turn, responded to the intended meaning of the children's piece. She commented, asked a question, or paraphrased the reading. She dated and recorded what they read on the back of their books in cursive writing, adding any notes on how they read their piece and if they offered any additional information. She also indicated any occurrence of match and mismatch between their reading and the print. The writing time varied with each session and between children. Prearranged activities were set up to accommodate the children's different completion times. The children were encouraged to read different "authors' work" which included their own writing as well as trade books. The teacher-researcher followed this general procedure: 1. The teacher moved about printing the date (unless a child prefers to print his own), reinforcing those who have started (Oh! You've got an idea already. Good for you!), 40 and observing how each child goes about the task. 2. The teacher expected the children to write on topics of their own choice. She did not suggest topics, spell, or otherwise take responsibility for the writing. She said,"It's your work. Do it in any way you can." or "Use your own spelling. You'll be able to read it." or "Just begin your drawing. It will give you ideas for your writing." 3. The children approached the teacher upon completion and read their writing. The teacher responded to the intended meaning. She might have: - reflected the meaning of the child's message (paraphrase) - contributed something from her own experience which related to the topic - asked a question for clarification or more information. 4. The teacher then might have commented on, or asked about the transcription strategies by saying: "Can you show me where it says ?" "Yes. I can see an s for sun." "Can you tell me about ?" (points to print) 5. The teacher transcribed in cursive form the child's message onto the back page of the notebook along with the date and anecdotal notes about the children's strategies. Following each session, The teacher-researcher made further notes on the children's writing using Gentry's (1982) stages of spelling development as reference, and on the content in terms of any perceived connections between teacher-initiated activities and the writing itself. 41 Interactions Recently, researchers have argued that interactions between observers and their subjects can add relevant information to the research (Allen, Combs, Hendricks, Nash & Wilson, 1988; Calkins, 1983; Graves, 1981a). The interactive nature of language development and the important part played by adults must be of particular interest to teacher-researchers (McKenzie, 1985). In the present study, the reseacher expected that the quality of the children's responses to the writing task would be directly affected by the total context of the situation. However, her main intention was to observe. She maintained an accepting and nondirective style of response. She discovered the more noncommittal comments such as "Uh, uh" or "Oh?" inspired the most language from the children. She responded only to the meaning of the children's communications. However, when the children's responses suggested a significant change in understanding, she probed for further information about the children's thinking. Some useful probes were: "Read it with your finger"; "Show me where it says..."; "Read it to me again"; "How did you know about this" (e.g. pointing to punctuation marks); "How did you know how to do it this way?"; "Tell me about this part". For the duration of the study, the investigator maintained her observer's role. This posed some difficulties especially at times when she felt some intervention needed to be made or when opportunities for on-the-spot teaching were presented. However, this disadvantage reversed itself when the investigator was then able to observe the 42 strategies which the children used to solve or to circumvent their problems when they were given the opportunity to find their own route to learning. Through this process, the teacher-researcher became better equipped to determine when intervention would be effective in that it may hasten the learning process, and when it would be harmful in that it may divert a child away from his natural learning style. The teacher-researcher also was put in the position to discover how literacy instruction could be incorporated in activities outside the writing sessions. This resulted in overt efforts to tailor her language in ways which allowed the children to make connections more easily between what was discussed during shared reading time and what they did during their independent writing time. This investigator agrees with Kantor, Kirby & Goetz (1981) when they stated that only by looking at the "whole" learning context, can the processes, the products and the interactions which ultimately affect both be revealed. This investigation allowed for change and the agents of change to be observed and documented. It permitted the type of research findings which may be most readily put into practice (Graves, 1980). Analysis The researcher examined the data in two ways - collectively and individually. Over the span of two years, close to 3740 writing samples were collected from the eleven children who completed the study. The first step was to look at each sample of the children's writing and analyze it against the stages reported for LI children (Temple, Nathan & Burns, 1984; Dobson, 1983; Gentry, 1982). The second step was to trace each child's progress in writing development and to examine it in detail for 43 common characteristics, strategies and implied knowledge and understandings. The third step was to compile this information to allow for a more comprehensive look at characteristics and behaviours generally found in each level of writing development. Gentry's (1982) stages of spelling development were used as a frame of reference for classifying the writing data collected in this study. This was especially appropriate since this present investigation was conducted under similar conditions as those described by Dobson (1983) and Hurst, Dobson, Chow, Nucich, Stickley & Smith (1983) who also referred to Gentry's stages. Five stages were identified in the writing development of young English-speaking children (Gentry, 1982; Temple, Nathan & Burris, 1982; Henderson & Beers, 1980; Beers & Beers, 1980). Gentry (1982) labelled these stages as Precommunicative, Semiphonetic, Phonetic, Transitional, and Correct. These stages were further elaborated on by Dobson (1983) and Hurst et al., (1983) to include other aspects of writing development besides spelling. Children progress through five levels of spelling, with each representing a different conceptualization of English orthography (Gentry, 1982, 1978). The first level (Precommunicative Stage) includes all writing responses ranging from scribbling to an approximation of real letters to true letters interspersed with numbers written in random order. At this level, children demonstrate no knowledge of letter-sound correspondence. At the second level (Semiphonetic Stage), children begin forming concepts of the alphabetic principle as they match letters to sounds. They provide a partial mapping of phonetic representations as they omit vowels and use one, two or three consonants that demonstrate letter-sound correspondence. At this level, a letter-name 44 strategy is very much in evidence with words being represented by letters on the basis of their letter-name (e.g. R = are). At the third level (Phonetic Stage), children attempt to provide a more complete phonetic mapping of letter-sound correspondence as they understand it. The children's spellings develop very systematically albeit unconventionally in adult terms. Nevertheless, the invented spellings are quite readable. The fourth level (Transitional Stage) is where children begin to move from reliance on phonology or sound for representing words in spelling to greater reliance on visual and morphophonemic representations which incorporate aspects of grammar and meaning. They display a growing knowledge of common vowel patterns and inflectional endings. At the fifth level (Correct Stage), children recognize and recall the correct representations. Dobson (1983) described this stage as showing correct vowel markers along with correct inflectional and derivational patterns. The term "level" rather than the term "stage" was used in an attempt to emphasize the cumulative nature of writing development. Forrester (1980), in describing a sequence of spelling development, noted that learning is not linear but one of gradual synthesis and integration. Gentry (1982) observed that change from one spelling level to the next is more or less gradual, and that features of more than one level may co-exist in a particular sample of writing. While learning to write, children draw increasingly from alternative strategies - phonological, visual and morphological. Development proceeds from simple to more complex, from concrete to more abstract form, toward differentiation and integration (Harste, Burke & Woodward, 1984; 45 Ferreiro & Teberosky, 1982; Gentry, 1982; Henderson & Beers, 1980; Clay, 1975; Read, 1975). The investigator noted her observations and recorded each child's individual growth within the framework of Gentry's stages (see Appendix B). Her intention was to observe young children's growth in writing and their behaviours when writing. Therefore, she began taking notes of other characteristics and behaviours beyond those already described earlier with the purpose of expanding upon Gentry's stages to include other aspects of writing growth as they pertain to ESL children (see Appendix C & D). Through careful analysis of the children's writing samples, her own notes and checklists, the researcher hoped to identify the underlying strategies and understanding implied in the children's behaviour as they moved along the writing continuum. 46 CHAPTER FOUR FINDINGS Introduction: This present study was set up to look at how ESL children learn to write. Its purposes were to: 1) investigate the processes and patterns of development in ESL children when acquiring a written language during the Kindergarten and Grade One years; and 2) evaluate the resultant data of these L2 learners against the stages reported for LI learners. An analysis of all the children's writing samples showed a systematic development of strategies which were refined over time toward the conventional. These strategies will be discussed further as I report my findings for each level of writing development. I found the general path of development to be consistent; however, the rate of development varied greatly amongst the children (see Appendix A). Any discussion must assume an understanding of this variance - that while one child may spend one month at one level of writing development, another may spend up to four or six months - and an acceptance that the levels are not distinct but overlap and flow into one another. At each level of writing development, discussions on the children's writing samples came under three categories: the Form; the Content; and the Children. Within each category, I looked at the examples for what they may imply in terms of underlying knowledge and understanding. The first two categories highlighted what children chose to write about when left to choose their own topics, and how they went about doing it. The third category highlighted the variety of strategies which children employed within each level of writing development. This category also allowed for 47 in-depth looks at each child in the study. It focused on the children as individuals who went about the task of learning to write in ways unique to them, and underscored the importance of a learner-centred approach to language instruction. LEVEL 1 - PRECOMMUNICATIVE  The Form This level was characterized by young children's first attempts to write, whether it was scribble writing, mock alphabetic writing (circles and lines), or alphabetic writing (series of letters). The primary feature of this level was that young writers did not demonstrate a knowledge of letter-sound correspondence. Young writers at this level showed both non-alphabetic and alphabetic forms of writing (see Figures 1,2,3). Figure 1: Non-alphabetic - scribble writing Transcription: Good night, Hen. Figure 2: Non-alphabetic - mock alphabetic writing Transcription: The owl can't sleep. The ghost, the girl is noisy to him. 48 Figure 3: alphabetic writing - letters and numbers V6A3 Transcription: Gingerbreadman, flower, sun, my name and snowman. All eleven subjects were at this level of writing development at the beginning of Kindergarten. The children who did not understand my instructions in English were able to induce what was expected of them by observing the other children and, I suspect, through their own experiences with the materials on hand. Crayons and coloured felts were used for drawing, and pencil (and the occasional felts) were used for writing. Baghban (1984) and Harste, Burke & Woodward (1984) reported that children respond appropriately to the type of instruments they considered right for writing. None of the eleven children refused to draw or write although one child held back until she observed another child scribble-write. Six of the children either scribble-wrote or made letter-like marks. The remaining five children began with real letters. Two children wrote in both their first and second languages (see Figure 4). Figure 4: English and Chinese writing Transcription: table, cup, chair, Chinese 49 At first, some of the children shrugged their shoulders or said "That's writing!" when asked about their written messages even though they were very willing to discuss their pictures. Eventually, all but one child related some verbal equivalence to the written forms. The lone child, Nana, wrote the letters of the alphabet in sequence or randomly interspersed with numbers and named each symbol correcdy until the day she made her first letter-sound matches. She clearly knew that her letters did not spell out words and she had no intention to "pretend-write or -read". She read her messages when she first began to show letter-sound correspondences in January of her Kindergarten year. Within this first level of writing, there appeared to be a developmental progression from the non-alphabetic form to alphabetic form. While there may have been the occasional playing with scribbles, once the alphabetic form was established, there was no real regression. Clay (1975) identified certain principles which children appear to follow in their early writing. Two critical principles are: the sign concept - letters and letter-like shapes carry some message although it may be unknown to the reader; and the message concept - messages the child speaks can be written down (see Figures 1,2,3,4). All the children in the study showed some understanding of these two principles in varying degrees. Within the ESL sample group, knowledge of the alphabet ranged from the production of all 26 letters to much repetition of a few known ones. Typically, the few known letters tended to be the ones in the children's names (see Figure 5). 50 Figure 5: Vladimir (also Vlado) I QV L t>jp)lV{ Transcription: Superman Clay (1975) referred to this repetition of the same basic forms as evidence of the recurring principle. All the young writers showed a strong preference for uppercase letters although the lower case letters appeared very occasionally. The notable exception, however, was their preference for the lower case "i" over the uppercase "I" (see Figure 6). Figure 6: The lower case letter "i" Transcription: Nobody likes me because he eat worms. This may have been due to the lower case "i" being more distinctive than the capital letter "I" which could have been easily confused with the lower case letter "L" and the number 1. Number symbols also occurred as part of the writing (see Figure 3). Initially, all except one child QNana) wrote randomly in every direction. The children wrote left-to-right, right-to-left, generally top-to-bottom but occasionally bottom-to-top. However, by the end of the Kindergarten year, all the children were 51 demonstrating what Clay (1975) referred to as the appropriate directional principle for the English writing system. The Content Temple, Nathan & Burris (1982) observed that most beginning writers were willing to pretend that the marks they wrote stood for something. As stated earlier, 10 of the 11 children were prepared to assign a message to their writing. These messages came in forms of: 1) labels - either by single words or sentences; 2) journals - true stories about themselves or their families; 3) retellings - partial or complete retelling of a storybook or television show; 4) stories - original make-believe stories of all kinds; and 5) songs - including chants, nursery rhymes and poems. The Children Every child showed a preference for a specific category of writing topics even though they experimented with many. The following children were chosen for discussion because their writing samples represented one of the aforementioned categories of topics. Wendy entered school knowing only a few English words. However, she was intent on learning English. During storybook time, she constandy interrupted and pointed at various items to indicate which labels she wanted to learn. She used writing time to practice her new words, drawing only the pictures for which she had English labels. Occasionally, she added a new item and asked me in Cantonese what it was. I always supplied her with the English words. Dyson (1981) found that frequently 52 young children did not write lengthy messages but wrote labels for people and objects. Writing, like drawing, was a means of symbolizing significant people and objects in the child's world. Wendy made no attempts to link her drawings together in any way (see Figure 3). Her main purpose, it appeared, was to practice her words. In her writing sample, Wendy demonstrated another one of Clay's (1975) principles, the inventory principle, which related to how children took stock of their own learning by listing or ordering aspects of their literacy knowledge. In the beginning, Wendy "read" her writing without any reference to her writing unless asked. By the end of November, she was underlining her writing left-to-right with her finger as she read. Vlado was also fond of labelling his drawings (see Figure 7). However, his oral language was more complex than Wendy's and his written messages reflected this. Figure 7: Labelling Transcription: This is the little box of candy, this is the big box of candy, this is the biggest box of candy. Vlado's writing was about three boxes of candy. This subject matter was most likely prompted by an earlier book reading session on Little, Big and Bigger. When reading his writing, Vlado first pointed at the scribbles on the far left, changed his mind and pointed to the smallest set of scribbles. He then read his version of Little, Big and  Bigger while matching his statements with the appropriate size of scribbles. Ferreiro 53 & Teberosky (1982) reported that many children initially hypothesize a concrete and direct relationship between graphic features and their referents. Vlado obviously felt that the largest box of candy was best represented by the largest set of scribbles. When reading, Vlado slowed his finger movements under his writing to insure that it matched his reading. As with Wendy, Vlado showed an understanding that writing represented speech and, therefore, the print should begin and end with the voice. Beginning in September, Rosalinda wrote only her first and last names and refused to read any more into her writing other than what it actually said. By January, she began to include her brother's and sister's names for the first time. She also read their names as written. Up to this point, Rosalinda, unlike many children, did not play with the letters in her name to create new messages. However, shortly after the appearance of her siblings' names, she began to create new letter strings. Her stories from then on focused almost exclusively on her family and the everyday events surrounding it. For Rosalinda, writing her brother's and sister's names allowed her to see how letters can be rearranged to make new meanings (generating principle: Clay, 1975). It also provided her with new letters to add to her limited repertoire. Not all children must learn to spell their siblings' names before they will write; however, all children do eventually begin to copy the print in their environment, thereby increasing their knowledge of print and how it is used (Y. Goodman, 1980). With her new-found knowledge and experiences with print, Rosalinda wrote longer and longer strings of letters. This, in turn, increased the length of her stories as she attempted to match her verbal messages with the length of her writing. 54 Of all the children, Kim appeared to have had the most experience with English storybooks at home. Many of the books which were read by the teacher were familiar to him. Kim's storybook experiences were reflected in his writing (see Figure 8). Figure 8: Retelling of The Great Big Enormous Turnip Transcription: Father find the turnip and tried to pull it and call the mother and she can't pull it and then they pull it. In retelling stories, Kim was practising book language and showing a growing knowledge of story conventions. He introduced his characters, set up a story problem, and then provided a solution which insured a happy ending for all. While reading his story, Kim repeatedly retraced his writing until his story ended. Like Wendy and Vlado, he recognized that there must be a print-voice match. All the children were continuingly refining their understandings about writing, and moving gradually towards the eventual conceptualization that letters can represent sounds in words. Ken created an original story about a dinosaur. His reading of this story was well supported by his drawing (see Figure 9). 55 Figure 9: Note details in picture Transcription: The dinosaur and the sun and worms and snakes come out. He want eat worms. He stomach got a ache. Drawing has verbal implications. Since drawing is expressive, its composing processes complement writing and speaking (Baghban, 1984). Many of the children used their drawing as a rehearsal for what they were to say in their writing later. However, for Ken, pictures carried his stories, not his writing. He demonstrated his uncertain concept of what writing did when he was asked to reread his story, this time with his finger. Ken began reading the print but quickly moved back to his drawing to support what he was saying. Nevertheless, by starting with the print, he did show a budding awareness that writing can carry a message which is complete in itself. Ken also showed a beginning understanding of story conventions when he introduced the characters first. His story reflected his current language level in English and a growing sense of "storyness" as he explained the consequences of eating worms. 56 Andrea arrived at school knowing only a few words of English and was unable to write her name correctly for most of her Kindergarten year. She also showed little interest in learning to write her name. While the other children wrote their names whenever, wherever and however, Andrea only attempted to write her name when requested. By April, Andrea was still using the scribble forms although she occasionally experimented with the alphabetic and mock alphabetic forms. In comparison, the rest of the children were all using the alphabetic forms and two of these children were moving into their second and third levels of writing development (see Appendix A). Nevertheless, when Andrea read her writing, she tracked her writing from left-to-right and she demonstrated an awareness that longer stories required lengthier pieces of writing. Andrea often used writing time as an opportunity to practise the songs and rhymes she was learning in English. The longer the song or poem, the more scribble-writing she did. The children and I were accepting of all Andrea's attempts at writing and responded to what she had to say. Occasionally, another child would either write on her page or provide her with letters when she requested help with spelling (see Figure 10). Figure 10: Note the name, Nana Transcription: Nana play doctor. 57 When asked for the spelling of a word, I responded with comments like "write it your own way" or "write it the way you think it looks like". Although Andrea was the last of her group to move into the second level of writing, she showed remarkable progress in her oral language development. Of the four children who were identified as non-functional in English (that is, very limited number of English words or none), Andrea stood out in her rapid acquisition of oral English. By the time she began Grade One, she was no longer as easily identified as ESL by her oral language as were the other children in the group. While it was not unusual for a Spanish-speaking child like Andrea to learn English more quickly than, for example, her Chinese-speaking counterparts, her progress nevertheless remained outstanding. All eleven children advanced through the first level of writing development in ways that were unique and similar at the same time. The characteristics and behaviours discussed so far were observed in all the children but in varying degrees, in modified forms, and at different points of their development. This can also be said for the general non-ESL population (Dobson, 1983; Gentry, 1982). The one factor that stood out as different was the high percentage of Kindergarten writing samples which were directly attributed to bookreading sessions. A direct link between the childen's writing and storybooks was considered existing when the writing contained some of the language of the books. Approximately 60 per cent of the children's writing samples showed this link. In my years of working with Kindergarten children, I have not observed this to be common amongst Kindergarten writers. Generally, these children created stories which expressed their own personal 58 experiences. A book on dinosaurs may prompt a story on the same subject but it is still connected to the child. It was more typical for children to be further along the writing continuum before they began to incorporate specific elements from a bookreading session initiated by a teacher. As a teacher-researcher, I found this to be particularly interesting because the sizable gap between spoken language and book language seemed to be closing at least at this level. LEVEL 2 - SEMIPHONETIC  The Form Children at this level were beginning to conceptualize the alphabetic principle. They were realizing that letters had sounds that could be used to represent sounds in words. They were able to partially map the letter-sound correspondences in English words (see Figure 11). Figure 11: Semiphonetic Writing Transcription: I saw a frog walking in the sea getting some crabs. Five of the eleven children in the sample group reached the second level of writing during their Kindergarten. By October in Grade One, all but two were at this level of writing development or higher (see Appendix A). 59 Initially, children made only one or two letter-sound matches usually at the beginning or at the end of their writing (see Figure 12). Not only was it easier to focus at the beginning or end of a sentence, but it also made it easier for the children to identify their matches when they read their written messages later to the teacher. Figure 12: Early semiphonetic writing Transcription: I am the King of the forests. Since many children at this level did not separate their words, and were not always able to retrieve the exact wording of their messages, some children found it difficult to locate and identify their matches. Often, the new awareness of letter-sound correspondences was so tentative that this strategy would be dropped for a while. However, once attempted, it was used again and in growing frequency. Some children gained control over this newly discovered strategy very quickly. They were representing every word in their messages in some way within a week. For most children, there was no turning back once they began to use the alphabetic principle in their writing. A few children struggled without it for long periods of time. While there may be many variables affecting this disparity in performance, two stood out in this study. One was the failure to reduce their message. When the children were pretend-writing, their stories became longer and longer as their oral language proficiency grew and as they became exposed to many book experiences. However, once they became aware of the letter-sound correspondences in English words, most children reduced their messages sharply, It was an effective strategy as 60 it allowed the children better control over their writing. These messages were easily read back which in itself was a motivator since it confirmed the children's thinking. On the other hand, the children who did not reduce their messages were struggling to read their own writing. The second variable was an uneven knowledge of the alphabet. Often times, I observed one child, Andrea, "sounding-out" a word and predicting the correct letter match but was unable to retrieve that letter from her memory. She then referred to her alphabet line, sang her alphabet song, missed her letter and picked the one beside it. Regardless of her choice, the fact that she didn't have a clear idea of what the letter looked like hindered her chance to recall it later. Thus, a poor knowledge of the alphabet severely limited the number of letter-sound matches a child could make. Much later, another variable became apparent. Many of the children eventually developed the strategy of "write-read-write"; that is, the children gradually adopted a strategy of writing some words, stopping to think, rereading what they had written and then continuing on. For some children, this strategy did not become evident until they reached the third level of writing development. In any case, this strategy became a very effective one when the children begun increasing the length of their writing. For those children who had not reduced their messages, this strategy compensated by creating opportunities for many readings of the same message. By the time a child was finished with his writing, he had his message almost memorized. This permitted him to read his complete message easily which, in turn, confirmed his hypotheses about English orthography. 61 Gentry (1982) referred to the spelling within the second level of writing development as semiphonetic. Semiphonetic spellers used one or two letters to stand for a whole word but these letters had some connection with the words they represented. As a basis for their choice of symbols, they used information they knew, including numbers and letter names. Often, one letter by virtue of its letter-name may represent a single word with similar or identical pronunciation (see Figure 13). Figure 13: Letter-word correspondence HHEffift" Transcription: Look at you. You are dirty (pronounced "dirdee"). Table 1 lists examples of the letter-name strategy in use. Table 1: Demonstrations of Letter-name Strategies letter/word number/word letter/syllable long vowels C (see) 2 (to) PD (party) se (sea) U (you) 4 (for) SN (singing) FLO (throw) R (are) 8 (ate) PYN (playing) NO (know) Y (why) It (want) HoD (holding) sa (say) M (am) BfO (beautiful) mi (my) N (and) AAPO (apple) Mak (make) Bot (boat) God (gold) 62 Children began to represent the consonant sounds first possibly because they were more easily determined. At this level, short vowel sounds were omitted. Table 2 lists the consonant sounds that have a stable and predictable letter-sound. Table 2: Letter - name Matches for Consonants Letter Sound Examples B Ibl • • bat Bt (bird) BK (because) Bic (bike) B (be) P w- Eig PK (picking) Prtc (protect) PD (party) F HI - fat frm (farm) FT (fighting) FADE (fighting) V hi - vase VADR (Vader) VADO (video) havu (have) T N - toe trto (turtle) tak (taking) TOCK (talking) D Id/ -dog DONT (donut) Dak (duck) DRT (Darth) S Isl - sun SP (sleeping) SNC (snakes) sad (said) Z IzJ - zoo Wz (was) WOZ (was) prazan (present) K Ikl - kite HK (hockey) BOK (bark) Kach (catch) J V Ijl- jump JoP (jump) JD (Justice) jaip (JumP) M Iml - man MAT (mountain) MtR (monster) SM (some) N Inl- nose NNe (nick-nack) GRN (gone) N (and) L N - lady LK (look) Los (lots) FL (falling) R hi - red ran (rain) SR (store) Hit (hurt) Young writers made predictions based on the sounds found in the letter-names and on the articulation points in the mouth when making those sounds. In other words, 63 the sound /b/ was heard when saying the letter name B /bi/. The mouth was shaped the same way and the tongue was in the same position. Therefore, it was not uncommon for children to confuse /b/ and /p/ since the only difference between the two sounds was in the vibration of the vocal cords. Other similar pairs in Table 2 were: /f/ and /v/; /t/ and /d/; and /s/ and /z/. ESL children were especially prone to making these types of substitutions (see Figure 14). Figure 14: B/P substitution i PH n\y SBIA w s<\ p. t> r\m . Transcription: I playing with (pronounced "rif') my Sup.erman. (Superman) was playing soccer in the home. In this writing sample, the word "the" was represented by the letter "D" which was very sensible since the child pronounced "the" as /d/ rather than /3cV. Also, the sound /tj/ in "-ing" was perceived to be similar to the sound /n/ in the letter-name "N". Other consonants, the letters C, G, H, W, Y do not have a letter-name match with the sounds they normally represent. Nevertheless, they regularly represented one sound and they appeared frequently enough to be learned easily by children. At the same time, these letters were used to represent sounds other than their own. Children used the name of the letter which was closest to the sound of the word or word part they wished to represent. They also chose the letter-name or letter-sound that was produced in a similar manner as the sound they wished to represent (see Table 3). 64 Table 3: Representational Consonants Letter Sound Examples C (name) /s/ CD (said) CE (sea) CKE (skiing) hxl CE (tree) C (sound) N DoC (don't) G /dr/ GF (drive) GM (drum) GAM (drum) Ici - chat GA (Chow) H ill - shot fih (fish) Y M YNT (went) YD (with) YK (walking) Q fkj - cat QN (Come in) SQ (soccer) J V 1)1 - George JJBMN (Gingerbreadman) oreithj (orange) /dr/ JrrP (drop) JM (drum) Ici - chat JAE (chasing) s M fis (fix) wax (was) ill - shot SP (ship) siep (ship) se (she) R M RiR (When I went) rif (with) K Ikl - cat BK (because) Igl KWN (going) N hk (hate) V in fv (wolf, run!) V IZI - that VBR (the bear) z Isl BCZ (because) 65 A growing control over word separation and a movement towards the use of lower case letters often seemed to accompany the appearance of sight words in text (see Figure 15). Figure 15: Word separations and sight words Transcription: My DOG got lost and I cry and my friend get it. The upper case forms for letters predominated in the first writing attempts. The change to lower case was gradual and accompanied the change to more proficient writing. Suggestions from teachers to incorporate spacing and appropriate lower case letters did not have the desired outcome. However, these changes systematically appeared with time. Some words continued to be capitalized, usually the first letter of significant words and the first word in a story. However, some letters, particularly the easily confused b, d, p continued to appear in the upper case form at every level of development (see Figure 16). Figure 16: Retention of upper case B,D,P Y Transcription: This is a jet plane. The jet plane is taking off. 66 Young children often retain the letters B, D, P because the capitals are less confusing visually than the lower case letters. In figure 16, the writer kept his capital letters for the above reason but he remained confused because /b/ and /p/ are also so similar auditorially. The Content In the second level of writing, illustrations continued to play an important part in the children's writing as they did in the first level. Pictures still carried the context of the story; however, their function was changing. At the first level, children were rehearsing their writing while they were drawing. Often, the topics changed several times during the course of the drawings. As evidenced by their discussions while drawing, many children did not even decide what they would write until their pictures were completed. After writing, the illustrations served as a reminder of what the story was about. Children at the first level came to realize that their writing could represent their whole story. One child, after he had discussed his picture, was asked what his writing said. He replied "Just what I told you". Another child when asked to read her writing said "You mean you want me to say the whole thing again?" These children gave their drawing and writing equal status and functions. On the other hand, Level 2 writers knew that they couldn't possibly write "the whole thing." Most children very quickly recognized that a reduced message meant 67 better control of the writing. At this level, children's drawings became more elaborate, and the context was shared pictorially and orally. The beginnings of the second level of writing was characterized by slow laborious work. It was also a very noisy period as each child repeatedly "sounded" his way through a word. Most written messages were reduced to single words, phrases or simple statements (see Figure 17). Figure 17: Reduced message Transcription: The hungry caterpjHar In Figure 17, there were two attempts at writing. Obviously unhappy with his first try, David crossed it out and rewrote his message again. This was his first attempt at self-correction. Once completed, he read his writing to me, then proceeded to talk about the food which the hungry caterpillar ate and identified the items in his drawing at the same time. David's story was inspired by an earlier bookreading session about a hungry caterpillar. His writing allowed him another opportunity to discuss the story and to review his food vocabulary. At this level, drawings and writing were not equal as much as they were complimentary. The written section began the story and the illustration completed it with much embellishment orally. Other children chose to write familiar refrains from their favourite story. By restricting themselves to a known language pattern, they were able to focus more on 68 the transcription of that pattern. Also, the problem of retrieving the exact wording of the written piece was greatly reduced (see Figure 18). Figure 18: Familiar refrains Transcription: Little pig, little pig, let me in NO NI Mi No, not my chimney. Nathan's wording was not exactly the book's but he approximated the book's language as closely as he could. I suspect that the phrase "not my chimney" made more sense to Nathan than "not by the hair of my chinney chin chin" since his version had some relevance to the story. For Nathan, his writing provided him with an opportunity to talk about his favourite story, The Three Little Pigs. The five categories of preferred topics in the children's writing at the first level continued to be popular at the second level. During the Grade One year, I added a sixth category as I observed a proliferation of stories on video games (see Figure 19). Figure 19: Video games Transcription: I p_lay Pacman game and Pac eat the ghosts. The novelty of these games probably made a greater impact than the games themselves. Nevertheless, it was a very challenging topic for both children and teachers. It was a rare child who was able to explain fully how a game was played. However, as I did 69 not know anything about the games, it was a natural question for me to ask. Illustrations became critical for those children who chose to write about Donkey-Kong or Pac-man. It was a great opportunity for children to use language for the purpose of instructing or informing others (Halliday, 1973). What they were eventually able to do orally, they would be able to do in writing later on. During the second level of writing development, some children were building a core of sight words. Many of the sight words were the names of other children or signs found in the environment (e.g. EXIT, STOP). The words were not always reproduced correctly (see Figure 20), but they were good indicators that the children were paying attention to the print around them and that they were perhaps moving towards a more visual strategy. Figure 20: Sight words (Note "in" and "my") Transcription: I have worms in my_ stomach. 70 Beyond that, at this point of writing development, correctly spelled words did not give much information about the children's understanding of how written language worked. It was the errors which provided insights into the children's thinking and underlying concepts of English writing (K. Goodman, 1969). The children, however, were always anxious to display their knowledge of conventional forms. One child incorporated his newly-acquired sight word into his story in a most ingenious way (see Figure 21). Previously to this, I had not noticed any sight words in his stories other than his name. Figure 21: Unconventional use of sight words, Example 1 Another child developed a conversation around her sight words. Her piece of writing was about two children discussing the words they know (see Figure 22). Figure 22: Unconventional use of sight words, Example 2 Transcription: A STOP (sign) went to school. A car crash him. Transcription: "I know how to say lots of words". "Say it out". "NO, CAT, DOG, -stop"that's all I know how to say". 71 The Children Three children were chosen to illustrate the variety of strategies employed by children in the second level of writing development. David, a Cantonese-speaking child, was described as "hyper" by his parents and his teachers. He moved and talked constandy. Unfortunately, his listening skills were very poor. His classroom teacher reported that David frequently did his assignments incorrecdy because he did not take the time to listen to instructions nor did he stay in one spot long enough to watch a demonstration. Of the eleven children, David's language was the most confused. One writing sample carried a message which David read as "I should is was Christmas and the Santa Claus is was giving boy and girl for toy". This was quite representative of his speech for that period. His English became more conventional both in his speech and writing over time as his experiences with and knowledge of standard English increased. David began making letter-sound matches at the end of June in Kindergarten but it wasn't until late October in Grade One that he finally settied down with his new strategy. In the beginning, he would show appropriate letter-name matches one day and then made no attempts to match for the next ten days. This was not typical of children who have discovered the letter-sound correspondences in English writing, especially for those who have displayed some competence in matching letters to sounds. However, when David finally decided to take control of his own learning, he wrote in full sentences with every word represented (see Figure 23). 72 Figure 23: Representations of complete sentences Transcription: The space ship was flying a-way. He was crash in the grass. In his sample, David showed only the initial consonant sounds of each word except for the last word which had three consonants represented. At this point, his work was very typical of Level 2 writers. Not as typical, yet not uncommon, was a little boy named Adrian. Like David, he had conceptualized the alphabetic principle in English writing. He, too, began with the initial consonant sounds. Unlike David, however, Adrian was unwilling to let go of his original notion of words as being made up of a set number of characters (see Figure 24). Figure 24: Early concept of words Transcription: I bought a new toy is Cycle-man. Adrian's writing sample showed letter-sound matches and "fillers" - letters that had no real purpose except to pad up words. Ferreiro & Teberosky (1982) talked about two basic hypotheses which children used: 1) that graphic characters were varied and 2) that their number was constant. Children at the early level of understanding, they said, seemed to work from the hypothesis that a certain number of characters was needed to make a word. However, this hypothesis generally gave way to the syllabic principle 73 which allowed one letter to stand for one syllable. At this point, letters may be assigned stable sound values. Those children who were able to utilize the strategies of reducing their messages and restricting their letter choices had a clear advantage over those children who were not. Adrian, however, worked around his disadvantage. Adrian had a few basic words in his sight vocabulary, the words "is, my, the, I, in, go, man". This in itself was unusual since young children generally have found most of these words difficult to read in isolation. Adrian accepted his sight words as they were regardless of his hypothesis of how many letters made up a word. On the other hand, all his invented spellings had three or more characters. In his writing sample, Adrian wrote "I" and "is" conventionally but because he did not have the article "a" as a sight word, he assigned five letters to represent that word. Earlier, I had mentioned another child, Andrea, who had difficulty remembering the letters in the alphabet I reported that she experienced extreme difficulty because she could not recall her exact wording nor could she find her letter-sound matches which left her no other clues to her writing other than the intended meaning. While her rereading retained the meaning, she lost the opportunity to have her hypothesis about letter-sound correspondences confirmed. Adrian, on the other hand, did not have the same problem. His sight words were few but stable and unlike Andrea, he had a good knowledge of the alphabet. The spaces he used to separate his words also allowed him to identify where the letter-sound matches were which in turn acted as word cues. Adrian could virtually ignore the rest of the letters in each word as he read his writing. If a familiar refrain 74 or song was chosen, which in itself was a good strategic decision, Adrian was able to read back his whole passage regardless of length (see Figure 25). Figure 25: Familiar songs M Transcription: Down by the bay ra^n w , THP u-. , -jr VMLOfflT T hp yyp^FUM0RLf:A^6)flg Where the watermelons grow WKf: T<> MY HfltoFl Back to my home UoPFA NHP- 6o I dare not go i fry?-• ' JM Forif 1 do>My PNAJ-N •• - mother wil1 say hfT" \}%~XmJ^ '~ Did you ever £e ~ LOP, ^inj^. see a g°°se M Q^A IV kissing a moose , Down by the bay. Nathan, unlike Adrian, did not have a good working knowledge of the alphabet. He could sing the alphabet song but he had difficulty identifying many letters in isolation. However, Nathan compensated by reducing his message to the point where he could have it well memorized, and by putting in "place-holders". When he couldn't match the initial consonant sound of his word, he would arbitrarily assign a letter to hold its place (see Figure 26). Figure 26: Letter place holders Ivf/V ' Transcription: I catch wolf. Wolf run. Nathan's writing lengthened as he learned more letters and as he acquired some sight words. 75 All the children at this level demonstrated their strategies in their own unique ways and at different rates; yet the overall progression in all the children's work remained the same. LEVEL 3 - PHONETIC  The Form Children were observed moving into the third level of writing development when their spellings incorporated many of the surface sound features of words, particularly vowels. At this level, the children attempted to provide a more complete mapping of all the sounds that they hear. Gentry (1982) referred to these spelling attempts as phonetic spelling. Within the ESL sample group, one child entered Grade One already writing at the third level of development. By June of the first grade, two children remained at Level 3 (see Appendix A). The children were representing long vowel sounds earlier since the letter-name strategy had been operating for some time. Examples of this is shown in Table 1. Long vowel sounds at the end of a word (e.g. mi = my) were typical of Level 2 writing. However, long vowel sounds in the middle of words (e.g. GOD = gold) were more common amongst Level 3 writers. Double-vowel markers (e.g. boat), the silent "e" (e.g. make), and the letter "y" vowel marker (e.g. stay) did not appear until much later because the children were still using primarily an auditory strategy. The appearance of short vowel sounds signified a major step towards conventional English orthography. Once the children became confident in making 76 letter-sound matches that were reasonably distinctive and stable, they were then ready to explore the less obvious sounds. Short vowels were not represented by letter-names nor did they provide any other good clues as to the way they should be spelled. Nevertheless, the children worked out a consistent strategy for spelling short vowels. The children organized their own classification of the vowel sounds on the basis of their perceived place of articulation in the mouth. This strategy revealed a surprising ability on the part of children to hear and make judgements about speech sounds. They used vowels in the appropriate places, but the vowels were not always the expected ones. Table 4 lists typical short vowel substitutions. Table 4: Pairing of Lax (short) Vowels with Letter-names Letter Sound Examples A /£/ as in bed MAS (messy) wat (went) prazan(present) E 111 as in hit PEC (pick) HEM (him) E (in) fen(friend) I laJ as in pot gini (gonna) WIS (wants) BIDM (bottom) O Ivl as in mud JoP(jump) poi (pull) drone (drunk) Word segmentation was generally, although not always, in evidence during the third level of writing. Of the eleven children, nine were separating their words before or around the time they began incorporating short vowel sounds in their writing. The remaining two children showed word separations a few weeks after they began representing short vowel sounds. Many children demonstrated their new understanding 77 of words as separate units by inserting spaces or markers (e.g. dots, slashes) (see Figure 27). Figure 27: Words as separate units :--A/V\T\ -w- m 1V. 2-HA//A/. Transcription: I saw Batman. Batman went to the Batcar. Then Batman went to the Bat cave. As with all aspects of writing, development was gradual. Movement from one phase of learning to another was not sudden. Children who showed exaggerated spacing between words were just as likely to collapse words back together again (see Figure 28). Figure 28: Spacing uTTcY Transcription: ...and the mud puddle is gone. I don't know the mud puddle come back. Some children understood that words should be separated but were still puzzled as to where those separations occurred. Table 5 lists examples of the children's overgeneralizations. 78 Table 5: Generalizing Concepts of Word Separation Influenced by syllabication Influenced by phonology a Way (away) Adt (ate it) too day (today) Itbt (little bit) tow day (today) aletobt (a little bit) a gen 's (against) PoATM (upon a time) ra bo (rainbow) bet fole (beautiful) up Pon (upon) Another way in which vowel sounds were perceived by young children was illustrated in Figure 28. The young writer wrote "DoC" to represent the word "don't". The letter "C" with its sound /k/ was a reasonable substitute for the sound N as both were produced orally in a similar manner. The nasal M, however, was omitted since it was difficult for the young writer at this level to identify the sound /n/ when it was followed by the sound /t/. Since the nasalization was more of an influence on the vowel than on the consonant, he assumed that what he heard in "don't" was a peculiar vowel rather than an extra consonant. Another little boy demonstrated this same thinking (see Figure 29). Figure 29: Omission of nasals Transcription: wqf to .MiygeraM 1 went to my g™^™'* house. 79 He wrote "wat" to represent the word "went". The long vowel /ey/ replaced the short vowel lei. This was logical within his terms of reference because the articulation points for the two vowel sounds are in the same place in the mouth. The nasal was again dropped and assumed to be a mere variation of the vowel /ey/. Often, the children's invented spelling diverged from conventional spelling because they perceive oddities of pronunciation that adults may not. Table 6 lists examples illustrating how the children spell words as they hear them and not as adults see them. Table 6: Digraph Consonants Digraphs tr dr ch sh sc/sk Sounds Ici as in chick v 1)1 as in jump /j/ as in "G" l\l as in jump /s/ as in sun Igf as in gone Examples chran (train) MonChreeyjol (Montreal) jrao (draw) jrvv (driving) Gam (drum) GRAGIN (dragon) jik (chick) tej (teach) siep (ship) se (she) GROQ (scored) sgrer (scared) sgiat (skate) Generally, children at this level used the lower case letters but continued to capitalize the words they considered most important. They were usually names, the first word in a story and other significant nouns. 80 The Content The children's writing in Level 3 was increasing in length and complexity as they gained more control over the spelling forms and as they became more proficient in their oral language. Still, they were not exploring topics as thoroughly in their writing as they were able to in their oral descriptions of their stories. One obvious reason was that talking is easier than writing. Another possible reason was that the children regarded illustrations and accompanying conversations as integral parts of their writing. When asked "what happened at the end?", one child responded readily and satisfactorily. However, when asked if what he said was in his writing, he answered "No, I just told you". Another child, when asked about a possible solution to a problem in a story, pointed to the picture and proceeded to make one up. She added to her story verbally but felt no need to include the same information in her writing. Generally, the children at this level did not write for an audience other than a significant adult. Once the children shared their writing with me or another teacher, that was the end of the story. These children appeared to gain more satisfaction from the writing process itself. Nevertheless, when discussing their stories and reviewing different options for their stories, they are preparing for a later time when they will look for a wider audience whose needs must be considered. They were gradually moving towards a more decontextualized language in both their oral and written language. In Level 3, the children's writing grew in fluency, flexibility and originality. At the same time, there were periods when it appeared that a child either regressed or 81 became extremely repetitive. On closer analysis, however, I noticed that these phases of limited or no progress always preceded a major breakthrough in writing format or in spelling. Level 3 writers were beginning to read conventionally rather than "pretend-read" (Dobson, 1987). All eleven children were able to choose a book appropriate to their level and read it independendy. The processes involved in learning to speak and in learning to read and write are very similar. The development in one area supports and reinforces the learning in the other (DeStephano, 1978). For the children at this level, there was so much more information to integrate and to synthesize than before. Many children developed coping strategies. However, as they advanced in one aspect of their writing, they may appear to have regressed in another. One child elected to have certain days when he kept to a specific sentence structure, thereby freeing himself to concentrate on practising his vocabulary and on his spelling (see Figure 30). Figure 30: Language patterns Transcription: A apple is red. A sky is blue. A egg is white. A plum is purple. A moon is yellow. A water is blue. A rain drops are grey. JL 31 "5SK()V FT~nonrF IA: 3L ~rno6n_~~/5 7TTOQZ~ "Wn -tT Bn I lor "TO "in- 1^ CLEI 82 Another child wrote many variations of the same story. These strategies allowed time for individual growth in specific areas. The fact that the children were encouraged to work out their own learning schedules and patterns of learning helped them to become confident writers. The children continued to write, sometimes spending days over one piece. Their stories reflected their growing experiences with books. Many children began their stories in conventional storybook manner (see Figure 31). Figure 31: Story conventions Transcription: upon a time there \(\X (\(A) d a 2. 3 liked to swim. They began to develop their own styles. One little boy wrote a monologue complete with good and bad characters, and a problem and a solution (see Figure 32). Figure 32: Monologue I .Do/ .The .ghod il Ul) U. BcT.cqn VU" ifV .Jef.WC/f. ..?To Help me. To (yef 1 Vii lo Uo Transcription: One day the ghost want the jet. What will I do? I can't get the ghost. He is big. I know what to do. I will get Batcar to help me to get my jet back. 83 At this level, a seventh category of writing topics was added. The children's many book experiences included non-fiction books. Expository writing began to appear. After a few days of reading and talking about whales, some children chose to write about what they had learned (see Figure 33). Figure 33: Expository writing \^QjfQ$£^v rc.ilara.ql i Transcription: Whales are mammals. F^~V~r(TV V)\U Vrr^^^iry They have blowholes. They or?.?, nc wo'r in 1CT ' breathe up (blow out) the water so they don't die. Whales have babies in their stomach and their fins help to swim. The Children Two children stood out as having "holding patterns" in Level 3 - that is, they held one aspect of their writing constant to buy time to explore others. While the other children experienced short periods of this holding pattern, these two maintained theirs for months. Cisy was a very verbal little girl who loved to talk. To her, writing meant more opportunities to talk. Her earlier writing, however, was frequently described as "rambling" as she constantly strayed from her point At the time, her writing reflected her speech. In contrast, her writing in Level 3 was characterized by a very structured 84 story format and by the repeated use of sight words made up primarily of people's names. When Cisy began demonstrating her tentative understanding of short vowel sounds, she also began writing stories using her classmates' names. Upon checking the data, I found that almost 80% of her work, classified as Level 3 writing, consisted of stories about her friends or her family. By restricting her topics, Cisy was able to practise her sight words and produce fairly conventional looking pieces of writing. At the same time, she freed herself to concentrate on the spellings of a very limited number of unknown words (see Figure 34). Figure 34: Names as a constant Some days, she was able to produce writing composed of only memorized sight words. I considered her correctly spelled words as memorized units because her invented spellings demonstrated only sound strategies. There was no evidence of a visual strategy other than in her sight words. Upon closer analysis of Cisy's writing samples, I noticed that she was working hard to discover the system in English orthography. Until she moved into the fourth level, Cisy struggled with the same words and gradually refined the spelling of those words. The word "beautiful" appeared first as "Butaif, then "bedfl", "Bedfou", Transcription: Tania and Maria Angela and me are phiying in the park. 85 "Btbole", "bet fole", "betdfole". While the final version was not conventional, it was a good phonetic representation. It even demonstrated a beginning awareness of the silent "e" vowel marker. Cisy's tightly restricted content permitted her ample time to develop a good understanding of the English forms. Nana was another one utilizing this very efficient strategy. Nana was the child who wrote the alphabet repeatedly until her first letter-sound match. By late April, in her Kindergarten year, Nana was already in the third level of writing development (see Figure 35). Figure 35: Sentence patterns Transcription: I am out 0 Po^ U-f LCXLR t0 pick a flower. Shortly after, Nana began her "I.M " stories. She went from picking flowers to singing, to walking , to swimming and so on. That one sentence pattern dominated Nana's writing to June of that year. Nana, like Cisy, channelled her energy into developing her spelling strategies. In September of Grade One, Nana was back to writing freely on many topics. She wrote several stories in the first month back to school including her version of The  Three Little Pigs (see Figure 36). Figure 36: The three little pigs -~X7\i i ——Awl—r-'5 = Transcription: The three little pig went to the big pig's house and they eat the wolf all up. 86 What was becoming more evident in this study was that children have an enormous capacity for learning and always approach their tasks intelligently. The children who were expected and permitted to discover their own route to learning a written language, always did. LEVEL 4 - Transitional At this level, the children underwent a transition from reliance on phonology or sound for representing words in spelling to greater reliance on visual and morphological representations which incorporated aspects of grammar and meaning (see figure 37). Contrast this writing by Nana in Figure 37 with one of her earlier pieces (see Figure The Form Figure 37: Transitional spelling Transcription: A bee hive All the bee are sleeping 38). Figure 38: Phonetic spelling Transcription: The bear draw a hole. 87 Nana wrote one piece in June of her Kindergarten year and the other in November of her Grade One year. Obviously, the second sample (Figure 38) was her earlier piece. In that writing sample, she used primarily an auditory strategy; she attempted to represent every surface sound that she heard in each word. It was on this basis that she represented the sound /dr/ as /j/, and assigned two loi sounds for the word "hole". At the same time, the word "the" was strictly a memorized unit. In her later writing sample (figure 37), Nana had largely given up her original letter-sound strategy for spelling. In its place, she began to employ features of standard English. However, although she had begun to notice and to use certain features of standard spelling in her writing, she did not always use them correcdy. The silent "e" was used to mark the long vowel sound in "nite". The vowel marker clearly indicated a visual strategy since the marker, itself, had no sound. Although Nana did not spell "night" conventionally, she was making some intelligent generalizations about English spelling. Words with irregular spelling patterns were usually misspelled by the children but their misspelling had the effect of making the spelling of the word "look the way they should" (Temple, Nathan & Burns, 1982). Nana was in a stage of experimentation with standard spelling forms, a stage commonly referred to as transitional spelling (Gentry, 1982; Temple et al, 1982). Transitional spellers show evidence of a new visual strategy; they are moving from phonological to morphological and visual spelling. Although transitional spellers no longer rely on a sounding-out technique, they nevertheless continue to depend on a phonetic strategy for unfamiliar words. In Nana's writing sample (Figure 37), she fell back on a phonetic strategy in her representation of "slppe" for the word "sleeping". 88 At the same time, she demonstrated a growing awareness of morphemes when she added "-ing" to that word which reflected a visual strategy. Nana, while writing, explained that there were two ways to spell "sleeping", "slppe" and "slppeing". Morphemes are meaning-bearing units; bound morphemes are parts of words, which have meaning yet cannot stand alone. The more common morphemes used by young children were "-ing", "-ed", "-s", "-ly", "-er", "-est". The children were demonstrating an understanding of morphemes, or morphological representations, when they assigned "-ed" to words as a tense marker regardless of how it was pronounced (eg. jumped A/, tasted /ad/). Occasionally, they overgeneralized as in the following examples: wond (won); singed (sang); sleped (slept). This overgeneralization was also characteristic of ESL children's oral language development. The awareness of certain spelling patterns also resulted in many overgeneralizations (e.g. sail = saw, rallck = rock). Recall of sight words was not always accurate (see Figure 39). Figure 39: Recall of sight words Transcription: The mother Bear made porridge the three bowls was hot so the three bears went for a walk. 89 In Figure 39, Cisy showed she used both sight and sound strategies. In her spellings of "mother" and "made", she demonstrated her knowledge of two sight words but embedded one sight word in a larger word and reversed the letters in the other. She also used a known sight word (saw) to represent a different word (so). The vowel marker in "wente", and the use of particular spelling patterns in "bolls" (bowls) and in "wock" (walk) were more examples of the overgeneralizations in her visual spellings. Cisy, like Nana (Figure 37), also demonstrated a growing understanding of narrative formats with her inclusion of a title. The children revealed an awareness of three vowel markers. They applied their concepts of vowel markers very generally as a strategy for spelling words when the spelling have not been memorized. In most cases, these markers created spellings that were easily read and understood by adults. Table 7 lists examples of the children's overgeneralizations of the three vowel markers: the silent "e"; the double-vowel combination; and the "y" as a vowel (e.g. play, key). Table 7: Vowel Markers Silent E Double vowels Vowel Y wate (wait) thae (they) thay (they) pike (pick) plae (play) maykeing (making) nite (night) liek (like) nayt (night) Knite (night) nead (need) outsayd (outside) wase (was) wean (when) sayd (said) care (car) tiem (time) inSiyd (inside) 90 Table 7, Continued eate (eat) maik (make) wiyt (white) wnde (wind) siep (ship) biying (buying) ware (wear) fiyr (fire) cibe (crib) speys (space) freande (friend) famoley (family) Visual approximations occurred more frequently. Occasionally, the children used known word units within longer words (e.g. fry day = Friday, myin = mine, wayols = whales). Other times, children transposed letters within a word. Table 8 lists examples of children's visual spellings based on their sight words. Table 8: Visual Spellings Approximations of sight words siad (said) uot (out) Oen (one) tow (two) uoy (you) mlik (milk) siwm (swim) dwon (down) candys (candies) 91 It appeared that transition to standard spelling occurred as the children became more experienced readers. The better readers reread and self-corrected their writing more often. They detected morphological rules in their reading and applied them in their writing. C. Chomsky (1973) described this as a shift from a phonetic to a lexical interpretation of the spelling system. The children also became aware of punctuation and experimented with periods, question marks, exclamation marks and apostrophes. Overgeneralizations occurred in this area also; however, they were always received with respect and encouragement The fact that they were using punctuation often indicated a growing understanding of the conventions of print (see Figure 40). Figure 40: Punctuation Transcriptions: to my house He brought the present. I like Santa's coming Santa went to my house With the advent of visual spelling, some children began to show representations of the nasals /n/ and ImJ (see Figure 41). 92 Figure 41: Nasals Transcription: \^:::MMpum^::z^ The rabbit win a egg LWJDT^yn^M-W^7'O^W't- I win two. I win a car ana and the ghost want my car. I will hide but I can't hide. But I can get him! Yes I can! In Figure 41, the ratio between correctly spelled words and invented spellings was high (22:3). Not all of Ken's writing at this level showed such a high ratio. Nevertheless, his sample serves to illustrate how children's sight vocabulary had grown since their tentative beginnings as far back as in Level 2 (see earlier example, Figure 20). In the analysis of Level 4 writing samples, the number of correcdy spelled words was a factor to be considered in the classification process. During the earlier levels, the children refined their invented spellings from the initial letter-sound correspondence in letter strings to partial representations of words to total mapping of all surface sounds. With each refinement, they established strategies which were effective enough to discourage regressions to previous, less efficient strategies. For the earlier levels, sight words were noted but virtually ignored as they brought litde insight into the children's understanding of English orthography which, at the time, was largely phonetically-based. 93 Movements between the earlier levels were relatively smooth and consistent as they involved gradual refinements of auditory strategies. Movement towards a visual strategy constituted a shift in children's understanding of conventional writing and a new conceptualization of English spellings. However, this process was very tentative and gradual. Often, signs of transitional spellings were almost lost in a piece of writing demonstrating predominantly phonetic strategies. There appeared to be a much greater overlap between Levels 3 and 4 when the children's spellings were gradually evolving and moving from a reliance on sound strategies to a reliance on visual strategies. This period, when the occasional signs of visual spelling occurred but when phonetic strategies prevailed, was indicated with a dotted line preceding the Level 4 category in Appendix A. The Level 4 classification was assigned when the children's writing demonstrated a heavier dependence on visual strategies than upon sound strategies. To be considered at this level, the children's writing need to show transitional spellings mixed in with correcdy spelled words whose forms the children may have either accurately invented or memorized (Temple, Nathan & Burris, 1982). It was recognized that the children who wrote "safe" pieces, those who restricted their topics and confined their vocabulary to simple words, were more likely to present themselves as transitional spellers earlier than those who didn't. The fluent writers with the more sophisticated vocabulary, by necessity, resorted more often to previous phonetic strategies. However, it did not seem that this distorted the overall picture of children's writing development nor their general patterns of progressions. 94 Level 4 was a time of integrating sight and sound strategies in the spelling of words. This integration was crucial to future development. While transitional spellers were aware of the features of standard writing, they had not yet integrated all of these features into a systematic understanding of English spelling that worked. In determining when the children's writing could be classified as Level 4 writing, I continued to follow Gentry's (1982) guidelines in order to remain consistent with the intent of this study and with my attempt to follow Dobson's (1983) investigation of young writers as closely as possible. By the end of the Grade One year, nine of the eleven children were in Level 4. Of these nine children, four entered the fourth level in the final month of school. The Content For the purpose of discussion, I have made a distinction between transitional spelling and transitional writing. As previously mentioned, the children who wrote short, simplified pieces were more easily identified as transitional spellers since the analysis of underlying strategies was limited to a fewer number of invented spellings. However, I believe that the young children who wrote freely and fully utilized the vocabulary and structures in their oral language were the more fluent and advanced writers (see Figure 42). 95 Figure 42: Beginnings of transitional writing Ihp /i ""UP Inof cn ErtK- TfW 1 o\' lie Sirp and its^v^. V need VL«<, <,ons fVj. V rjor him OP 1 V Transcription: The Final Battle V Once a ship landed on Earth. There was red men was coming out of the ship and it was V. They need food. They eat people meat Mark was Sean's friend. V got him for food. The face is mask. It's plastic. Inside, it's ugly. It's a creature. V was taking over the world for food. Adrian's writing sample showed the beginnings of transitional spelling. Of the 23 invented spellings, Adrian relied on a phonetic strategy for 17 and a visual strategy for 6. Since the phonetic strategies prevailed, his correcdy spelled words were not given as much consideration. This was consistent with my process in analyzing my data. In earlier discussions of Level 2 and Level 3 writers, I had mentioned that some children acquired a core of sight words which appeared to be memorized as they generally showed no other signs of visual strategies (see earlier example, Figure 34). Therefore, it was more productive to focus primarily on the misspellings when attempting to discover the strategies the children employed and the concepts underlying their writing. 96 Adrian's partial retelling of a television mini-series, The Final Batde V, was incomplete but the main ingredients were there. The red uniformed aliens, V, were disguised as Earthmen in an attempt to take over the planet so that the V-creatures could have the Earth people as their food supply. The conclusion to this dilemma was missing because Adrian had yet to see the final show. Compare this writing with Nana's writing sample (see Figure 43). Figure 43: Nana's writing, a comparison JV\fc on3 my lor-DcL Picked Iv.eJ- QkOa --S.ceck . d.rjd rr,y t>rocJr Vr>st<*ol<„ Transcription: Me and my brother played hide-and-seek and my brother was hiding. He was hiding in the haystack. At the time of this writing sample, Nana had already been identified as a transitional speller for almost four months. In her story, Nana and her brother were playing hide-and-seek and her brother hid in the haystack. While both writing samples were charming, the first sample (Figure 42) perhaps was more challenging in terms of identifying key points in the story and organizing that information. In terms of content, Adrian's writing was more representative of the beginnings of transitional writing. Transitional writing can be described as writing which is in transition between "talk written down" and book-like language. The children were showing signs of 97 transitional writing at the time when they began to increase the lengths of their writing pieces. They intuitively recognized that when they wanted to talk about details or events beyond what the illustrations provided, there had to be some internal organization in their writing to carry that information in a meaningful and cohesive fashion. They also began to demonstrate an awareness that their writing must be able to stand alone without support from their illustrations. Many children at this level elected not to draw pictures or chose to illustrate after the writing was completed. Through their reading experiences and their increasing exposure to many kinds of books, the children began to induce the rules or conventions of written expression. They began to elaborate on what they had to say. Instead of writing only an introduction, a conclusion or a summary statement, the children attempted to develop their writing to lead into a conclusion. Personal views were expressed and supported. Storybook conventions were recognized and adhered to. Over the years, I had not observed many Grade One children whom I would have described as transitional writers. However, many children were beginning to show signs of transitional writing during that first grade. Again, learning is not a linear process; movement from one level of development to another is gradual and overlapping. Experiences with books and exposure to different kinds of writing seemed to play an important part in the children's development of composing strategies. Harste, Woodward and Burke (1984) suggested that perhaps experience rather than intellect and maturation, is the greatest factor in writing development. 98 The Children Young children often assumed a shared context with their audience both in their oral conversations and in their written communications regardless of whether that assumption was warranted or not. They failed to take into account the fact that some of their audience may not have shared in their experiences and, therefore, required much more background information than someone who had. They took for granted that their opinions and their viewpoints were universal and need not be explained. The children in the study group reflected this same behaviour and attitude. In conversation, they provided necessary information and elaborated on their subject when questioned. During writing sessions, there was the same kind of dialogue. The children's writing acted as a conversational opener. They elaborated further on what they read and clarified specific points. Initially, I prompted and probed for more information to get at the intended meaning of the writing. There was no attempt to have the children add details to their writing. I responded only to the content of the writing as I would in any normal conversation. Through their growing experiences with books and the many discussions arising from their writing, the children began to anticipate my needs and volunteered relevant information orally before any questions were asked. Over time, what they were able to express and the manner in which they did were reflected in their writing. Vlado, while in Grade One, developed an overwhelming interest in hockey games. Around March, he began to keep his teachers and classmates informed of the hockey results almost on a daily basis. The following examples illustrate his increasing sense of audience, his growing understanding of how written language looked, what it sounded like, and how it was used. Writing Samples Sample 1: I well tell you hou well plye at Satrday the Enmith Oaljro a gens Sneluwes Blus Sample 2: I well tell you hou wane last nayt The Vankuvr Canoks a gen's Cor gory Flamps the Flamp wene 5-1 and the Vankuvr Canoks are out the Pljofs Transcriptions I will tell you who will play at Saturday. The Edmonton Oilers against St. Louis Blues. I will tell you who won last night. The Vancouver Canucks against Calgary Flames. The Flames win 5-1 and the Vancouver Canucks are out the play-offs. Sample 3: I well tell you Who well plye today the Ednmoton Oleyrs a gans Maneesoouda Norstorss I Think Ednmoton Oleyrs well wen and the N-y Ilendrs a gans MonChreeyjol Cenireajens I Dount now who well wen there to good Hokeiy maybey MonChreeyjol Cenareajyensce in MonChreeyjlol Co I will tell you who will play today. The Edmonton Oilers against Minnesota Northstars. I think Edmonton Oilers will win. And the New York Islanders against Montreal Canadians. I don't know who will win.They're two good hockey (teams). Maybe Montreal Canadians in Montreal could. Sample 4: to Morow the New-York Ilendrs Tomorrow the New York Islanders will play a gen's Edmoton Olyjers Andy Moog or grent fojhur Wayne Geksy and jry kryie or my best plors and Mike bossy Edmoten is trig to get the Selny Cap will play against Edmonton Oilers. Andy Moog or Grant Fuhr, Wayne Gretsky and Jari Kurry are my best players and Mike Bossy. Edmonton is trying to get the Stanley Cup. Sample 5: At satr day Edmoton Olyjers play a gans New-york Ilendrs It was 5-2 and I was for Edmoton Olers Olers wan the Sanli Cap At Saturday, Edmonton Oilers play against New York Islanders It was 5-2 and I was for Edmonton Oilers. Oilers won the Stanley Cup. 100 These writing samples were written within a two-month span. Each was on the same subject matter but there were quantitative and qualitative differences among them. The first sample mentioned only who was playing; it was an introduction. The second introduced the teams, the results of the game and the consequence for the losing team. The third sample introduced the teams who were to play and predicted a winner for one game but tentatively guessed at the winner for the other because the two competing teams were so equal in ability. The fourth piece demonstrated a more conventional introduction, gave an opinion on who the best players were, and explained the purpose of these games. The fifth included a personal stance, provided the essential information and the conclusion. In terms of content, it was fairly complete. Eventually, Vlado would be able to put elements of sample 3, 4, and 5 together to produce a well thought out report. When Vlado began to provide more information, to venture opinions, and to explain cause and effect and future goals, he was demonstrating a growing understanding of what effective writers do. As with his spelling attempts, he was approximating adult models. It seemed that Vlado was becoming aware of the conventions in composition and style in much the same way young writers became aware of spelling conventions. The children who were demonstrating their awareness of the differences between "talk" and written language, who were moving towards decontextualized language, were the children who showed the beginnings of transitional writing. 101 Summary Four levels of writing development were discussed using Gentry's (1982) spelling stages as the starting point. The fifth level which incorporates standard English orthography was not reached by the children and, therefore, was omitted. During the Kindergarten year, the children gradually refined their strategies and made closer and closer approximations toward conventional spelling. Their writing ranged from scribbling to letter-like characters to real letters. All eleven children began at the first level of writing but their strategies indicated different levels of knowledge and understanding about how written language worked. However, the children showed similar patterns of overall development The children's reading of their own work demonstrated their understanding of the functions of writing. Their writing included: labelling of objects; retelling of favourite stories and T.V. shows; original stories; journal pieces; songs; and nursery rhymes. As the children began to relate speech to print, and as their alphabet knowledge increased, they eventually came to the realization that letters can represent speech sounds. By the end of the Kindergarten year, five children were using letters to stand for speech sounds. One child was already at the third level of writing while the remaining six children continued to explore the forms and functions of writing. By October of Grade One, all but two children were demonstrating letter-sound correspondences in their writing. Overall, the children were making the developmental progressions in spelling as originally identified by Read (1971). By the end of the first grade, nine of the eleven children were integrating the sight and sound strategies in their spelling of words. Their writing reflected a greater 102 reliance on the visual and morphological representations over the phonological. As their spelling became more automatic, the children began to write more and to experiment with different kinds of writing. The children's composing strategies were also developing and moving toward the conventional models. The many book experiences and the increasing exposure to various kinds of writing were accompanied by the children's gaining sense of how written language differed from speech. Observations over the two years supported the findings of early writing research. Within the terms of reference set out by Gentry (1982), the ESL children in this study demonstrated that they used similar strategies and followed the same general patterns of development in their acquisition of written language as those reported for LI children. These ESL children also showed that their written language can develop side by side with their oral language. A closer analysis of the children's written compositions indicated an increasing awareness of English forms and usage commensurate to their growth in oral English. 103 CHAPTER FIVE DISCUSSION Two questions were raised in the previous chapters. They were: 1) What do ESL children do when they are left to discover the English writing system for themselves?; and 2) Will the writing development of L2 children be similar to the writing development of LI children when the same optimal conditions for language learning are in place? In an attempt to address these questions, a two-year project was set up to look at how ESL children acquire a written language. Its purposes were to: a) investigate the processes and developments in ESL children's writing during the Kindergarten and Grade One years; and b) evaluate the resulting data of these L2 learners against the stages reported for LI learners. Eleven children who were learning English as a second language were designated as the target group to be observed. The children were in regular Kindergarten and Grade One classrooms. Since they spoke little or no English at the beginning of Kindergarten, they were referred by their classroom teachers to the English Language Centre (ELC) for extra help in learning English. These children began writing in the ELC within the first month of school. This research showed an overall pattern of development from less sophisticated to more sophisticated writing strategies and knowledge. It also revealed how children proceeded in their path of development towards conventional writing at a pace and in ways unique to them. The findings concurred with current research on children's 104 development, there appears to be a general path of development which is generalizable to all children (Y. Goodman, 1985; Calkins, 1983; Dobson, 1983; Graves, 1983; Gentry, 1982; Read, 1975). Vygotsky (1978) stated that writing development does not follow a simple, clear-cut path of conversion from one stage to the next but that it is more like "metamorphoses; that is, transformations of particular forms of written language" (p. 106). In this study, the ESL children moved from more primitive forms such as scribbling, drawing and making letter-like forms, to using strings of letters and phonetically-based invented spelling, and finally to using morphological and visual strategies for spelling. Regular orthography was not achieved by the children in the study. However, this was very typical within the general population of Grade One writers as a whole. The children's writing samples were analyzed and classified using Gentry's (1982) stages of spelling development as a frame of reference. This was to ensure that the investigation remained as consistent as possible with Dobson's (1983) study of Grade One children's writing. The Kindergarten Year All eleven children began "writing" right at the beginning of Kindergarten. They approached their writing tasks sensibly by creating forms which were most similar to the forms in their environment The children's discussions about their writing indicated that they all had made the distinction between writing and drawing, and that they understood what writing did (Baghban, 1984; Harste, Woodward & Burke,1984). However, their understanding of what "writing" looked like varied. Two children 105 began with scribble-writing. Five children began using true letters from the very first day with some numbers thrown in occasionally. At first, they had a limited repertoire of letters but this repertoire increased as they gained experience with print through books and by watching demonstrations of print in use. Four children began with mock letters and within this group, two wrote correctly spelled names amongst the letter-like forms. These names were always presented as single, unanalyzed units; there were no attempts to use the letters in the names singly or for any purposes other than to represent the people named. This changed later, however, as the children's conceptualization of writing evolved to include new understandings. The perception of commonly known words as single units made way for a new awareness that these units were made up of smaller units (letters) and that these letters could be rearranged to create new meanings. From the very beginning, the children in this study demonstrated emergent writing behaviours similar to those reported for English-speaking children (Clay, 1975). All the children except for one readily assigned messages to their writing. This one child, Nana, wrote the letters of the alphabet in and out of sequence and read each letter correctly back to the teacher. She made it clear that her letters did not make up words. When asked what words looked like, she pointed to the print in books and at the signs posted around the room. Nana did not begin to read her writing until she made her first letter-sound matches in January of her Kindergarten year. At that time, she also began showing evidence of visual memory (e.g. EM = me; YM = my, STOP); however, an auditory strategy for most of her spellings prevailed. Unlike the two children mentioned earlier, Nana did not accept her few sight words as unanalyzed 106 units. In fact, she appeared to be quite aware that letters were combined to create words. This may explain why she often confused the letter sequence of some of her sight words (e.g. "my" as YM). Possibly, Nana's growing awareness of the systematic nature of English orthography played an important part in her refusal to pretend-write or pretend-read. She knew that there was a more concrete "system" to writing than merely assigning meaning to random strings of letters. Her reluctance to play-write was not uncommon amongst children who entered school already reading. Nana's first attempt to make sense of English orthography was to represent words with letter-names. She wrote "I.M5" (I am five). This was the beginning of a phonetically-based strategy which eventually allowed her to represent any word she wanted. The rest of the group continued to write in their own way. They readily read what they wrote. Eventually, another child admitted that her writing was not "for real" but continued anyway. She, like the others, needed more time to explore and to role-play the experiences in which they wanted to participate. For these children, pretending was an effective "engagement" strategy; that is, it allowed them to be engaged in writing tasks in spite of their growing awareness that they lacked the necessary skills for conventional writing (Harste, Woodward & Burke, 1984). The children continued to refine their conceptualizations of what writing looked like and what it did. They began to organize their drawings and print in more conventional fashion. They began to make judgements about the lengths of their writing in relation to what they wanted to say. Again, these children's writing behaviours were very similar to the behaviours reported for young English-speaking writers. 107 Two children were observed experimenting with writing in different languages (see Figure 4, Chapter 4). These children who were growing up with two different writing systems, produced letter-like forms that resembled the orthographic system of both cultures. Both children were able to articulate the differences between the two systems particularly in regards to directional orientation; for example, Chinese characters were written vertically from right to left while English letters were written horizontally from left to right. What this data demonstrated was that long before formal instruction began, young children were actively making sense of the world, including the world of print (Y. Goodman, 1985; Harste, Woodward & Burke, 1984). Children's early writing samples were examples of how children made "organized" language decisions and that these organizational decisions were sociologically and contextually rooted (Hall, 1987; Harste, Woodward & Burke, 1984; Teale, 1982). By the end of the Kindergarten year, all the children had established correct letter forms and correct linear and directional orientation for English orthography. In the Kindergarten year, the children in this study appeared to have followed the general path of development described by researchers investigating young English-speaking writers. The most obvious difference between LI writers and L2 writers was in the language which the children used when reading their writing. Many of the children knew only a few English labels at the beginning of the school year. Some of the children repeatedly drew those items for which they had names. Other children drew pictures and scribble-wrote, then proceeded to read in their first language. Regardless of what and how they wrote, all the children in this study were gaining important insights into what print looked like and did. 108 During the Kindergarten year, it became evident that for the ESL children in this study, writing was an oral language activity. Again, this was consistent with the findings reported by researchers who had investigated the writing processes of English-speaking children (Graves, 1983). Through the children's drawings, the teacher-researcher was able to infer the children's areas of interest and to build many oral language activities and games around them. The children and the teacher also were able to establish common grounds on which their conversations were based. In terms of ESL methodology and programming, this aspect of writing cannot be ignored. The Grade One Year While the first level of writing was characterized by pretend-writing which ranged from scribble-writing to letter strings, the second level was marked by the children's growing insights into the alphabetic nature of writing. Four children began Grade One at Level 2 (see Appendix A). At this level, the children began to form the concept of letter-sound correspondence as they tentatively used single letters to represent words or parts of words on the basis of the initial or final sounds of the unit. They provided a partial mapping of words by omitting the vowels and using one or two consonants that demonstrated letter-sound matches (see Figure 11, Chapter 4). A letter-name strategy was also very much in evidence as words were being represented by letters on the basis of their letter-name (e.g. R = are). Initially, the letter-sound matches were few and very tentative. However, as the children found that their efforts in writing were received and valued, they developed the confidence to continue. They gradually represented more and more of the 109 perceived sounds in the words they used. By the time the children reached the third level of writing, they were representing all the sound features of the words being spelled including vowels (see Figure 29, Chapter 4). The ESL children in this study showed many letter-sound matches which, due to their pronunciation of many English words, were not as common amongst English-speaking children. However, with the increased exposure to books and to print, the children began to demonstrate a surprising ability to reconciliate what they heard auditorially in their speech with what they came to know about print in their environment. The children in this study continued to move ever closer to conventional English orthography. They demonstrated the same strategies and followed similar routes to writing as those identified for English-speaking children. The appearance of short vowels signified a major step towards conventional English orthography. The children first made letter-sound matches of the consonant sounds that were reasonably distinctive and stable (see Table 3, Chapter 4), then advanced to less obvious sounds (see Tables 4 & 6, Chapter 4). Short vowels were not represented by letter-names nor did they provide any other clues as to the way they should be spelled. Nevertheless, the children organized a consistent strategy which classified the vowel sounds on the basis of their perceived place of articulation in the mouth (see Table 5, Chapter 4). The children's strategies revealed a surprising ability to hear and make judgements about speech sounds, thereby confirming Henderson's (1980) claim that invented spellings are intelligent and systematic creations by children. By the second month of the Grade One year, two children were still identified as first level writers, six classified as second level writers and two as third (see 110 Appendix A). One child, Wendy, was difficult to categorize at this time because she was writing and reading words which she found around the classroom. She first began copying words in Kindergarten but made no references to them when "reading" her writing. This was not uncommon amongst young children at the first level of writing. However, in September of the Grade One year, she began to copy only the words which she was able to read and made no other efforts to write beyond her list of known words. This continued until mid-November when she began writing sentences and from necessity, worked out a strategy by which she was able to represent the words she could not copy. Possibly, Wendy refrained from writing freely until she figured out an efficient strategy which enabled her to do so. The children's development in writing was highly individual; the avenue to writing for one child may have been by copying words found at home or in the classrooms; for another, it may have been through using known letters to create meanings. This variance in how strategies were employed within a general path of writing development was also reported for English-speaking children (King, 1985). The children's gradual refinement of strategies brought them closer and closer to the conventional form. Factors which significantly influenced a child's progress in writing through Levels 2 to 3 fell into two categories. They were: 1) the strategies used when composing and 2) the concepts in place at the time of the production. Three strategies stood out as being particularly useful in helping the children gain control over the writing process. One effective strategy was seen in the reduction of messages once the children began to make their letter-sound matches. This gave the children better control over their writing and made rereading easier. When the Ill children were able to read their own writing, they were then able to validate their new hypotheses about written language. On the other hand, those who did not reduce their messages struggled with their rereading and frequently failed to locate their letter-sound matches. This was not surprising since the children in this study often changed the wording of the messages without changing the content. Dobson, (1983) and Hurst et al., (1983) also reported this in their investigation of young English-speaking writers. One child in particular experienced great difficulties in establishing this strategy. If the child had not been observed making letter-sound matches while writing, this investigator would not have known at which point she began to conceptualize the alphabetic principle. It was a long time before this child was able to locate her letter-sound matches since she almost always reworded her messages. There were many variables which may have affected why she experienced more difficulties than the others. On one hand, her oral language gains were notable and she was successfully learning to lengthen her verbal utterances; on the other hand, this may have conflicted with the need to reduce her messages when writing. She also stood out as the only child in the study who showed very little interest in writing, and did not even learn to write her name until mid way through the Kindergarten year. In addition, she was the youngest child in the target group. If this child had entered school at 5.1 years of age which was the mean age of entry for this study, she might not have been as noticeable. However, it must be remembered that the two most advanced children were also younger than the mean age of the group (Appendix A). One of these children reported many home experiences with books. The parents of the other child did not read to her, however, books were readily available at home. Both children were older 112 siblings. Within this study, it would appear that age and linguistic development may not necessarily be the greatest factors in writing development although their influences cannot be overlooked. Harste, Woodward & Burke (1984) suggested that experience rather than age and maturation is the greater factor in literacy learning. The point to remember, however, is that all the children in the study continued to develop as writers in very similar ways regardless of where they began on the writing continuum upon entering school. The second composing strategy could be observed in young writers' behaviour of first writing a partial message and then stopping to read over what they have written before continuing on. Some children may reread their writing many times before completing their written message. This behaviour helped to consolidate emerging concepts and to create opportunities for self-corrections and minor revisions. This strategy was particularly effective when children began to increase the length of their writing. For the ESL children, this strategy also supported their English language acquisition. Often, the children who were repeating their messages were also refining their understanding of English syntax. The third strategy was demonstrated when children held one aspect of their writing constant to allow themselves time to explore another area of writing. For example, some children repeatedly used the same sentence pattern so that they could focus on the spelling of words rather than on the structuring of sentences (see Figure 30, Chapter 4). This behaviour is not unusual in language learning situations. New linguistic understandings are best expressed through old functions or forms (Slobin, 1979). For the ESL children in this study, strategies which were used in their oral 113 language learning were often observed in their writing behaviours. This would also be consistent with how skills learned in first-language learning support second-language learning (Cummins, 1984). Certain concepts made children's access to a phonetically-based strategy easier. Children who understood that words were separate entities and readily put in spaces or space-markers between their words, were more able to locate their letter-sound matches than children who collapsed their letters into one string. This was especially true for those children who continued to hold onto the hypothesis that words must meet the minimum requirement of three or more characters (Ferreiro & Teberosky, 1982). Often too, the children were not consistent about the number of surface sounds they chose to represent. One boy wrote DRTVADR for "Darth Vader" but only S for "sword". Since he wrote a lengthy piece with no word separations, he had great difficulty indentifying his word units. In this study, once the young writers demonstrated an understanding of the alphabetic principle, they continued to refine their invented spellings until all the surface sounds were represented (see Level 3, Chapter 4). For the most part, the children's progressions in spelling development were easily followed. However, there was one notable exception to this observation. Over a five-month period, David demonstrated the alphabetic principle only occasionally and at irregular intervals. There were long periods of no letter-sound matches in his writing. The investigator observed that during this time, David "wrote" long and involved stories. In his readings, David's language was best described as confused. On the days when he made letter-sound matches, however, David's writing consisted of two or three word sentences which 114 were syntactically correct. When David's speech became more easily understood, he also began writing in full sentences with letter-sound matches for every word. It would appear that, for David, writing presented opportunities to practise what he knew about English speech. He was only able to focus on his transcription skills during writing when he was working with very familiar and well-established structures. When his control over English syntax increased, his writing development forged ahead. Of course there were other factors to be considered in David's case. At the time of the study, David was living between two homes. This created external confusion in his environment which reflected on his behaviour and in his learning. Nevertheless, David's overall progress in writing was similar to the rest of the children in the study and was consistent with the findings reported for young English-speaking writers. Knowledge of the alphabet was another factor which facilitated young children's progress in writing development. The child who knew a limited number of letters with which she could make a letter-sound match was at a disadvantage when compared to the children who knew all the letters of the alphabet. In general, ESL learners were at a disadvantage since many English-speaking children enter school already knowing the names of the letters in the alphabet. However, for most the children in the study, it was not a major obstacle. One child overcame this obstacle by placing a "filler" to represent the word for which he had no letter-sound match. This investigator observed that the "fillers" were restricted to four letters which appeared to be an effective visual cue for unknown letter-sound correspondences. The child rarely attempted to use those letters to cue a word. He merely filled in a word 115 which made sense within the context of his pictures and what he was able to read. This example illustrates the ability to problem-solve on the part of all the children in this study and how effective they were at developing coping strategies. Overall, it appeared that regardless of how individualized development was in terms of strategies employed and rate of progress, the path of development was similar for all children. The findings at all levels of development in the this study remained consistent with the finding by researchers who had investigated the emergent writing of English-speaking children (Dobson, 1983; Gentry, 1982; Temple, Nathan & Burris, 1982; Henderson & Beers, 1980; Read, 1971). Level 4 Transitional Spelling The discussion on Level 4 writing development was presented separately to highlight the qualitative change in the children's conceptualization of English orthography. Gentry (1982) referred to the spelling attempts of Level 4 writers as transitional spelling. At this level, the children underwent a transition from reliance on auditory strategies in spelling to greater reliance on the visual strategies. During the earlier levels, the children refined their invented spelling and established new strategies which were effective enough to discourage regressions to previous less efficient strategies. Movement between Levels 3 and 4, on the other hand, involved a shift in children's understanding of conventional writing and a new conceptualization of English spellings. This process was very tentative and uneven. 116 Often, signs of transitional spelling were almost lost in a piece of writing demonstrating predominantly phonetically-based strategies (see Figure 42, Chapter 4). Certain features of standard English began to appear at this fourth level of writing. Transition to conventional spelling occurred as the children became more experienced readers. As readers, they became aware of the morphological rules in their reading and began to apply them to their writing (C. Chomsky, 1973). The use of punctuation marks, however unconventional, was also a sign that young children were observing certain features of written language (see Figure 40, Chapter 4). The number of sight words increased; visual approximations (e.g. mlik = milk) occurred more frequently. Overgeneralizations of spelling patterns (e.g. sail = saw) and of silent vowel markers (e.g. nite = night) began to appear. The children regularly misspelled words but their misspelling had the effect of making the spelling of the word "look the way they should" (Temple, Nathan & Burns, 1982). Level 4 of writing development was a time of integrating sight and sound strategies in the spelling of words. While transitional spellers were aware of the features of conventional writing, they had not yet integrated all these features into a systematic understanding of English spelling that worked. In this study, nine of the eleven children were classified as Level 4 writers according to Gentry's (1982) model of spelling development. This percentage of writers in the fourth level of writing by the end of Grade One was consistent with other researchers using Gentry's model (Gunderson, 1989; Dobson, 1983). However, these researchers did not make the distinction between transitional spelling and transitional writing. Their classifications of children's work were based on the spelling 117 strategies implied in the children's writing. Over time, it has become evident that there were qualitative and quantitative differences amongst writing samples identified as transitional writing. Within this study, there was a period of time when a clear distinction could be made between transitional spelling and transitional writing. Transitional Writing In Chapter four, a distinction was made between transitional spelling and transitional writing at the fourth level of writing development (see Level 4: Content, Chapter 4). It was suggested that while the spelling strategies and composing strategies of the earlier levels were closely related and interdependent, they were much less so at Level 4. As the children became more comfortable with their invented spellings, they no longer needed to concentrate so intensely on each word produced. Writing was becoming free of the forced phonological considerations of the earlier levels (Temple, Nathan & Burris, 1982). The children began to write more widely and freely as their spelling became more automatic. The children also began to write as readers (McKenzie, 1985). Through their reading experiences and their increasing exposure to many kinds of books, the children began to induce the rules or conventions of written expression. In terms of the distinction made between transitional spelling and transitional writing, none of the children in this study was identified as a transitional writer. Many, though, showed signs of the beginnings of transitional writing. Transitional 118 writing can be described as writing which is in transition between "talk written down" and book-like language. The children's writing samples were examined to determine how much they had shifted from "talk written down" to using written language structures. This shift was one of gradual synthesis and continuous movement. The boundaries between one end of the continuum and the other were elusive if, in fact, they existed at all. At the same time, transitional spellers who have not yet become transitional writers were easily identified. Calkins (1986) talked about the "general growth currents" underlying the writing behaviours of six and seven year-olds (i.e. Grade 1 & 2). She stated that most children seemed to move in these directions: " a) From writing for oneself toward writing also for an internalized audience. b) From writing for the sake of the activity itself (process) toward writing also to create a final product. c) From less to more fluency. d) From writing episodes that do not begin before or last beyond the actual penning of a text, toward broader writing episodes that encompass looking ahead and looking back, anticipating and critiquing." (Calkins, 1986: p. 67) In this study, the children often used language from books (e.g. "not by the hair of my chinney-chin-chin"), but their writing did not reflect the behaviours or attitudes of more mature writers who kept their audience in mind and who demonstrated an awareness of the elements of the various forms of composition. Just as there were overlaps between Level 3 and 4 in children's spelling when they shifted from auditory 119 to visual strategies (see Appendix A), there were overlaps in the children's compositions as they gradually evolved and moved from one or two sentences to longer, more detailed pieces of writing. Semantic spelling units such as "one The PoAtM" (Once upon a time), which signalled fairytales, were signs of transitional writing in much the same way "nite" (night) represented a sign of transitional spelling. These "signs" increased until the rudimentary structures of mature writing were self-evident. Vygotsky (1978) best described the growth in children's writing as "metamorphic" as opposed to stage-like. For Level 4 writers, lengthier pieces were moves toward transitional writing. For example, the child who was prepared to write a longer piece of writing was more likely to include more elements of a story grammar than the child who restricted herself to two sentences (see Figures 42 & 43). In this author's experience, it is not uncommon for children to jump from a four-sentence story to a four-page story in a matter of days, although this is more typically found in the second grade rather than the first. The pieces of writing by transitional writers were generally incomplete. These writers usually wrote long elaborate introductions and rarely concluded their stories satisfactorily in spite of the obligatory "The End". Calkins (1986) referred to these children who were typically in Grade Two, as being in "a land of opposites" where extremes in writing behaviours were most apparent. Some of the children in this study had only begun increasing the length of their writing near the end of the Grade One year. Their work indicated the beginnings of transitional writing. Their writing behaviours began to show "the growth currents" described by Calkins (1986) in her observations of English-speaking children. In this 120 study, it would appear that the shift in spelling strategies must be made before the shift in composing strategies can be accommodated. This is consistent with the shifts of priority and focus in all aspects of children's writing development. It must be remembered that this investigation followed the children only to the end of the Grade One year. If it had been possible to follow the children into their second and third year, there may have been at least one student who would not have necessarily conformed to the order of shifts suggested in the discussion of Level 4 transitional writing. It has been this author's experience that there are usually some students who clearly demonstrate more advanced abilities in composing than their spelling abilities would indicate. While these students are not typical, they are not uncommon either. Unfortunately, a large number of these children are referred to Learning Assistance Centres for extra help in the language arts because they are perceived to have "learning problems." However, it has been this author's experience that when these students are put in the same environment as the children in this study, they also flourish and become fluent writers in spite of their spelling handicaps. Clearly, teachers must evaluate each of their students individually and longitudinally, and empower their students to discover their own routes to learning. The Impact of Literature on ESL Learning In any discussion on children's compositions, consideration must be paid to the content of the children's writing. What the children chose to write about was just as important as how they wrote it. Choice of topic reflected the children's interests and 121 their attempts to make sense of their world. For ESL children, it was a means to making sense of a new language, too. What stood out in this study was the high percentage of Kindergarten writing samples which had direct links to book-reading sessions. Non-ESL Kindergarten children's writing efforts were more typically connected to their own identities. Writing was a matter of social learning, of playful exploration and of self-expression; thus, children's writing centred around themselves and their world (Dyson, 1985). On the other hand, the ESL children's writing reflected the world of children's books. In retrospect, it was not surprising that the ESL children in this study should show "book talk" in their writing. Trade books were the basis of their oral language program. Widdowson (1975) encouraged the use of literature in ESL programs. He supported the view of literature as discourse and as a way of looking at how a language can be used to express reality. Children's picture books were particularly accessible since the large picture format transcended the language constraints of print. In this study, there appeared to be two main factors affecting the amount of book-talk in the children's writing. The first related to how the children were "taught" English. Since a major part of the group language activities revolved around book-reading sessions, the children began to pick up words and phrases from the books they shared. The language they heard and practised was almost always connected with books. For example, the story of The Three Bears showed a tremendous influence in the children's English language learning. One child, when asked about his weekend, replied "not too good, not too bad, just right". Another child, in response to the teacher's impatient request to wait just a few minutes until the bell, said "No, I really 122 have a big-size pee". The language in storybooks became their jargon. It was no wonder then that there was such an inordinate amount of book-talk in the children's writing. The second factor was related to the children's free choice of writing topics. In reality, the children's choices were restricted by their lack of English. The children quickly realized that if they were to talk (and write) about their pictures, they needed to choose a topic which made the best use of their limited repertoire of English words. Not surprisingly, they chose favourite songs, nursery rhymes or stories. One child scribble-wrote to the bottom of the page and sang a complete verse of Old MacDonald  Had a Farm. A second child, after hearing several readings of Eggbert the Egg (author unknown), drew a picture of an egg and responded to Eggbert's question, "What will I be?" She wrote a string of letters and read "I be a snake". Yet another child in Kindergarten pretend-wrote: "Father find the 'ten-up' and tried to pull it and call the Mother and she can't pull it and then they pull it" (The Big, Enormous Turnip by Alexei Tolstoy). These examples, amongst many others, illustrated how literature supported the children's oral language development and how their reading, writing and speaking skills were developing concurrently. C. Chomsky (1972) found a strong relationship between children's exposure to written stories and their rate of linguistic development. This is becoming more evident in the Kindergarten writing data. Snow & Ninio (1986) reported how reading a book over and over again exposed children to more complex, more elaborate and more decontextualized language than almost any other kind of interaction. The contextualized language of the talk surrounding book-reading events enabled children to understand and to produce the decontextualized language of books which are crucial prerequisites to literacy (Wells, 123 1985; Snow, 1983). For the ESL children, these kinds of literacy events established a bridge between oral and written language right from the beginning. The same language strategies were used to make sense of oral communication and print. By the time the children entered Grade One, their English language had improved enough to allow a much wider range of topics in writing. There were many more stories about themselves and their world but the influence of literature was still evident. In this study, the children's writing tended to be more like narratives than expository text. Mason & Allen (1986) claimed that narratives were easier because they contained structures which were more commonly found in speech. As the children's English proficiency increased, they began to tell longer, more detailed stories. This stopped abrupdy, however, when they began to conceptualize the alphabetic principle in English orthograghy (see Level 2, Chapter 4). Nevertheless, their experiences in storytelling held them in good stead when they began to write longer stories at a later time (see Level 4, Chapter 4). Expository-like writing came later despite the children's early exposure to non-fiction books (see Figure 33, Chapter 4). Possibly, this was due to the difficulty of expository texts for children who typically lack the necessary content as well as structural knowledge for comprehension (Boch & Brewer, 1985). Another possible reason may be in the fact that a fewer number of non-fiction books was being read to the children in comparison to the number of narrative books. The children in this study varied gready in their amount of book experience at home. The quality of interactions over books also differed. However, there appeared to be no appreciable difference in the children's overall ability to bridge the gap 124 between oral and written language. The children with more book experiences tended to be further along the writing continuum but all the children in the study demonstrated the same ability to move along similar continuums. The findings in this study clearly support the important role of literature in literacy and language learning within an ESL program. Hall (1987) reinforced the current viewpoint that exposure to books read aloud provided children with access to a variety of styles of written language use and to the conventions involved in talking like a book. It would appear that the writing samples in this study have demonstrated the strong connections between book-reading sessions and the oral language development of young ESL children. Limitations and Implications It was not the intent of the author to make any conclusive statements based solely on the findings of this study. Certainly, the size of the sample group alone would make this inadvisable. Nevertheless, given the substantial support of current research in young children's acquisition of writing, the present study provides a strong argument for researchers and educators in the area of second-language acquisition to consider seriously a position on language learning, oral and written, which is more in line with emergent literacy research. Graves (1981b) talked about how good research needs to come from four levels of investigation. He emphasized how case-studies (level 1) and small group studies (level 2) are essential prerequisites or co-requisites to larger-scale research (levels 3 & 4). Data collected in longitudinal and smaller-scale studies usually point the way to 125 discovering new variables not seen in the larger data gathering. However, these variables may, in turn, be the focus for larger group studies later. Also, the product analyses of larger groups can be further investigated for their process implications in case-studies or smaller group studies (Graves, 1981a, 1981b, 1980; K. Goodman & Y. Goodman, 1977). As a teacher-researcher, the author was in a good position to observe how the classroom setting influenced the children's learning and how all the participants within the classroom environment affected each other. Graves (1981a) warned that research about writing must include the full context of human behaviour and environment before its findings can be translated into classroom practice. The results of this investigation have significant implications for ESL methodology. The findings showed that these ESL children were able to discover the English writing system for themselves and that they could do this regardless of how much or how little English they had. It appears that these children acquire their new language, oral and written, in the same way and under the same conditions English-speaking children learn their first language. This would seem to give credence to the L1=L2 hypothesis and suggest a re-evaluation of how children traditionally are taught a second language. The implications in terms of learning environment and the roles of teachers are evident. An environment which encourages and provides the time for students to find their own route to learning is crucial to making school success attainable for all students. The findings supported the picture of an ESL child as an active "constructor of language" rather than as a passive recipient of English structures and vocabulary. 126 Traditional programs which teach "simplified" sequences and subskills go against the order of events found in this study. Children who were expected to compose a message and then transcribe it onto paper began learning a written language through the whole act of writing. The transcription skills were learned in the process and only later, were analyzed in their part. This would suggest that skills lessons in phonics, printing and spelling would be much more meaningful and effective if they were taught after the children have demonstrated some knowledge of them. This kind of instruction occurs within what Vygotsky (1962) described as children's "zone of proximal development", or the area between what the learners know and what they come to know with assistance. Teachers need to find out what the children already know in order to tell where effective instruction can start. ESL teachers, in particular, need to capitalize on the knowledge that ESL children bring to the task of learning a language. It would seem that children's learning flourishes when they are allowed some degree of control over their own actions and when they can interact with adults who are receptive, who are less concerned with lightness and wrongness and more likely to respond in ways that stretch thinking. Literature played a key role in this investigation. Although the stated intention of the study was to examine how ESL children learned to write, the literacy events and the demonstrations of literacy within the learning environment were forces to be considered. The children showed in their writing the extent to which books influenced their speech development and their "talk written down". As the children's writing progressed, they easily incorporated many conventions of written language. The development in one area of language supported and reinforced learning in the other. 127 DeStephano (1978) argues that oral language, reading and writing are "outputs from a cognitively managed set of communicative competencies" and that one enriches the other through the many experiences in all. The connections between the reading, writing and speech development of ESL learners became evident in this study; however, further research is needed to articulate the nature of these connections where second-language learners are concerned. The findings so far, indicate that a literature-based language program for ESL children should be seriously considered in an ESL curriculum. Further study is also needed to take a closer look at ESL children's emerging forms of composition beyond the Grade One year. The interactive nature of language development, oral and written, and the part played by care-givers in that development must be of particular interest to teachers. Research which extends the work done by Calkins (1986) and Hansen (1987) would give greater insights into how the changing roles of the teacher affect the writing processes of ESL students. Teachers need to move from the perspective of "teaching writing" to "teaching writers" (Calkins, 1985). Conclusion The results of the present study point to the similarities between how LI children and L2 children learn to write when challenged to discover the English writing system for themselves. The ESL children in this investigation appeared to have progressed along the same general route of development towards conventional orthography as did their English-speaking counterparts in other studies. At the same time, they developed at differing rates and utilized strategies in ways that were highly 128 individualized. The theoretical perspective of writing as a developmental process was evident throughout the study. The findings were consistent with the current research on literacy learning as reviewed by Hall (1987), Mason & Allen (1986) and Teale & Sulzby (1986). It would appear that an accepting, social environment as described by these researchers is critical to successful language learning, oral and written, and to learning in general. It followed, then, that a teacher's most important role is to set up the "prepared environment" which will allow learning to occur naturally (Montessori, 1948). Teachers have a responsibility for creating a classroom context which offers as many opportunities for learning as possible. At the same time, they need to build in many subde constraints so that students will be guided by the obvious aspects of the situation (Newman, 1987). For example, when the children in this study were presented with books wth lined paper, they were also being extended a specific invitation to write. Regardless of the manner in which a child responds, the teacher expects that the child will eventually come to make sense of the print and the demonstrations of literacy in his environment. The nature of adult interactions also appear crucial in influencing children's learning and language development. The teacher follows the children's lead and acts as a faciliator and a collaborator, thus allowing the students a greater control over their own learning. Not only does this empower learners to think and act, but it also ensures a more success-oriented environment for all students regardless of ability and previous knowledge. 129 In this study, there were no general intelligence scores, detailed information of parental background or first-language assessments on the children available. Information concerning home literacy events, including book experiences, were attained through informal conversations with the children and their parents. The child who had the most book-reading experience at home was the most outstanding student overall; that is, she progressed more rapidly through the earlier levels of writing development. However, there were other children who reportedly had few or no previous home experiences with books who, despite a slower beginning, were catching up by the end of the first grade. The children did not necessarily progress at the same rates nor was it consistent within an individual child. Some children seemed to develop faster than other children, and faster at some times than at others. Generally speaking, however, it would appear that the children who demonstrated the most advanced concepts about writing at the beginning of the study were also the more advanced students by the end. Throughout the investigation, the children were actively engaged in their task of learning to write. They were continually refining and re-defining their existing concepts about written language and getting ever closer to the conventional forms. Their explorations of the forms and functions of writing were always intellectually- and logically-based. Within a school environment which attemped to replicate the positive experiences surrounding first-language learning, the children in this study flourished and emerged as young writers in control of their craft. REFERENCES 130 Allen, J., Combs, J., Hendricks, M, Nash, P. & Wilson, S. (1988). Studying change: Teachers who become researchers. Language Arts, 65(4), 379-387. Altwerger, B. Edelsky, C. & Flores, B. (1987). Whole language: What's new? Reading Teacher, 41(2), 144-154. Anderson, A. & Stokes, S. (1984). Social and institutional influences on the development and practice of literacy. In H. Goelman, A. Oberg & F. Smith (Eds.), Awakening to literacy. Exeter, NH: Heinemann Educational Books. Applebee, A. N. (1978). The child's concept of story: Ages two to seventeen. Chicago, IL: University of Chicago Press. Ashworth, M. (1988). Blessed with bilingual brains. Vancouver, BC: Pacific Educational Press. Baghban, M. (1984). Our daughter learns to read and write: A case study from birth  to three. Newark, DL: International Reading Association. Beers, J. & Beers, C. (1980). Vowel spelling strategies among first and second graders: A growing awareness of written words. Language Arts, .57(2), 166-171. Benna, L. (1985). Young ESL students and writing as a process. Unpublished master's paper, University of British Columbia, Vancouver, BC. Birbaum, J. (1980). Why should I write?: Environmental influences on children's views on writing. Theory into Practice, 19(3), 202-210. Bissex, G. (1980). Gnys at wrk: A child learns to read and write. Cambridge, MA: Harvard University Press. Bissex, G. (1981). Growing writers in the classroom. Language Arts, .58(7), 785-791. Bissex, G. (1986). On becoming teacher experts: What's a teacher-researcher? Language Arts, 6J2(5), 482-484. Boch, J. & Brewer, W. (1985). Discourse structure and mental models, (Technical report No. 343). Urbana, IL: University of Illinois, Centre for the Study of Reading. Bogdan, R. & Biklen, S. (1982). Qualitative research for education: An introduction  to theory and methods. Boston, MA: Allyn & Bacon. 131 Boomer, G. (1987). Addressing the problem of elsewhereness: A case for action research in schools. In D. Goswami & P. Stillman (Eds.), Reclaiming the  classroom: Teacher-researcher as an agency for change. Upper Montclair, NJ: Boynton/Cook. Block, E. (1986). The comprehension strategies of second language readers. Tesol  Quarterly, 20(3), 463-494. Bloom, B. S. (1964). Stability and change in human characteristics. NY: Wiley & Sons, Ltd. Bloome, D. & Green, J. (1984). Directions in the sociolinguistic study of reading. In P. D. Pearson (Ed.), Handbook of reading research. NY: Longman. Brown, R. (1973). A First language: The early stages. Cambridge, MA: Harvard University Press. Bruner, J. (1966). Toward a theory of instruction. NY: W. W. Norton and Co. Bruner, J. (1975). From communication to language: A psychological perspective. Cognition, 3, 255-287. Bruner, J. (1983). Child talk. Oxford: Oxford University Press. Bruner, J. (1984). Language, mind, and reading. In H. Goelmann, A. Oberg& F. Smith (Eds.), Awakening to literacy. Exeter, NH: Heinemann Educational Books. Calkins, L. (1983). Lessons from a child. Exeter, NH: Heinemann Educational Books. Calkins, L. (1985). Learning to think through writing. In A. Jaggar & M. T. Smith-Burke (Eds.), Observing the language learner. Urbana, IL: National Council of Teachers of English. Calkins, L. (1986). The art of teaching writing. Portsmouth, NH: Heinemann Educational Books, Inc. Cambourne, B. (1984). Language, learning and literacy. In A. Bulter & J. Turbill (Eds.), Towards a reading-writing classroom. Rozelle, NSW, Australia: Primary English Teaching Association. Cazden, C. (1972). Child language and education. NY: Holt, Rinehart & Winston. Cazden, C. (1982). Contexts for literacy: In the mind and in the classroom. Journal  of Reading Behaviour. 14, 414-447. 132 Cazden, C. (1983). Adult assistance to language development: scaffolds, models and direct instruction. In R. Parker & F. Davis (Eds.), Developing literacy: Young  children's use of literacy. Newark, DL: International Reading Association. Chomsky, C. (1971). Write first, read later. Childhood Education, 47(5),296-299. Chomsky.C. (1972). Stages in language development and reading exposure. Harvard  Educational Review, 42(1), 1-33. Chomsky, C. (1973). Reading, writing and phonology. In F. Smith (Ed.), Psycholinguistics and reading. NY: Holt, Rinehart & Winston. Chomsky, C. (1979). Approaching reading through invented spelling. In L. Resnick & P. Weaver (Eds.), Theory and practice of early reading. Hillsdale, NJ: Lawrence Erlbaum. Chomsky, N. (1957). Syntactic structures. The Hague: Mouton. Chomsky, N. (1968). Language and mind. NY: Harcourt, Brace & World. Chow, M. (1986). Nurturing the growth of writing in the Kindergarten and Grade One years: How are the ESL children doing? TESL Canada Journal, .4(1), 35-47. Chow, M., Early, M., Lowe, H. & Yeung, E. (Eds.). (1978). Suggestions andactivities  for the teaching of primary ESL students. Vancouver, BC: Vancouver School Board. Clark, M. (1976). Young fluent readers. London: Heinemann Educational Books. Clay, M. (1975). What did I write? Auckland: Heinemann Educational Books. Clay, M. (1982). Learning and teaching writing: A developmental perspective. Language  Arts, 59(1), 65-70. Clay, M. (1984). Observing young readers. Exeter, NH: Heinemann Educational Books. Cochran-Smith, M. (1984). The making of a reader. Norwood, NJ: Ablex Publishing Corporation. Cummins, J. (1984). Bilingualism and special education: Issues in assessment and  pedagogy. Clevedon, Avon, England: Multilingual Matters, Ltd. Deford, D. (1980). Young children and their writing. Theory into Practice, 19(3). 157-162. Deford, D. & Harste, J. (1983). Child language research and curriculum. Language  Arts, 59(6), 590-600. 133 DeStephano, J. (1978). Language, the learner and the school. NY: Wiley & Sons, Ltd. deVilliers, P. & deVilliers, J. (1982). Early language. Cambridge, MA: Harvard University Press. Doake, D. (1979, April). Book experience and emergent reading behaviour. Paper presented at the Preconvention Institute, International Reading Association Annual Convention, Atlanta, GA. Doake, D. (1985). Reading-like behaviour: Its role in learning to read. In A. Jaggar & M. T. Smith-Burke (Eds.), Observing the language learner. Newark, DL: International Reading Association. Dobson, L. (1983). The progress of early writers as they discover written language for themselves. (Report No. 83:11). Vancouver, BC: Educational Research Institute of British Columbia. (ERIC Document Reproduction Service No. ED 235 505). Dobson, L. (1987). Connections in learning to write and read: A study of children's development through Kindergarten and Grade One. Unpublished doctoral thesis. Simon Fraser University, Vancouver.B.C. Dobson L. & Hurst; M. (1986). How do young children learn to read and write when meaning is the major focus? (Report No. 86). Vancouver, BC: Educational Research Institute of British Columbia. Donaldson, M. (1978). Children's minds. Glasgow: William Collins Sons& Co., Ltd. Dyson, A. (1981). Oral language: The rooting system for learning to write. Language  Arts, 58(7), 776-784. Dyson, A. (1982). Teachers and young children: Missed connections in teaching/learning to write. Language Arts, 59(7), 674-680. Dyson, A. (1984). "N spell my Grandmama": Fostering early thinking about print. The Reading Teacher, 3, 262-271. Dyson, A. (1985). Puzzles, paints, and pencils: Writing emerges. Educational Horizons, M(D, 13-16. Early, M. (1976). The categorization of the speech sounds of English by non-native children. Unpublished master's thesis. University of British Columbia, Vancouver, BC. Ferreiro, E. & Teberosky, A. (1982). Literacy before schooling, trans. K. Goodman Castro. Exeter, NH: Heinemann Educational Books. 134 Forester, A. (1980). Learning to spell by spelling. Theory into Practice. 19(3), 186-193. Genishi, C. & Dyson, A. (1984). Language assessment in the early years. Norwood, NJ: Ablex Publishing Corporation. Gentry, R. (1978). Early spelling strategies. Elementary School Journal, 79, 88-92. Gentry, R. (1982). An analysis of developmental spelling in GNYS AT WRK. The Reading Teacher, 36(2), 192-200. Giacobbe, M. (1981). Kids can write the first week of school. Learning, 9, 130-132. Gibson, E. & Levin, H. (1978). The psychology of reading. Cambridge, MA:MfT Press. Goodman, K. (1969). Analysis of oral reading miscues: Applied psycholinguistics. Reading Research Quarterly, 5(1), 9-30. Goodman, K. (1985). What's whole in whole language? NY: Scholastic, Inc. Goodman, K. & Goodman, Y. (1977). Learning about psycholinguistic processby analyzing oral reading. Harvard Educational Review, 42(3), 317-333. Goodman, Y. (1980). The roots of literacy, Claremont Reading Conference, 40, 1-32. Goodman, Y. (1982). Making connections. Language Arts, 59(5), 433-436. Goodman, Y. & Altwerger, B. (1981). Print awareness in preschool children: A working paper. Tucson, AZ: Program in Language and Literacy, Arizonal Center for Research and Development, University of Arizona. Graves, D. (1980). A new look at writing research. Language Arts, 57(8),913-919. Graves, D. (1981a). Where have all the teachers gone? Language Arts, 58(4), 492-497. Graves, D. (1981b). Writing research for the eighties: What is needed. Language Arts, 58(2), 197-206. Graves, D. (1982). Patterns of child control of the writing process. In R. D. Walshe (Ed.), Donald Graves in Australia. Exeter, NH: Heinemann Educational Books. Graves, D. (1983). Writing: Teachers and children at work. Exeter, NH: Heinemann Educational Books. Gunderson, L. (1985). A survey of L2 reading instruction in British Columbia. The  Canadian Modern Language Review, 42(1). 44-55. 135 Gunderson, L. (1989, March). Whole language and ESL students: What is it? Does it work? Paper presented at TEAL '89, Teachers of English as an Additional Language Conference, Vancouver, BC. Haley-James, S. (1982). When are children ready to write? Language Arts, 59(5). 458-463. Hall, N. (1987). The emergence of literacy. Portsmouth, NH: Heinemann Educational Books. Halliday, M. (1973). Explorations in the functions of language. London: Edward Arnold. Halliday, M. (1975). Learning how to mean: Explorations in the development of  language. NY: Elsevier North Holland. Hansen, J. (1987). When writers read. Portsmouth, NH: Heinemann Educational Books. Harste, J. & Burke, C. (1980). Examining instructional assumptions: The child as informant. Theory into Practice, 19(3), 170-178. Harste, J., Burke, C. & Woodward, V. (1982). Children's language and world: Initial encounters with print. In J. Langer & M. T. Smith-Burke (Eds.), Reader meets  author/bridging the gap. Newark, DL: International Reading Association. Harste, J., Woodward, V. & Burke, C. (1984). Language stories and literacy lessons. Portsmouth, NH: Heinemann Educational Books. Heath, S. (1982). What no bedtime story means: Narrative skills at home and school. Language in Society, 2, 49-76. Heath, S. (1983). Ways with words: Ethnography of communication, communities, and  classrooms. Cambridge, England: Cambridge University Press. Henderson, E. (1980). Developmental concepts of word. In E. Henderson& J. Beers (Eds.), Developmental and cognitive aspects of learning to spell: A reflection  of word knowledge. Newark, DL: International Reading Association. Henderson, E. & Beers, J. (1980). Developmental and cognitive aspects of learning to  spell: A reflection of word knowledge. Newark, DL: International Reading Association. Hickman, J. (1982). A new perspective on response to literature: Research in an elementary school setting. Research in the Teaching of English, 12(4), 343-354. Hildreth, G. (1936). Developmental sequences in name writing. Child Development, 1, 291-303. 136 Holdaway, D. (1979). The foundations of literacy. Sydney: Ashton Scholastic. Huey, E. B. (1908). The psychology and pedagogy of reading. NY: Macrnillan. Humes, A. (1983). Research on the composing process. Review of Educational  Research, 53(2), 201-216. Hurst, M. (1982). The influence of evolving theory and practice on teaching emergent readers and writers. Unpublished master's paper. University of British Columbia, Vancouver, BC. Hurst, M., Dobson, L., Chow, M., Nucich, J., Stickley, L. & Smith, G. (1983). A program to foster literacy: Early steps in learning to write. Vancouver, BC: B. C. Teachers' Federation Lesson Aids.(ERIC Document Reproduction Service No. 244 511). Hymes, D. (1972). Introduction, In C. Cazden, V. John & D. Hymes (Eds.), Functions  of language in the classroom. NY: Teachers College Press. Jagger, A. & Smith-Burke, T. (1985). Observing the language learner. Newark, DL: International Reading Association. Kantor, K., Kirby, D. & Goetz, J. (1982). Research in context: Ethnographic studies in English education. Research in the Teaching of English, 15(4). 293-309. King, M. (1985). Language and language learning for child watchers. In A. Jaggar & M. T. Smith-Burke (Eds.), Observing the language learner. Newark, DL: International Reading Association. Labov, W. (1969). The study of non-standard English. Urbana, IL: National Council of Teachers of English. Legrun, A. (1932). Wie und schreiben Kindergarenzoglinge? Leitschrift fur  padagogische psychlogie, 33. 322-331. Lindfors, J. (1985). Understanding the development of language structure. In A. Jaggar & M. T. Smith-Burke (Tids.), Observing the language learner. Newark, DL: International Reading Association. Lindholm, K. (1980). Bilingual children: Some interpretations of cognitive and linguistic development. In K. E. Nelson (Ed.), Children's language (Vol. 2). NY: Gardner Press. McConochie, J. (1982). All this fiddle: Enhancing language awarness through poetry. In M. Hines & W. Rutherford (Eds.), On TESOL '81. Washington, DC: TESOL. 137 McCracken, R. & McCracken, M. (1985). Reading is only the tiger's tail. Winnipeg, Canada: Peguis Publishers Ltd. McKay, S. (1982). Literature in the ESL classroom. TESOL Quarterly, 16(4), 529-536. McKenzie, M. (1985). Classroom contexts for language and literacy. In A. Jaggar & M. T. Smith-Burke (Eds.), Observing the language learner. Newark, DL: International Reading Association. Mason, J. & Allen, J. (DRAFT, April 4, 1986). A review of emergent literacy with implications for research and practice in reading. Unpublished manuscript. Meek, M. (1982). Learning to read. London: The Bodley Head. Menyuk, P. (1971). The acquisition and development of language. Englewood Cliffs, NJ: Prentice-Hall. Milz, V. (1980). First graders can write: Focus on communication. Theory into  Practice, 19, 179-185. Montessori, M. (1948). The discovery of the child. Adyar, Madras, India: Kalakshetra Publications. Murray, D. (1977). Write research to be read. Special report to the Ford Foundation. Newman, J. (1987). Learning to teach by uncovering our assumptions. Language Arts, £4(7), 727-737. Ninio, A. & Bruner, J. (1978). The achievement and antecedents of labelling. Journal  of Child Language, 5, 1-15. Piaget, J. (1974). The language and thought of the child, trans. M. Gabain. NY: The New American Library, Inc. Piaget, J. (1977). The development of thought: Equilibration of cognitivestructures. NY: Viking. Piaget, J. & Inhelder, B. (1969). The psychology of the child. NY: Basic Books. Piper, T. (1987). Stories and the teaching of language in a Grade two ESL class. Canadian Journal of English Language Arts, 10(2), 13-17. Povey, J. (1967). Literature in TESOL programs: The language and the culture. TESOL Quarterly, 1.(2), 40-46. Read, C. (1971). Preschool children's knowledge of English phonology. Harvard  Educational Review, 41(1), 1-34. 138 Read, C. (1975). Children's categorization of speech sounds in English, research report no. 14. Urbana, IL: National Council of Teachersof English. Read, C. & Hodges, R. (1982). Spelling. Encyclopedia of educational research, 5th ed. NY: Macmillan. Reid, S. (1988). The 1988 survey of pupils for whom English is a second language  in Vancouver schools. (Research report 88.07). Vancouver, BC: Vancouver School Board. Rich, S. (1985). Restoring power to teachers: The impact of "Whole Language". Language Arts. 62(7), 717-724. Schieffelin, B. & Cochran-Smith, M. (1984). Learning to read culturally: Literacy before schooling. In H. Goelmann, A. Oberg and F. Smith(Eds.), Awakening to  literacy. Exeter, NH: Heinemann EducationalBooks. Scollen, R. & Scollen, S. (1982). Narrative, literacy and face in interethnic  communication. Norwood, NJ: Albex Publishing Co. Sims, F. (1982). Dialect and reading: Toward redefining the issues. In J. A. Langer & M. T. Smith-Burke (Eds.), Reader meets author/bridging the gap. Urbana, IL: National Council of Teachers of English. Slobin, D. (1979). Psycholinguistics (2nd ed.). Dallas, TX: Scotts- Foresman. Smith, F. (1981). Demonstrations, engagement and sensitivity: The choice between people and programs. Language Arts. 58(6), 634-642. Smith, F. (1986). Insult to intelligence. NY: Arbor House. Smith-Burke, M. (1985). Reading and talking: Learning through interaction. In A. Jaggar & M. T. Smith-Burke (Eds.), Observing the language learner. Newark, DL: International Reading Association. Snow, C. (1983). Literacy and language: Relationships during the preschool years. Harvard Educational Review, 53, 165-189. Snow, C. & Ninio, A. (1986). The contracts of literacy: What children learn from learning to read books. In W. Teale & E. Sulzby (Eds.), Emergent literacy. Norwood, NJ: Albex Publishing Corporation. Snow, C, Perlmann, R. & Nathan, D. (1984). Why routines are different: Situational effects on children's and mothers' speech. Paper presented at the Third International Congress for the Study of Child Language, Austin, TX. 139 Spack, Rr (1985). Literature, reading, writing, and ESL: Bridging the gap. TESOL  Quarterly, 19(4), 703-725. Teale, W. (1982). Toward a theory of how children learn to read and write naturally. Language Arts. 59(6), 555-570. Teale, W. (1984). Reading to young children: Its significance for literacy development. In H. Goelmann, A. Oberg & F. Smith (Eds.), Awakening to literacy. Exeter, NH: Heinemann Educational Books. Teale, W. (1986). Home background and young children's literacy development. In W. Teale & E. Sulzby (Eds.), Emergent literacy. Norwood, NJ: Albex Publishing Corporation. Teale, W. & Sulzby, E. (1986). Emergent literacy. Norwood, NJ: Albex Publishing Corporation. Temple, C, Nathan, R. & Burris, N. (1982). The beginnings of writing. Boston, MA: Allyn & Bacon. Tough, J. (1977). The development of meaning: A study of children's use of language. London: Allen & Unwin. Vygotsky, L. (1962). Thought and language. Cambridge, MA: Massachusetts Institute of Technology. Vygotsky, L. (1978). Mind and society. Cambridge, MA: Harvard University Press. Weinstein, G. (1984). Literacy and second language acquisition: Issues and perspectives. TESOL Quarterly, 18(3), 471-484. Wells, G. (1981). Learning through language: The study of language development. (Vol. 1). Cambridge, England: Cambridge University Press. Wells, G. (1985). Preschool literacy related activities and success in school. In G. Olson, N. Torrance & A. Hildyard (Eds.), Literacy, language, and learning: The nature and consequences of reading and writing. Cambridge, England: Cambridge University Press. Wells, G. (1986). The meaning makers. Portsmouth, NH: Heinemann Educational Books. Widdowson, H. (1975). Stylistics and the teaching of literature. London: Longman Group, Ltd. 140 Widdowson, H. (1982). The use of literature. In M. Hines & W. Rutherford (Eds.), On TESOL '81. Washington, DC: TESOL. Zamel, V. (1985). Responding to student writing. TESOL Quarterly. 19(1), 79-101. APPENDIX A 141 The Progress of ESL Writers Through Developmental Stages During the Kindergarten and Grade One Years (Gentry, 1982; Dobson, 1983) Legend Level 1 mock letters real letters Level 2 Level 3 Level 4 * Subjects within target group are ordered by age from oldest to youngest. (Mean age = 5 years 1 month at school entry) ** Age at entry level = 5 years 6 months *** Age at entry level = 4 years 8 months APPENDIX B Levels of Writing Development Level 1 (Precommunicative) scribble shapes linearity mock-letters numbers random letters unstable direction prints own name Level 2 (Semi-Phonetic) single letters - knows must be specific direction controlled complete message represented semi-syllabic, letter-name strategy short vowels not represented Level 3 (Phonetic) approximating short vowels sight words appearing most surface sounds represented as perceived Level 4 (Transitional) beginning to represent vowels conventionally beginning of inflectional patterns (ed, ing) beginning to use silent vowel markers, usually unconventionally increasing use of visual patterns nasals represented word separation established Level 5 (Correct) vowel markers inflectional patterns derivational patterns adapted from Dobson (1983: Table II) based on Gentry (1982) APPEND IXC 143 RECORD OF WRITING GROWTH Name  ML PRODUCTION: cursive-like  letters (Q - capital  - small  asks/koks for help  name  a/v copy - try  relates messaoe  any single 1. for word  alphabetic appears  ALPHABETIC  specific sinqie 1. to w.  sounding aloud  spelling alouci  rereads in process  begin. & final l.'s  most consonants  lono vowels  segments - orouoinos ' - syllables  - single words  short vowel rep.  * of approx. si w's - try  -_£ most surface sounds rep'd  TRANSITIONAL  short vowels consistent  inflected patterns (ed.ino)  s of vowel markers  self-corr. - (G.Se.Sv)  COMPOSITION  l8bels  journals  Story  REREADS - match oral to writ.  finds initial match - try  W - part : relocates - J  -JM rereads  self-corr. - 6fSe,Sy - •  -_m improves product - O.Se.Sy  edits during reading - G.Se.Sy  RECORD OF WRITING GROWTH APPENDIX D Name 144 Date  Writing Process - no writing  - direction  - scribble :  - svmbols  - letters - capital?  __ - small  - numbers  - dlff. print from drawing  - relates message while writing  - composes aloud while writing  - rereads in process  Source of Message - no message  - uses illustration only  - illusr. & print unrelated . - uses illustr. but Id's print  - uses print  Quality of Message  - labels : - fragmented  - speech-like descrip. of illustr. •  - elaboration of illustr. : - oramm. acceptable . - semant. acceptable  Matching Oral to Written Language ' - oral to Illustr. only  - dips into print  - begins & ends w. print  - word to a symbol unit  - word to specific letter  - word to word  - reads with finger - adjusts or corrects  Comments upon;  - letters  - words  - punctuation  Dobson & Hurst (1987) 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items