Text-related variables in narrative picture books: Children's responses to visual and verbal texts.

Walsh, Maureen

Introduction

The description of text-related variables in narrative picture books detailed in this article developed from a larger study of the beginning reading behaviour of young children who were learning English as a second language (L2) (Walsh 1997). Several variables were investigated that may affect `reading' from the perspective of a socio-psycholinguistic model of reading (Walsh 1997, 1999a). Reader-related variables for both learners for whom English is their first language (L1) and L2 children were identified and compared with their reading behaviour (Walsh 1999b). A framework of text-related variables was formulated and investigated.

The investigation of text-related variables had a two-fold purpose. Firstly, to identify those cultural and language features of a text that may be difficult for L2 learners compared with L1 learners (Wallace 1986; Steffenson 1987). Secondly, since narrative picture books are regularly used in reading lessons for Infants' classes, the whole aspect of `reading pictures' compared with `reading words' was examined within the developing emphasis on visual literacy.

The framework of text-related variables in narrative picture books

The framework of text-related variables for narrative picture books derives from a model of reading (Walsh 1997, 1999a, 1999b) that incorporates the wider-sociocultural context that exists within the written production and reading of a text. This model, represented in Figure 1, shows that reading is a constant interaction between reader-related variables, text-related variables and the immediate and wider context of reader and text, as well as author. It represents an interactive view of reading with each component dependent on the other, with the interaction between text-related variables and reader-related variables occurring at a number of levels. Each variable is separate, as indicated by the broken lines through the middle, yet interdependent within the reading process, as indicated by the circles. The relationship between the reader and the text within the whole reading process is a two-way recursive interaction, with both reader-related variables and text-related variables contributing to reading behaviour. These occur within both an immediate context and a wider socio-cultural context.

[Figure 1 ILLUSTRATION OMITTED]

Within this context a narrative is developed through both verbal and visual texts. Thus the framework of text-related variables comprises:

* the wider socio-cultural context

* the narrative

* the verbal text

* the visual text.

Following is an explanation of these text-related variables as they occur in narrative picture books.

Wider socio-cultural context

A wider socio-cultural context exists in both the production and reading of any text. It incorporates the notion of a wider context or context of culture (Halliday 1985; Martin 1992; Wallace 1992) along with the ideological positions embodied in the social construction of a text (Luke 1995). This wider socio-cultural context may include either explicit or implicit ideology or both. For example, a multi-layered text such as Sendak's Where the Wild Things Are presents explicit, if ironic, ideology for a young reader about behaviour and parental love as the narrative shows Max learning the limits to being a `wild thing'. The implicit ideology embodies an underlying discourse of childhood being a time for imaginative play and fantasy, reflecting a changed view of childhood that had developed in Western society when Where the Wild Things Are was first published in 1963.

Different literary genres are part of the wider socio-cultural context of reading literature, as are the literary devices that comprise such structures, and intertextuality (Halliday & Hasan 1985; Meek 1988; Stephens 1992b). Readers draw on their expectations of different genres, e.g. realism or fantasy, when they experience a new text. Expectations assist readers when the traditional conventions of a genre are varied and intertextual links become part of a reader's repertoire.

The narrative

Since a narrative cannot exist in isolation, the components of narrative are located in the wider socio-cultural context (Halliday 1985; Stephens 1992b) as well as the immediate context of the book itself. The narrative embodies the `tale' and its `telling' (Toolan 1988), developed through both visual and verbal texts in picture books. The `tale' may suggest a particular theme, idea, or message with its presentation of characters, its setting in time and place, and the actions and events that develop as the plot. The `telling' of the tale (Toolan 1988, pp. 1-3) includes the various devices used by the author, such as the narrative voice and the use of linguistic, literary and stylistic effects.

In the telling of a tale, there is usually a narrative voice that presents a perspective, termed by some theorists as `focalisation' (Toolan 1988; Stephens 1992b). In the case of narratives for children, the narrative voice is often that of an all-knowing adult or `omniscient narrator', although more recent narratives have used a first person narrative with the voice of a child as focaliser. There is a `point of view' that may be explicit or implicit, and this is effected through the focalising character and the positioning of the reader (Stephens 1992b). These occur as part of the narrative but are effected through the verbal and visual texts, and particularly through the ideological constructs of the wider socio-cultural context (Luke 1995). Elements of a narrative are juxtaposed and intermingled through the visual and verbal semiotics in a variety of ways in different picture books.

Verbal text

The term `verbal text' is used here to refer to the way language is used through a particular narrative in a picture book. It may involve spoken or written modes, register, vocabulary, lexico-grammar, rhetorical devices and cohesion. It includes literary devices such as simile, metaphor, rhyme and alliteration.

Visual text

The term `visual text' is used here to mean the non-verbal semiotics that comprise a picture book. It includes the illustrator's choice of objects, their positioning on the page, colour, saturation, hue, textual medium, patterning, shape, contour, layout, composition and continuity (Doonan 1993). Techniques that determine the way visual semiotics convey meaning are use of line, angle, perspective, framing, foregrounding, horizontal and vertical placements, size of elements, symmetry and vectors (Kress & Van Leeuwen 1990; Doonan 1993).

Applying the framework of text-related variables to the two picture books

This framework of text-related variables is now applied to the two picture books, I Went Walking (Machin 1989) and Felix and Alexander (Denton 1985). These books were the two books used by the researcher in teaching whole class lessons to Kindergarten and Year 1, respectively, and in reading sessions with individual children. I Went Walking is an apparently simple picture book for young children, yet it combines an intricate weaving of verbal and visual texts that offer a beginning reader an induction to literature. Felix and Alexander develops a more substantial narrative than I Went Walking and combines different themes suitable for the child who has had more experience with stories. It is particularly popular with children because of the visual representation of houses turning into `monsters'.

Discussion of I Went Walking

Wider socio-cultural context

The wider socio-cultural context of this picture book constructs an image of childhood as idyllic and innocent. The rural setting is suggested through the illustrations in such a way that the country and animals are shown as a place for a child to explore and discover. The absence of framing in the illustrations reinforces the sense of freedom that is often associated with childhood play.

The narrative

The narrative (more a recount than a traditional narrative) is a child's account of the animals encountered during a walk. The story is told through a cumulative question and answer dialogue that alternates between the child and an implied adult. The interchange is not separated by speech marks but by different pages. The dialogue is tightly linked to the illustrations, as each statement/question alternates on each page, for example: I went walking. What did you see? I saw a black cat looking at me.

This refrain is repeated throughout with the inclusion of the next animal being the only variation: I went walking. What did you see? I saw a brown horse looking at me.

Each of the lines is accompanied by a brightly coloured illustration that shows the child, the particular animal, then part of the next animal to be introduced. The illustrations are designed so that the reader is led to predict the next animal by a hint in the illustration at the outer edge of each right hand page. Thus continuity is established with the illustrations providing an integral link between the words and the pictures.

A sequence of animals (red cow, green duck, pink pig, and yellow dog) is introduced in the same way until the final summative line: I saw a lot of animals following me!

There is no plot development, although this final line of `knowing' surprise expresses the cumulative result of the story in the illustrations. A cumulative text such as this is frequently used in stories for young children (for example A Dark, Dark Tale (Brown 1981)), My Cat Likes to Hides in Boxes (Sutton 1978)) and draws on an oral tradition of story rhymes.

This narrative operates at different levels. For example it presents conceptual knowledge appropriate for a young reader (names of animals, their colours, the actions of `walking' and `looking'). At another level it suggests an imaginative dimension that is mainly evoked through the illustrative techniques. These are discussed in the section on the visual text.

Features of the verbal text

The repetitive nature of a question/answer pattern is a feature of the verbal text and is suited to the beginning reader. Perspective is reversed in that there is no third person narrator as is common in children's stories. Instead of an `omniscient' adult narrator, an adult is present only as an audience for the child who is the teller of the tale.

At the lexico-grammatical level interpersonal meaning(2) is foregrounded through the dialogue. The dialogue reveals the relationship between adult and child, particularly through the use of the pronouns I and you. The adult does not do anything, except respond to the child with the question What did you see?. The repetition of this question tells the reader that the adult is there and that the child is secure. The comfortable status between the adult and child is conveyed by the alternation of the pronouns. They place the focus on the child, as in: I went walking. What did you see? I saw a black cat looking at me.

The ideational meaning is developed through the presentation of the animals that are introduced with their names and classifying colours. There are two main actions in the recount and these are represented through the material verb `walking', or the behavioural verbs of `looking', `see' or `saw'. The interchange of `see', `saw' and `looking at' suggests an interplay of interest and curiosity between the child and the animals. The illustrations take this suggestion further.

Features of the visual text

The visual text complements the verbal text by presenting the events with much more detail and adding actions not mentioned in the words. The adult speaker in the verbal text is not presented in the visual text, whereas the child appears on every page of the visual text. Details of the child's activities are not given in the words but are shown in the illustrations. The illustrations are clearly representational of a child and real animals and these occur on the equivalent pages with the verbal text. Colours are strong and bright against a white background. The setting is suggested rather than represented, as there are only a few instances where details of setting are included. The strong colours could be described as `full colour saturation' (Kress & van Leeuwen 1990, p. 54) which creates a `less real' effect, reinforcing the imaginary nature of the depictions.

Perspective focuses the reader on the child and the animals while simultaneously developing the interpersonal relationship between them. On a few pages the child and the animals are shown `looking' at each other as the verbal text tells us. Where this occurs there is a depiction of this mutual curiosity of child and animal. Yet in most cases the child and the animals are not just looking at each other. Rather the illustrations show what the child is doing with the animals and this usually involves some physical, tactile contact. In one illustration, for instance, the child is shown patting and cuddling the black cat, with gentle, rounded contours presenting child and animal as one.

Other illustrations provide further information such as the child patting the horse, sitting on the cow, climbing a tree, hosing the pig and being licked by the dog. There is a `story' in the visual text that both complements and extends the verbal text. This technique of `dual narrative' (Michaels & Walsh 1990) has become a predominant feature of many picture books and is developed with more sophistication in some texts (for example, Outside Over There (Sendak 1985), Changes (Browne 1990), Piggybook (Browne 1986)) to allow for a reader's response at a number of levels.

Visual continuity is maintained with parts of the animals being introduced and then continued on to the following pages. A new page presents a different animal with the child in a double-page spread. Perspective at first focuses on the child, who is represented earlier in the book as filling the pages. As the animals are introduced they become larger and the child becomes smaller. Clothing is used thematically in the visual text to suggest the playfulness and spontaneity of childhood, along with the rapscallion image of his/her hair. The child, with androgynous features, is shown dressed in bright, unmatched clothes. The image of these clothes, socks and shoes suggest that she/he is well cared for. As the child's playfulness with the animals increases, layers of clothes, shoes and socks are taken off. Yet control and safety are maintained as she/he is shown carrying these clothes at the end and the reader knows from the verbal text that there is an adult presence.

Distinctions between the verbal and the visual text are that the story in the visual text shows the relationship between the child and the animals while the verbal text presents the relationship between the child and the adult listener. Of course the two complement each other, but it is interesting to see how the `grammar' of verbal and non-verbal semiotics present different aspects of a story.

Discussion of Felix and Alexander

Wider socio-cultural context

A major feature of this narrative is the use of intertextuality. There are a number of intertextual levels within this story in both the verbal and visual texts. At a broad level the narrative draws on the fairy-tale genre with the motifs of the journey into danger then the rescue, with the accompanying change from light to dark and back into the light. This story contains allusions to Hansel and Gretel, with the characters finding their way home by following the stuffing from the toy animal's side, while drawing on the modern tradition of the adventure story with children being endangered, then being rescued without the presence or intervention of parents. The story also draws on many traditional stories of a character being rescued from the `jaws' of a `monster', while the image of Felix's side being torn by a nail reflects the Christian symbolism of sacrifice. While a young reader will not have the experience to make all these intertextual links, the story contains a number of traditions that will become part of the repertoire of the reader for future encounters with stories.

The narrative

Felix and Alexander is more detailed in its concepts, content and language than I Went Walking. As noted above, it presents a traditional theme--a child being lost and rescued--within a modern urban setting. The context of the story operates on both a real and a fantasy level. The realistic level presents the boy, Alexander, who lives in a block of flats in a city and becomes lost when he goes for a walk one afternoon. The fantasy level involves his toy dog, Felix, acting as humans do. Felix searches for Alexander and rescues him from the houses that are shown through the illustrations as changing into monsters.

The story contains the structure of a conventional narrative with each stage of the genre clearly detailed for a young reader. For example, the `orientation' introduces the two characters, Alexander and Felix, in the urban setting of a `block of flats', explains why Alexander is not allowed to have a `real pet', and establishes the friendship between the two. The `development' explains why Alexander goes for a walk every day but does not take Felix. The `complication' begins with Felix's concern when Alexander does not return and his decision to leave the house to search the city streets for his friend. There are two plot complications: Alexander needs to be rescued from the `monsters', and Felix and Alexander need to find their way home. There is a `resolution' as they return home safely and Felix's side is repaired. The `coda' adds a humorous comment about the `stitch' that Felix will have.

The story contains a number of themes that are significant for young children: having a pet, friendship, adventure, being lost and rescued. The book appeals to children because of the powerful effect of the illustrations, as evidenced in the children's responses detailed later in this paper. Felix and Alexander appeals at the imaginative and affective levels because, like such stories as The Velveteen Rabbit (Williams 1983), it is a story of a `toy' coming alive, showing human qualities of loyalty and becoming a hero by rescuing a lost child and returning him home safely.

Features of the verbal text

The verbal text clearly details the characters, their relationship and the events. There are specific markers for each stage of the narrative. Initially Alexander is introduced. The second sentence beginning with `even though' explains why Felix, the toy dog, is Alexander's `best friend'. The development is marked by a change from `Every afternoon' to `one afternoon' indicating the beginning of the complication. Suspense is increased by the change from `After a while ...' to: Night was falling and the air was becoming colder. Would Felix ever find his best friend Alexander?'

and the climactic `Suddenly ...' Markers continue throughout the narrative in this way. This linguistic style is suited to lead the young reader through the stages of a narrative. It varies between giving concrete description that is very specific and sequential, for example: Felix climbed out of the window, edged across to the drainpipe and slid down into the garden.

to introducing literary language in the description of Felix tearing his side on a nail on the fence: Tiny balls of stuffing like pale pink pearls fell from the tear.

The third person narrative is more conventional and less personal than the tone shown in I Went Walking. In contrast, the illustrations develop a strong interpersonal meaning.

Features of the visual text

The characters and setting are represented in detail in the visual text. The first illustration orients the reader by showing the inside of Alexander's room, with the cityscape through the window. While the characters are foregrounded, they are not in the centre of the picture. The thematic subject of home and the world outside is established from the beginning. Alexander is shown as dominant in the way he looks down on Felix. This positioning, and therefore status, between the two changes through the story in the visual text. Felix is drawn larger at the climactic point of the story when he rescues Alexander. Through the remaining illustrations he is positioned alongside Alexander, showing that their status is more equal after the experience.

Colour and perspective change to suit the development of anxiety and fear. As Felix becomes more worried about Alexander the pastel colours that depicted an everyday city setting become darker. Buildings loom larger in the darkness. Felix becomes central in the pictures and then perspective is used to distort reality as the houses in the street become darker and take on the mouths, eyes and faces of monsters that have captured Alexander. It is only the light of the torch in the darkness that restores reality. This illustrative technique evokes children's fears of the dark or nightmares about monsters. As the two characters find their way home safely, light is gradually reintroduced to represent safety and perspective returns to normal. The second-last illustration shows Felix and Alexander positioned in the centre and further up the page as they reach the front door of home. This door is normal and safe. The perspective of the final illustration is of safety as the reader is positioned looking through a window from outside in to the two asleep in bed inside the room. The cityscape of night is again shown in the background through another window but it is smaller, further away and no longer malevolent.

Relationship between children's responses and the description of text-related variables

Children's oral responses to both the picture books were analysed and categorised. These categories reveal significant aspects about the oral language development of L2 children and provide insights about the relationship between emergent literacy and early writing (for further details see Walsh 1997, 1999a, 1999b). Some of the children's oral responses are considered here in relation to the framework of text-related variables in order to determine whether such a framework is a useful description of text-related variables, and whether it is helpful in demonstrating how text-related variables impact on readers. Predominantly, the children's responses to the two books were related to the narrative itself and particularly to the narrative presented in the visual text, rather than to the verbal text or wider socio-cultural context. A discussion of the responses follows.

Details regarding this study

The responses of children discussed in the rest of this article were selected from a larger investigation of the reading behaviour of 98 young second language learners in classes of Kindergarten, Year 1 and Year 2 (Walsh 1997). The study was conducted in four Catholic schools. Three were in inner metropolitan areas of Sydney, where the majority of children came from lower socio-economic backgrounds and were predominately L2 speakers and the fourth had a majority of L1 children. The responses used in this article were elicited from the researcher's individual reading sessions following whole class lessons given by the researcher to Kindergarten, using I Went Walking, and to Year 1, using Felix and Alexander.

In these sessions each child was encouraged to talk about the book with statements/ questions/ prompts such as, `Tell me what you remember about the book'; `Can you remember the name of this book?'; `Tell me what you like about the book'; `Have a look at the pictures to help you remember'; `Have you ever heard a story like this before?'; `Would you like to read some of it to me/read along with me?', `Have you ever seen ... ?', `Would you ever ... ?', with variations to ensure each child was encouraged to respond spontaneously during the session at her/his own level of ability.

Responses to the wider socio-cultural context

While there were no responses that could be directly identified as responses to a wider cultural or social context, overall responses showed children had an awareness of wider contexts.

The Kindergarten children's responses to I Went Walking were focused on the visual depictions of the subject's different activities, so in this sense the children's responses were cueing in to the sense of `play' evoked in the illustrations, with an understanding of animals in a rural setting. In particular, L2 children, it was found, had more difficulty with the verbal text, as they were clearly not used to the spoken dialogue being used in a book as it is used in I Went Walking. Cueing in to this interchange involves not just knowledge of linguistic context, but familiarity with the way an adult may talk to a child in a particular social situation, as well as knowing how this spoken mode changes when it is produced in a literary form.

The wider literary context of Felix and Alexander--of someone being lost and found; of a toy acting like a human--was taken as a given by the Year 1 children, with only one child commenting, `It's not true because toys don't work'. Some children made intertextual links to Hansel and Gretel, referring to the droplets of stuffing from Felix as `breadcrumbs' as Felix and Alexander followed the stuffing to find their way back home. While there was no conscious reference from the children to the various other motifs (the `tear' in Felix's side from the nail; Felix shining the torch into the darkness; Felix being prepared to risk his own existence for Alexander) the children were evidently absorbing all aspects of the story. This book is an example of a rich literary text that can be understood while accumulating a range of associations for the reader for future reading experiences. As Stephens (1992a, p. 23) has shown in his detailed examination of intertextuality in literature: ... no text has its meaning on its own--in that its language, form, genre and allusions are shared by other texts and culture generally--the significance of a text will always lie to some extent in the relationships between texts and between the text and the culture within which it exists.

Responses to the narrative

Children's responses strongly demonstrated that they were responding to the narrative of each text in terms of verbal and visual modes in both picture books. Since the narrative embodies the `story' with the characters, setting, events and development of plot, it was clear that the majority of the children's answers were related to the story created in the words or the pictures in some way, and these types of responses would be expected from young children.

Kindergarten children's responses to I Went Walking revealed that the focus for the children was on the animals, the colours of the animals and what was happening in the pictures. Responses reflected that the principal effect of the book was achieved by the illustrations rather than the event of the boy `walking' or the dialogue of the two voices. The responses ranged from children who recalled all the animals of the story with their colours; children who remembered the animals and some of their colours; children who remembered the animals but not their colours; to children who remembered a few animals. Of the twenty-three children in this class seven were L1 and sixteen were L2. Of the seven L1 children six remembered all the animals and their colours, while one remembered the animals and not the colours. Of the L2 children only five remembered all the animals and their colours; one remembered the animals and some colours, while ten remembered the animals but not their colours.

While such a comparison is not a significant statistical finding, it does, however, add to the cumulative examples from data (Walsh 1997; 1999a) that show the L2 children are not able to complete classroom tasks associated with reading as readily as L1 children. These responses served as an oral check on the children's recall of the main features of the text, or literal comprehension.

There is no evidence to determine the reasons for the differences between the responses of the L1 and L2 children, other than to suggest that the L2 children were still struggling with mastering the vocabulary of English for animals and their colours, especially the variation of colours presented in the illustrations. It could be argued that these children, from inner city backgrounds, were more familiar with animals found in their immediate context such as cats and dogs than they might be with animals from farm settings. One child of Tongan background made a reference to her mother cooking a rabbit. I Went Walking, therefore, is an example of how texts for young children are constructed to present a particular social image that can be unfamiliar to the experience of the reader (Baker & Freebody 1989).

The majority of the children's responses were linked to the narrative. These were responses about the animals, features of the animals, or what the boy was doing. There were specific comments on the progress of events, such as `He didn't know all the animals were following him' which summarised the effect of the end of the story. Other interpretive responses such as `He's washing the pig because it's dirty' showed the children's awareness of events happening during the boy's walk. Affective responses, such as `I like that picture', showed the children's reactions to specific aspects in the illustrations of the text. Most of these responses focused on their own personal like or dislike of events in a story, or their ability to empathise, with only one child (L1) who was able to justify why she liked the cat: `I like the cat because the cat was soft and the boy was hugging it'.

The Year 1 children's responses to Felix and Alexander showed more variation, as would be expected from children a year older, yet they were mostly responses to the narrative of the visual text, as discussed in the section on `visual text'. Comments were about the houses changing to monsters and trapping Alexander, about Felix saving Alexander, about Felix's fear as he searches through the city, and about the children's satisfaction that Felix and Alexander find their way back home safely. There were comments on the characters, the characters' feelings, and the narrative events--for example: `The dog going to find friend.' Other comments were more affective or textual, with these responses linked to the story in the illustrations, for example `I liked the dog and the scary bits', or `I like the part where the houses turned into monsters'.

Responses to the verbal texts

Responses of the children to the verbal text in either of the two books were not explicitly evident, although research (particularly Fox 1983 and Wells 1985, 1986) has shown that frequent experience with the language of books assists children's language development. Only one L1 Kindergarten child made a comment that could be categorised as both `textual' and `intertextual' because of the reference to `another book in our class that says...'. This type of response demonstrates the child's awareness of language patterns of the book and that the book is a constructed text. The effect of the rhetorical pattern of this book was evident as children tried to read it or read along with the researcher. They were remembering and repeating the pattern of the question/answers: `I went walking. What did you see? I saw a .... looking at me.' Some children incorporated the word `following' into their comments. Here was evidence that language patterns in books were being internalised over time and with different reading experiences.

The children's responses were very closely tied to the narrative elements. The children talked about the characters represented in the narrative, their actions and characteristics, gaining most of this information from the illustrations. The Kindergarten children were asked who was saying `I Went Walking' and who was saying `What did you see?'. All were able to answer `the boy' to the first question but most did not understand the answer to the second question, or were not able to verbalise an awareness of a second speaker in the dialogue. Two L1 children commented that it was `someone' while one L2 child suggested it was `the cat'. The interpersonal mode of such spoken dialogue within a text seemed to be confusing to most of these children, suggesting that they were not as accustomed to spoken mode in written texts. The dialogue of the text is an interesting example of spoken mode transferred to written text, with no third person narrative to bridge the difference. The repetitive refrain of `I went walking' rather than `I went for a walk' is more stylised, placing the language within the poetic domain of literature.

Year 1 children gave more responses that were `metatextual', or reflecting the children's awareness of the structure of the text, although these often seemed to be influenced by the illustrations rather than the language of the text, for example: `The story's about a boy being lost and the dog saved him. They sewed the dog back up and put him to bed'. Such metatextual comments were more often made by L1 children, suggesting that the L2 children had not developed the language to `talk about' texts.

Responses to the visual texts

Most of the responses from both classes showed that children were responding to the illustrations, or visual text, in both books and using the pictures for understanding and retelling events in the story. Seventeen of the eighteen responses from the Kindergarten children were all prompted by or linked to the illustrations, with only one linked to the words of the text. Evidence for this is demonstrated in the way so many comments referred to actions that were not described in the verbal text, but were shown in the visual text, such as: The boy was climbing the tree ... The ducks are drinking water out of a trough. He's washing the pig ...

It is representative of these children's stage of early reading development that they would be using and relying on the pictures to help them with reading, as has been established by previous research (Clay 1979; Ferreiro & Teberosky 1982; Sulzby 1994, 1996). At the same time, it was found in this study that the illustrations clearly captured the children's attention.

The events that captured the children's interest, particularly the child climbing the tree, sitting on the cow and hosing the pig, are shown in the `story' in the pictures. The powerful impact of the magic of a literary text is reflected in the children's responses, and here we see that `reading' is much more than decoding words. Illustrations can reinforce, in the case of this text, conceptual knowledge of different animals, different colours, and knowledge of language, e.g. walking, looking and seeing. Illustrations provide a context and can represent a range of experiences about what setting animals may be in, what children might do in a rural setting, suggesting humorous and adventurous elements.

The responses of the Year 1 children to Felix and Alexander showed that the children also focused on the illustrations. The twenty-three responses were all linked to the illustrations in some way, while eight of these (four from L1 children, four from L2 children) were partly related to the verbal text. In some instances it was difficult to determine whether some comments were more affected by the words or the pictures, as in a comment like: `I like the end when they are back home in bed'.

While the children referred to the impact of being lost, it was not usually the first thing they recalled. Six children, five of whom were L1, referred in some way to the theme of being lost. In contrast, 15 children referred to the way the houses changed, thus showing the impact of the illustrations of this particular book. A few children referred to both words and pictures or some variation. Again these responses show that the main impact was from the illustrations of the houses turning into `monsters', `dinosaurs', or `robots', and on the effect of Felix's `torch' that made the monsters `change back' into houses when Felix shone the torch on them.

An issue raised by the examination of these responses was whether the L2 children were more dependent on the illustrations and on such concrete representations, because many were not reading well at this stage and because, again, they did not have the experience with stories in English to respond to the overall genre.

Recurringly, children's responses showed that the illustrations assisted or complemented their comprehension of the text. While this would be expected, it is interesting to consider whether the illustrations were assisting both the L1 and L2 children with `contextualising' meaning, and whether there were differences for L1 and L2 children. The illustrations in both the picture books were highly contextualised with the illustrations in Felix and Alexander contextualising an imaginary fear. One child did comment that `It's his imagination that's doing it'. It would be of interest in a future study to see how children would respond to picture books where the illustrations go beyond being ancillary to the text.

Teaching implications

Responses of the children to the two picture books support findings in the larger study (Walsh 1997, 1999a, 1999b) that difficulties for L2 children occurred in the written language of texts rather than in the illustrations. While difficulties occurred if children did not have any background experience or knowledge related to events depicted, in most cases the illustrations provided a context to assist understanding of what was happening, and thus begin reading. The implications for teaching, therefore, are to emphasise the importance of building on children's repertoire with every text. Even more important is the need for teachers to continually provide L2 children with varieties of spoken and written language and the way these are used in spoken and written contexts. Teachers need to thoroughly examine the linguistic structures within a text in order work with ways of using these to expand children's linguistic experience.

This discussion of the two narrative picture books reveals some of the complexities in the way texts can be structured, and therefore read, on a number of levels. The artistry of apparently simple picture books for young children can reveal different levels of meaning through the interaction between verbal and visual texts. While a young child is often able to understand the literal meanings in texts such as I Went Walking or Felix and Alexander, it requires experience of different stories, experience of the language of stories, and an understanding of conventions in illustrations to be able to `read' a picture book with multi-layered meanings in both words and pictures.

The pattern of children's responses showed that the children responded more to the narrative events, particularly to the way narrative is represented in the illustrations. Such responses are part of emergent reading behaviour, yet are of concern for children who are still at this stage in Year 1 or even Year 2. It is important for teachers to understand that some L2 children may be at an `emergent reading' stage in English because they are developing English language proficiency. If teachers can help children build a repertoire of the way stories and the language of stories are constructed, children will learn to predict, anticipate and identify patterns that provide clues to some meanings. L2 children who have not had experience of different stories in the home need far more exposure to varieties of literary genres at school to assist in the development of their literary repertoire and schemata, and their ability to use intertextuality to understand new stories.

It is important for teachers to consider how they can exploit all the variables of a text to enrich and extend children's understanding and responses. Since children are fascinated by illustrations, it is appropriate to extend this interest by exploration and discussion of visual elements. Similarly, patterns of verbal elements need to be used and scaffolded so that children have models of rich, varied language. The `unpacking' of text-related variables can assist teachers in developing literacy outcomes for their children.

(1.) Children in Infants' classes range from ages 4-8 years.

(2.) The terms `interpersonal' and `ideational' used in this article are from systemic-functional grammar (see Halliday, 1985; Martin, 1992).

References

Baker, C. D. and Freebody, P. 1989, Children's First School Books, Basil Blackwell, Oxford.

Brown, R. 1981, A Dark, Dark Tale, Scholastic Publications, London.

Browne, A. 1986, Piggybook, Julia MacRae Books, Great Britain.

Browne, A. 1990, Changes, Julia MacRae Books, Great Britain.

Clay, M. M. 1979, The Early Detection of Reading Difficulties, 2nd edn, Heinemann, Auckland.

Denton, T. 1985, Felix and Alexander, Oxford University Press, Melbourne.

Doonan, J. 1993, Looking at Pictures in Picture Books, Thimble Press, Great Britain.

Ferreiro, E. & Teberosky, A. 1982, Literacy Before Schooling, trans. K. Goddman Castro, Heinemann Educational Books, New Hampshire.

Fox, C. 1983. `Talking like a book: Young children's oral monologues', in Opening Moves Work in Progress in the Study of Children's Language Development, ed. M. Meek, Bedford Way Papers 17, Institute of Education, University of London, pp. 12-25.

Halliday, M.A.K. 1985, Spoken and Written Language, Deakin University Press, Victoria.

Halliday, M.A.K. & Hasan, R. 1985, Language, Context, and Text: Aspects of Language in a Social-semiotic Perspective, Deakin University Press, Victoria.

Kress, G. & van Leeuwen, T. 1990, Reading Images, Deakin University Press, Victoria.

Luke, A. 1995, `The social practice of reading' in Celebrating Difference Confronting Literacies, Conference Papers for the Twenty-First National Conference of the Australian Reading Association, July, pp. 167-188.

Machin, S. (illustrated by J. Vivas) 1989, I Went Walking, Omnibus Books Adelaide.

Martin, J. 1992, English Text: System and Structure, Benjamins, Amsterdam.

Meek, M. 1988, How Texts Teach What Readers Learn, The Thimble Press, Exeter.

Michaels, W. & Walsh, M. 1990, Up and Away, Oxford University Press, Melbourne.

Sendak, M. 1985, Outside Over There, Penguin Books, London.

Sendak, M. 1963, Where the Wild Things Are, Harper and Rowe, New York.

Steffensen, M.S. 1987, `The effect of context and culture on children's L2 reading: A review', in Research in Reading English in a Second Language, eds. J. Devine, P. Carrell, & D. Eskey, TESOL, Washington D.C.

Stephens, J. 1992a, Reading the Signs, Kangaroo Press, Victoria.

Stephens, J. 1992b, Language and Ideology in Children's Fiction, Longman, London.

Sulzby, E. 1994, `Children's emergent reading of favourite storybooks: A developmental study', and `Postcript' in Theoretical Models and Processes of Reading, 4th edn, eds. R.B. Ruddell, M.R. Ruddell & H. Singer, International Reading Association, Newark, Delaware, pp. 244-280.

Sulzby, E. 1996. `Roles of oral and written language as children approach conventional literacy' in Children's Early Text Construction, eds. C. Pontecorvo, Orsolini, B. Burge & L.B. Resnick, Lawrence Erlbaum Associates, Mahwah, New Jersey.

Sutton, E. (illustrated by L. Dodd) 1978, My Cat Likes to Hide in Boxes, Penguin Books, London.

Toolan, M.J. 1988, Narrative: A Critical Linguistic Introduction, Routledge, London.

Wallace, C. 1986, Learning to Read in a Multicultural Society, Pergamon Press, Oxford.

Wallace, C. 1992, Reading, Oxford University Press, Oxford.

Walsh, M. 1997, An Investigation of the Beginning Reading Behaviour of Young Second Language Children Learning to Read in English, PhD thesis, University of Sydney.

Walsh, M. 1999a, `Reader-text interaction: An analysis of young ESL children's responses to text-related variables in two narrative picture books', paper presented to Global Citizenship Languages and Literacies -- Joint National Conference of AATE, ALEA, AFMLTA, Adelaide.

Walsh, M. 1999b, `Literacy development of young second language learners beginning to read in English: the relationship between reader variables and reading behaviour,' paper presented to the Lern Conference, Penang, Malaysia.

Wells, G. 1985, Language Development in the Pre-school Years, Cambridge University Press, Cambridge.

Wells, G. 1986, The Meaning Makers. Children Learning Language and Using Language to Learn, Heinemann Educational Books, London.

Williams, M. (illustrated by M. Hague) 1983, The Velveteen Rabbit, Henry Holt and Company, New York.

Maureen Walsh is a senior lecturer at Australian Catholic University, Sydney, and lectures in English and TESOL Curriculum in undergraduate and postgraduate courses. She has published several articles on English teaching as well as authoring Story Magic (OUP, 1991) and co-authoring Up and Away (OUP, 1990). She is interested in all aspects of first and second language literacy with her PhD investigating the beginning reading behaviour of young second language learners.

Address: 5 Burraga Place Lindfield NSW 2070 Email: m.walsh@mary.acu.edu.au