Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Pathways through the woods : how the cohesive resources of colour and repetition contribute to the construction… Shoemaker, Kathryn E. 2014

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


24-ubc_2014_spring_shoemaker_kathryn.pdf [ 1.44MB ]
JSON: 24-1.0166910.json
JSON-LD: 24-1.0166910-ld.json
RDF/XML (Pretty): 24-1.0166910-rdf.xml
RDF/JSON: 24-1.0166910-rdf.json
Turtle: 24-1.0166910-turtle.txt
N-Triples: 24-1.0166910-rdf-ntriples.txt
Original Record: 24-1.0166910-source.json
Full Text

Full Text

PATHWAYS THROUGH THE WOODS: HOW THE COHESIVE RESOURCES OF COLOUR AND REPETITION CONTRIBUTE TO THE CONSTRUCTION OF COHERENT NARRATIVE PICTUREBOOK TEXTS   by Kathryn E. Shoemaker B.A., Magna Cum Laude, 1969, Immaculate Heart College M.A., 1969, The University of British Columbia   A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF    DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Language and Literacy Education)   THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)    April 2014  ? Kathryn E. Shoemaker, 2014  ii  Abstract This research focuses on how visual cohesive resources construct coherent picturebook texts and contains a close analysis of two acclaimed picturebooks: Where the Wild Things Are and How to Heal a Broken Wing. I analyze, in particular, how the cohesive resources of colour and repetition construct meaning across the visual sequential narrative of the picturebook format. Deep scholarship exists on the picturebook format, especially on the meanings, how and what readers glean from them, and how they work and support literacy and literary development; however, little systematic study has been undertaken on the construction of meaning and even less has been done that uses a social semiotic theory of multimodal meaning construction. The broader implications of my study stem from the use of the multimodal picturebook as the focus of analysis. The picturebook is a widely acknowledged literary format?a multimodal sequential narrative form that is brief, concise, and ideal for the delicate analysis of multisemiosis. This investigation is timely and relevant to the growing imperative to involve future educators and students in multimodal learning, and literary response and assessment activities.  My study reveals three significant and new resources for picturebook analysis that have the potential for application beyond children?s literature and picturebook creation as resources for multimodal learning and assessment. The first resource is a framework of resource systems and the textual environments that they construct. These systems and environments are central to the analysis and creation of picturebooks and multimodal sequential narrative texts. The second resource is a defined set of cohesive colour configurations that offer resources for assessing and editing visual cohesion in sequential visual narratives and other multimodal forms. The third resource provides a set of descriptions of the ways visual repetition constructs continuity and cohesion, and pays particular attention to the affordances of salience and framing in establishing critical resources for multimodal analysis and creation. iii  Preface This dissertation is original, unpublished, independent work, including all illustrated figures and tables, by the author, Kathryn E. Shoemaker.  iv  Table of Contents Abstract ..................................................................................................................................... ii Preface ...................................................................................................................................... iii Table of Contents ..................................................................................................................... iv List of Tables ........................................................................................................................... ix List of Figures ........................................................................................................................... x Glossary of Terms .................................................................................................................... xi Acknowledgements ................................................................................................................ xvi Chapter 1: Selecting A Path: In Search of How Meaning is Constructed in the Picturebook ............................................................................................................................... 1 1.1 Perspectives................................................................................................................... 1 1.2 Systems for Making Meaning ....................................................................................... 2 1.3 Defining the Picturebook as a Unique Format .............................................................. 5 1.4 The Differences Between Picturebooks and Illustrated Books ..................................... 6 1.5 The Multimodality of the Picturebook .......................................................................... 8 1.6 Genre and Register in Picturebooks .............................................................................. 9 1.7 Defining Text ................................................................................................................ 9 1.8 Outline of the Thesis ................................................................................................... 10 1.9 Significance of This Study .......................................................................................... 11 Chapter 2: A Brief History of the Picturebook Format ........................................................... 13 2.1 Introduction ................................................................................................................. 13 2.2 The Origins of the Picturebook ................................................................................... 14 2.2.1 Early Visual Narrative Arts ............................................................................... 14 2.2.2 Early Books ........................................................................................................ 15 2.2.3 Early Books Created for Children ...................................................................... 17 2.2.4 The Comic Divide .............................................................................................. 18 2.2.5 The Nineteenth Century?s Middle Class and the Increase in the Number of Literate Readers and Illustrators ........................................................................ 20 2.2.6 The Invention of Photography and Photolithography Initiate Developments in Narrative Art Formats ........................................................................................ 21 v  2.2.7 Playful Leisure Time Activities and Shifts in the Purposes and Formats of Illustrated Children?s Books .............................................................................. 22 2.2.8 Influential Illustrators and Creators ................................................................... 26 2.3 Changes in the Purposes of Illustration ...................................................................... 28 2.4 The Illustrated Storybook and the Picturebook .......................................................... 28 2.5 Additional Key Influences Promoting and Hindering the Evolution of the Picturebook ................................................................................................................ 31 2.6 Innovations in the Picturebook Format ....................................................................... 33 2.6.1 William Nicolson and the First Modern Picturebooks ...................................... 33 2.6.2 The Evolving Picturebook Format from 1929?1963 ......................................... 34 2.6.3 Integrating Text and Image ................................................................................ 34 2.6.4 Refining Composite Illustration in Picturebook Formats and Adding Colour ................................................................................................................ 35 2.6.5 Mechanical Colour Separation ........................................................................... 36 2.6.6 Edward Ardizonne: Using Comic Conventions ................................................. 38 2.6.7 Comic Books and Mass Trade Books for Children ........................................... 39 2.6.8 The Visualizers and Picturebook Worlds .......................................................... 40 2.6.9 Animators and the Illusion of Movement .......................................................... 41 2.6.10 Cinematic Points of View ................................................................................ 42 2.6.11 Continuous Narrative ....................................................................................... 43 2.6.12 What Comes First in a Picturebook: Words or Pictures? ................................ 44 2.6.13 Style ................................................................................................................. 45 2.6.14 The Full-Colour Picturebook ........................................................................... 47 Chapter 3: A Review of the Academic Search for How Meaning is Constructed in Picturebooks ............................................................................................................................ 50 3.1 A Review of Scholarship on the Forms and Meanings in Picturebooks ..................... 50 3.2 Form and Meaning in Sequential Narrative Picturebooks .......................................... 52 3.3 Codifying Principles of Visual Construction .............................................................. 53 3.4 Pioneering Semiotic Analysis of Picturebooks ........................................................... 54 3.5 The Application of a Theory of Language to Picturebook Analysis .......................... 56 3.6 A Systematic Approach Through Social Semiotics .................................................... 57 vi  3.7 The Ecology of the Picturebook?s Multimodal Text .................................................. 57 3.8 Categorizing Picturebooks by Word/Picture Relationships ........................................ 58 3.9 Relationships Between Words and Pictures ............................................................... 60 3.10 Review Summary ...................................................................................................... 61 Chapter 4: A Theory of Language as Social Semiotic and Its Application to the Analysis of Multisemiosis in Picturebooks ................................................................................................ 63 4.1 Introduction ................................................................................................................. 63 4.2 Stratified Model of Language ..................................................................................... 64 4.2.1 Context in Halliday?s Theory of Language ....................................................... 64 4.2.2 Strata and Context as Central Resources for Adaptations of SFL by Other Authors .............................................................................................................. 66 4.2.3 Realization and Instantiation .............................................................................. 67 4.2.4 Metafunctions .................................................................................................... 68 4.3 The Nature of Text ...................................................................................................... 69 4.4 Textual Resources of Theme and Rheme and the Information Unit of New and Given ................................................................................................................... 71 4.4.1 Challenges to the Use of Structural Resources in Visual Modalities ................ 73 4.5 Cohesive Resources for Constructing Coherent Texts ............................................... 76 4.6 Adaptations of SFL to the Analysis of Multimodal Texts .......................................... 80 4.7 The Cohesive Work of Composition .......................................................................... 83 4.7.1 Information Value .............................................................................................. 84 4.7.2 Salience and Framing ......................................................................................... 86 4.7.3 Typography and Colour as Additional Integrating Systems .............................. 87 4.7.4 Visual Cohesion and Composition as the Integrating Code of Image ............... 88 4.8 Applications of SFL to Sequential Narrative Forms................................................... 88 4.9 SFL Resources Supporting My Study......................................................................... 89 4.10 Cohesion ................................................................................................................... 90 Chapter 5: How the Cohesive Resources of Colour and Repetition Contribute to the Construction of Coherent Narrative Picturebooks .................................................................. 95 5.1 Introduction ................................................................................................................. 95 5.2 Methodology ............................................................................................................... 95 vii  5.2.1 Introduction ........................................................................................................ 95 5.2.2 Evolution of My Methodology .......................................................................... 96 5.2.3 Colour Configurations that Construct Cohesive Relations ................................ 99 5.2.4 A Systematic Perspective ................................................................................... 99 5.2.5 The Intra-textual Environment of a Text ......................................................... 100 5.3 The Integrating Code of Rhythm in Sequential Narratives ...................................... 107 5.4 Colour ....................................................................................................................... 110 5.4.1 Colour as an Integrating System ...................................................................... 111 5.4.2 Colour and Cartography ................................................................................... 112 5.4.3 Colour and Memory ......................................................................................... 113 5.4.4 Colour as a Meaning Making Resource ........................................................... 115 5.4.5 Colour and Continuity...................................................................................... 115 5.4.6 Colour Variation .............................................................................................. 116 5.4.7 Colour and Contrast ......................................................................................... 117 5.4.8 Colour and Salience, Framing, and Texture .................................................... 117 5.4.9 Colour and SFL ................................................................................................ 118 5.4.10 Colour and Typography as Integrating Codes ............................................... 121 5.4.11 Colour and Contexts ...................................................................................... 121 5.4.12 Colour as a Cohesive Resource...................................................................... 123 5.5 Repetition as a Cohesive Resource ........................................................................... 134 5.5.1 Introduction ...................................................................................................... 135 5.5.2 Repetition and Context .................................................................................... 135 5.5.3 Repetition and Contexts of Situation ............................................................... 135 5.5.4 Repetition and Key Narrative Elements and Activities ................................... 137 5.5.5 Repetition and the Information System, New and Given ................................. 138 5.5.6 Repetition and Continuity ................................................................................ 139 Chapter 6: A Close Analysis of the Cohesive Work of Colour and Repetition in Two Narrative Picturebooks.......................................................................................................... 141 6.1 Introduction ............................................................................................................... 141 6.2 Topographic, Typographic, and Narrative Environments ........................................ 141 6.2.1 Where the Wild Things Are .............................................................................. 141 viii  6.2.2 How to Heal a Broken Wing ............................................................................ 144 6.3 Colour Cohesion ....................................................................................................... 146 6.3.1 Colour Cohesion in Where the Wild Things Are .............................................. 146 6.3.2 Colour Cohesion in How to Heal a Broken Wing ............................................ 157 6.4 Repetition as a Cohesive Resource in the Book ....................................................... 159 6.4.1 Repetition in Where the Wild Things Are ........................................................ 159 6.4.2 Repetition in How to Heal a Broken Wing ...................................................... 171 Chapter 7: Reflections on Looking for Ways of Meaning .................................................... 180 7.1 Introduction ............................................................................................................... 180 7.2 The Frameworks of the Resource Systems of the Picturebook and the Intra-textual Environment and its Textual Environments ............................................................ 181 7.2.1 Resource Systems ............................................................................................ 182 7.2.2 Intra-textual Environments .............................................................................. 186 7.3 Conclusions on Colour .............................................................................................. 187 7.3.1 Introduction ...................................................................................................... 187 7.3.2 The Cohesive Work of Colour ......................................................................... 189 7.4 Conclusions about Repetition ................................................................................... 190 7.5 Significance of the Findings and Outcomes for Colour and Repetition ................... 192 7.6 The Construction of literariness in picturebooks ...................................................... 193 7.7 Contributions to Picturebook Research .................................................................... 196 7.7.1 Contribution to the Field of Children?s Book Illustration ............................... 197 7.7.2 Contribution to the Study of Sequential Visual Narratives ............................. 197 7.8 Application of My Findings and Outcomes to Other Fields ..................................... 198 7.8.1 Educators .......................................................................................................... 198 7.8.2 Creators ............................................................................................................ 201 7.9 Strengths and Limitations of this Study .................................................................... 203 7.10 Looking Ahead........................................................................................................ 204 7.11 Final words.............................................................................................................. 205 Bibliography ......................................................................................................................... 206 ix  List of Tables Table 1 Visual Cohesive Resources ........................................................................................ 77 Table 2 Verbal Cohesive Resources ....................................................................................... 78 Table 3 Resource Systems in a Picturebook ......................................................................... 183 Table 4 The Intra-textual Environment of the Text and Its Textual Environments .............. 183 x  List of Figures Figure 1 Example of full colour bleed, partial bleeds, and spots .......................................... 104 Figure 2 Cross-hatched leaves with variations of colour and density of cross-hatching ...... 117 Figure 3 Textures created by different patterns, media, and techniques .............................. 118 Figure 4 Max in his salient white wolf suit........................................................................... 120 Figure 5 Reading path of salient red beats across a sequence of small images .................... 126 Figure 6 Storyboard showing a sequence of skyscraper grays and blues ............................. 128 Figure 7 A collection of wild things arranged by colour association ................................... 147 Figure 8 ?And a forest grew.? A sequence of illustrations ................................................... 163 Figure 9 Two sketches of pizzicato beats and tree trunks .................................................... 164 Figure 10 Episodes in a composition .................................................................................... 169 xi  Glossary of Terms The following terms are specific to the discussion and analysis of how visual cohesive resources construct coherent picturebook texts using a social semiotic theory of multimodal meaning construction as examined in this dissertation. Coherence: A text has coherence when it hangs together and has the quality of wholeness. The expression hangs together refers to the logical sense of it as a complete message and that cohesive relations within it help to make it coherent as a text. Cohesive: A text is cohesive when there are semantic relations between the words, the clauses, and (in the case of picturebook text) between the words and images. Cohesive resources: In the lexicogrammar of language, the major cohesive resources include reference, ellipsis, substitution, and conjunction. In the grammar of image, cohesive resources include line, colour, tone, texture, shape, size, and scale. They also include the dynamics of composition and its related resources (framing, position, and salience) as well as rhythm and its resources (reference, ellipsis, repetition/reiteration, substitution, and conjunction).  Comics and graphic novels: Literary forms that use the visual conventions of panels, speech, thought, or sound balloons, and captions to convey the narrative. Composition: Layout or design of a single page or a double-page spread and the overall configuration of the visual elements in that space. Context of situation: Semiotic contexts of situation pertain to the interactions between individuals and groups. Material situational settings pertain to the physical setting within which activities are taking place that are relevant to the interaction through semiosis. Any number of semiotic contexts of situation may take place within a material situational setting. Continuity: The consistency of verbal and visual depictions of events within a narrative sequence. xii  Form: Types of narratives such as novels, nonfiction books, poetry, or screenplays.  Format: The material composition and arrangement of printed and bound pages of a book. Illustrated books come in a number of formats including the picturebook format and the pop-up novelty book format. I distinguish format from literary forms since a literary form such as poetry, a novel or a biography can be created in a picturebook format. Framing: A term used in literary analysis to describe a narrative structural enclosure, such as a story within a story or similar anecdotes that bookend a story. In visual literature, a frame may be a visual enclosure such as the formal frame created by borders around a picture or the informal frames within a composition such as those created by a window or doorway. Genre: This term describes variations in types of narratives and categories of content and style that occur in many kinds of book forms and formats such as novels, illustrated books, comic books, and picturebooks. A detective story, for example, is a genre of fiction that can be rendered in any of the aforementioned forms and formats. The term is used here differently from its use in SFL by Jim Martin and his colleagues. Illustrated books: Any book with illustrations that have been drawn, painted, or photographed. A category that includes all books that are illustrated such as picturebooks, pop-ups, board books and illustrated novels. However, a picturebook is a special kind of illustrated book in which words and pictures work together so that neither one alone can convey the meaning achieved by interaction between the two.  Integrating codes: Kress and van Leeuwen posit that for spatial items composition is the dynamic that put constructs relationships and integrates elements (2006, p. 177). Another integrating code is rhythm, the patterns that construct cohesion in temporal forms, which includes picturebooks, graphic novels, live action films, and animated movies (Kress & van Leeuwen, 2006, p.177). Medium: The materials such as paper, paint, ink photographic film, or digital technologies used to render images and typography.  xiii  Metafunctions: The three semiotic metafunctions include ideational, interpersonal, and textual. They are probably constitutive of all social semiotic systems. Ideational (the logical and the representational) involves activities such as doing, saying, and the language of reflection. Interpersonal addresses ways of meaning that enact relationships such as turn taking and ways of addressing others in discourse such as expressions of a speaker?s certainty about propositions. Textual is the weaving of disparate features into coherent text (Halliday & Hasan, 1985). Modality: In applications of SFL to a grammar of visual design (Kress & van Leeuwen, 2006), modality refers to the level or quality of realism in a rendered image. In other words, modality is the ?truth value or credibility? (Kress & van Leeuwen, 2006, p. 155) of an image.  Mode: The channel selected for communication such as spoken word, written word, drawing, and photography to name a few. Multiliteracies: This term acknowledges that the notion of literacy applies to the ability to construe meaning from multiple modes that include printed language on paper, language displayed digitally, spoken language, and images displayed in a variety of formats. Multimodal: Communication in at least two modes such as spoken words accompanied by facial and body gestures. This term also applies to print and screen use of image and written language and can include motion and sound. Picturebook: An illustrated book format in which words and pictures work together in a variety of complex way to tell and show a narrative. Neither mode (words or pictures) alone can tell the complete story as there is a collaboration with one mode filling in gaps in the other mode. xiv  Position: In applications of SFL and Arnheim?s (1969) principles of design, Kress and van Leeuwen (2006) consider position (the location of an element or activity in a composition or page layout) important to understanding how an item is construed. For example, items positioned in the upper half of a composition can be construed to indicate the ideal or the powerful. Position is sometimes evaluated in terms of how it may be constructing relationships of power. Salience: In the visual mode, an item has salience when it attracts immediate attention by its intensity of colour, and or its prominent size, shape, or scale.  Sequential visual narrative: The family of arts that tell stories visually and includes pantomime, puppetry, drama, opera, live action film, animation, comics, graphic novels, illustrated books, and picturebooks. Social semiotics: Halliday?s (1978) view of language in sociocultural context, social semiotics relates semiotic expression to its sociocultural context of origin. Social semiotics is Halliday?s term for his own theory of language (1978).  Stratified model of language: Halliday (1994) describes the three strata of language as 1) the semantic or the meaning strata, 2) the lexicogrammar, and 3) graphology, which is expressed by the written word or by the spoken word (phonology). The stratum of context is used as a fourth strata when language is viewed from a descriptive perspective (Matthiessen, 2007). System: A set of related features or rules/dynamics that construct a whole. In this study, system refers to a collection of resources that offer choices for the construction of meaning in language or image. Systemic Functional Linguistics (SFL): SFL is based on M. A. K. Halliday?s social semiotic theory of language. The social aspect of SFL draws on the idea of ?interpreting language within a sociocultural context, in which the culture itself is interpreted in semiotic terms?as an information system, if that terminology is preferred? (Halliday, 1978, p. 2). xv  Text: The largest unit of meaning in SFL theory. Linguistic interaction in a context of situation results in a text. It may be as short a traffic sign, as long as a novel, or the complex multimodal text of a picturebook. According to Halliday and Hasan ?text is best regarded as a SEMANTIC unit: a unit not of form but of meaning? (1976, p.2). xvi  Acknowledgements These notes of appreciation are for the generous people in my life who helped me find my path through the academic woods of children?s literature.  To a pair of word lovers, my mother, a librarian and my father, a journalist, and to my stepfather, a maker of magic, who brought a love for creating whimsy, celebrations, and magic into our family?s life. To Norma Charles who deserves the true credit for this endeavour because during her service as a B.C. Book Prize board member she asked me to create a display honouring Sheila Egoff using books from UBC?s Special Collection Library. Because of that exhibit, while reading all of Sheila?s books and a book by Phillip Aries on the history of childhood, and handling a selection of exquisite rare children?s books, an academic book bug bit me. To Professors Judi Saltman and Sheila Egoff who inspired me to dip my toes in the academic waters of children?s literature. And, a special note of appreciation to Judi Saltman for starting the Master of Arts in Children?s Literature program.  To my kitchen cabinet, my colleagues, all best friends, and members of a long-running picturebook club: Linda Bailey, Norma Charles, Nan Gregory, Bernie Lucas, Ainslie Manson, Ellen Schwartz, and Beryl Young. To my brilliant colleague, fellow illustrator and fellow Master in Children?s Literature (MACL) program participant, Cynthia Nugent To the Vancouver Children?s Literature Roundtable (VCLR) and to its founder, Dr. Ron Jobe, for his energy and dedication to bringing great children?s literature creators to Vancouver. To Dr. Margot Filipenko for the very great pleasure of working together to continue Dr. Jobe?s work overseeing the VCLR and for the pleasure of her company, friendship, and wisdom.  To all my VCLR steering committee colleagues for their love of children?s literature. xvii  To my dear friends, Jane Flick and Bob Heidbreder for always elevating the discourse, whether written or spoken, and for inspiring me to be more precise and more playful. To Dr. Theresa Rogers for encouraging me to apply to do a PhD in the department of Language and Literacy Education and for her generous guidance. To Dr. Geoff Williams who came to UBC with his love for the poetics of picturebooks, and agreed supervise my work and mentor me through the academic woods. To my committee, my dream team of Dr. Maureen Kendrick, Dr. Theresa Rogers, and Dr. Geoff Williams who have questioned me into clarity, word by word. To the Department of Literacy and Language Education (LLED) community of faculty, staff, and students who have enriched my life for many years and to whom I am forever indebted. To my editor, Karen Taylor, for her exacting editorial work and for teaching me the fine points of academic writing.  To all of these wonderful and generous people I will be forever grateful for the support patience and wisdom they have shared over the years.    1  Chapter 1: Selecting A Path: In Search of How Meaning is Constructed in the Picturebook 1.1 Perspectives Years ago, 12 eager people sat before me in a room, each one excited to learn how to illustrate children?s books. I began the class equally keen to share my knowledge because I thought I knew how to illustrate children?s books. But, even though I had illustrated over 20 books, I could give little more than a road map to one level of the illustration process. What I had to offer was my love of children?s books and illustration and a sound portrayal of the practical tasks and organization required to illustrate a book or create a portfolio. As the course progressed I began to recognize my need for a much more explicit set of descriptive resources to understand and articulate how tools of illustration create meaning. This insight was the impetus for my search for a deeper knowledge of how the resources of word and image create meaning in a visual narrative. It was also the beginning of what would become the focus of this study, a pursuit that would unite my love of illustrating with a deep interest in the socio-cultural nature of art and language.  Now, years later, preparing to teach a class in illustrating children?s books, I still bring my love of children?s books and illustration and I still present a road map for some of the tasks. However, what was once a single layered practical approach has evolved into a multilayered systematic exploration of form and the semiotic resources best suited for constructing meaning. My approach comes from an ongoing examination of what I require to articulate, as precisely as possible, the potential ways for semiotic resources to create meaning in multimodal forms of children?s literature. Also, I now bring to my current work and teaching, deeper knowledge of the social practices related to the creation, production, distribution, and use of children?s literature. This knowledge of social contexts is essential to the critical editing of my own illustration work. Thus, my current course content contains the information I needed 30 years ago when I began to illustrate books. Perhaps one of the most important pieces of knowledge I now bring to my illustration work is a deep understanding of formats and their particular affordances. Early on I did not 2  distinguish between illustrated books and picturebooks and between the finer points of the meaning-making potential associated with each format. Sitting down to illustrate my first book I had little more than my drawing ability, a few key remembered learning moments with picturebook creators Maurice Sendak, Robert McCloskey, and Leo Politi, and a burning ambition to illustrate children?s books. Now, when I begin a book, I bring my drawing skills, the same burning ambition to illustrate children?s books very, very well, hoping that this time ?I?ll get it right,? and a very special brew of knowledge. This brew, percolating in my subconscious, is a collection of important, continually deepening notions about illustrated children?s books: their various formats and their affordances, ideas about narratives, and knowledge of the visual and verbal resources that can express narrative meaning on the page. I still do some of the same work I did early in my illustration life: I read the story over and over again, I research the text and its themes, and then I sketch characters, narrative elements, and scenes to develop consistency in the rendering. Once the narrative is firmly rendered on paper and the characters and setting are developed with a view to consistency and continuity, I make a dummy (paper, page-turning facsimile) of the book. Up until that point my creative conceptual work draws intuitively on my store of different kinds of knowledge. At this point, the ever-percolating understanding of verbal and visual potentials becomes a resource for editing. 1.2 Systems for Making Meaning  My brew of ever-deepening knowledge and understanding of visual and verbal dynamics now includes a set of six meaning making systems. This addition to the brew is the most important change in my teaching approach. It is the distillation of my knowledge about illustration into a framework of the six systems of resources for constructing narrative meaning: (1) the system of narrative resources; (2) the topographical material resources of the form; (3) the verbal resources; (4) the visual resources; (5) the typographic resources; and (6) the multimodal cohesive resources?a mixture of resources from the other systems that integrate the text. Here I use the term system as it is used in Systemic Functional Linguistics (SFL) where a system is a set of features or resources from which a speaker or writer selects. Similarly an image creator may select from each system particular resources to convey meaning. This set of resource systems also supports multimodal assessment. In Chapter 3  Seven I will describe this framework in greater detail so that I can refer to notions defined in Chapter Four. This framework is inspired by the work of Halliday (1994), Halliday and Hasan (1976), Kress and van Leeuwen (2006), Matthiessen (2007), O?Toole (1994), and Nodelman (1988), which I will discuss in detail in Chapters Three and Four. Linguistics Professor Emerita Ruqaiya Hasan?s (1985) belief that all literary forms ought to be examined through the lens of a robust theory of language is an important influence on my study. She posits that investigating the construction of meaning in a particular semiotic system requires a systematic theoretical framework that describes its resources and their meaning-making functions. Her view is the inspiration for my creating and using that collection of meaning-making systems for my teaching, my academic work, and my illustration.   The cohesive resource system as I outlined is a multimodal system that integrates the elements and resources of the other five systems, particularly the verbal and visual systems. My cohesive resource system is inspired by and based on Halliday and Hasan?s (1976) work in their book Cohesion in English.   The application of a social semiotic lens to consider the social practices related to the picturebook is another important strategy vital to the critical editing process. My illustration work does not begin with a specific social practice in mind, but once I have roughed out a picturebook or any other kind of illustrated text, I edit it with a view to its probable use and audience. I do this to be sure that its cohesive dynamics are accessible to its probable audience. To my academic analytic study of picturebooks, I bring the same body of accumulated knowledge that I bring to my own illustration work and my teaching, along with a persistent desire to deepen my understanding of how to construct a coherent narrative across the pages of a picturebook. However, inspired by the work of Halliday and Hasan and their consideration of the potential resources for constructing meaning, my approach is now systematic and brings a social semiotic perspective to my review of the social practices related to textual creation, production, distribution, and use.    4  My interest as an illustrator, teacher, and academic is in how meaning is constructed. Of course I am interested in the meaning construed in a picturebook and in the process of investigating how meaning is constructed there is a good deal of meaning construal that takes place. I make this distinction between how meaning is constructed and how meaning is understood or construed because as an illustrator my interest is in how to create meaning in the picturebook. This is different from, but not unrelated to, an interest in what meaning is conveyed. My intention is to increase academic knowledge about how complex relationships between semiotic modes and meaning-making systems construct meaning, and specifically how they construct cohesive, coherent multimodal picturebook texts. Thus, the focus of my study is on the construction of textual meaning, which is one of the three kinds of meaning making that will be described in the detailed discussion of SFL in Chapter Four. I mention this now to highlight my intention to analyze only the textual function-the function that deals with how a coherent text is constructed by cohesive resources. The equally important representational and interpersonal functions will also be defined in Chapter Four but the analysis of those functions is beyond the scope of this work.   One of the most important changes in my approach to illustration is the shift from a focus on requisite practical skills, such as the ability to repeatedly render characters and settings with consistency and continuity, to a systematic approach that considers the potential in each of the key meaning-making systems in an illustrated narrative. Also important to my illustration and academic study is the accumulation of knowledge about the affordances of various formats (Shulevitz, 1985, p. 15). This knowledge is critical to the selection of the best format for a particular narrative. Sorting out a format?s affordances requires a systematic approach that categorizes the layers of structures or networks of resources actively creating meaning in a text. The social semiotic perspective provides a systematic layered approach to reviewing the contexts of situations related to social practices within and outside the text (Halliday and Hasan, 1985, p. 45). The notion of text and how it is constructed relates closely to the notions of literary formats and genres (Hasan, 1985, p. 91).  5  1.3 Defining the Picturebook as a Unique Format I emphasize the value of distinguishing various formats within the family of illustrated books; therefore, I preface the overview of my study with descriptions and definitions of some related key terms.  First is the term picturebook, a term broadly used, conflated, and frequently applied to any illustrated book for children. The conflation of illustrated books and picturebooks often stems from a failure to ascertain the differences in an illustration?s purposes. Most of the time, determining whether or not the text can stand alone sorts out the issue. Conflation of the two notions also comes from the widespread use of the term picture book to describe all books for children containing pictures. This distinction is highlighted by academics through the use of the compound word picturebook (Lewis, 2001; Sipes, 1998, 2009).  The picturebook is a particular format in the family of illustrated books. It is now widely recognized as a 20th century literary art form. It is a complex, ever-evolving format in which words and pictures create meaning that neither mode could convey on its own. The picturebook is not always intended for children. As a literary art format, it is now used for narratives and content for teens and adults. The focus of my study is on the narrative picturebook format for young children, two to five years of age. The reason I make a clear distinction between formats is that it is the nature of different formats to possess different affordances and to be associated with different reading conventions and social practices.  Over the past 60 years as the picturebook has gained recognition as a literary artifact, it has been and is frequently described as a literary art form. However, to avoid confusion over the use of the term form, I use format because it is a term often associated with the design and production of printed materials. The term form comes into frequent use in describing textual and visual forms and systems. Also, to reduce further possible confusion I use the term format because it describes the picturebook as a physical entity that is designed and assembled.  There is precedence for using the term format to describe the layout of printed text (also known as the typographic rendering of written language). The picturebook text is a 6  multimodal text, so to describe it as a format fits well with a definition that deals with the layout and composition of its multimodal text of language and image. Here I want to note that throughout this work I use the term composition in place of the term design.   While I, and others, consider the picturebook a format, some academics regard it as a literary genre. I argue that the picturebook is a format that can realize a naturalistic narrative genre (or any other) among many literary narrative genres. While the literary narrative genre determines the conventions of narrative, participants, elements, plot structure, and length within a particular format the picturebook format is defined by having its meaning conveyed by the collaborative and integrative work of image and written text. While there are some conventional lengths, particularly 32 pages, the picturebook may possess a variety of physical properties, including its number of pages. Its materiality may vary; that is, its paratextual features such as the book jacket, cover, endpapers, colophons, half title page, and other particular features that offer potential resources for expressing the narrative.  Sometimes the material format is determined by technological constraints and matters related to the production of the printed and bound picturebook. Ultimately, the important thing about the picturebook format is that within its covers words and pictures work together in a variety of complex ways to tell and show a narrative (Bader, 1976, p.1). As a point of clarification my research discusses narrative picturebooks in which both words and pictures convey narrative features. My discussion will later focus on visual resources and dynamics that may be found in narrative wordless picturebooks. However, I want to acknowledge that the wordless picturebook is a particular format that is outside the scope of this study because I choose to use the word/picture definition of the picturebook as articulated by Bader (1976).    1.4 The Differences Between Picturebooks and Illustrated Books The picturebook?s material format affords the telling and showing of the narrative with what was famously described by Barbara Bader as ?the drama of the turning of the page? (Bader, 1976, p. 1). The turning of the pages allows forward and backward reading, which until very recently was unlike the viewing of the sequential narrative in a film or drama, since the reader/viewer can pace the experience and move ahead or go back as frequently and quickly 7  or slowly as desired. Other formats in the family of illustrated books offer similar affordances, but a key distinction between picturebooks and other illustrated book formats is that picturebooks? collaboratively constructed meaning across the visual and verbal modalities forms a cohesive, coherent whole.  An illustrated storybook may contain fine and important visuals that support, enhance, and expand upon the text, but it is not usually conceived and constructed as a text in which verbal and visual modalities depend upon each other to form a whole. In an illustrated book there may be a narrative that is complete without images, as is often the case in illustrated fairy and folk tales. The illustrations add important renderings of particular times, places, and situations but they are not essential additions to the narrative verbal text. Another example is an illustrated information book in which there is a collection of texts and images on a particular topic. Of course there may be visual cohesiveness to the composition of the material, but the pictures, charts, descriptions, labels, captions, and other paratextual features, while conveyed with a general visual cohesion, are usually construed as a collection of related texts rather than a singular text. It is the difference between the general visual cohesion that relates a series of separate texts and a singular text that is visually and verbally constructed as a cohesive whole. The difference could also be likened to the difference between a complex multimodal painting made up of painted, drawn, or collaged bits of images, and text intended to be considered as a whole and the creation of an exhibit or display of various multimodal bits of information selected based on a subject or theme. The painting is meant to be a discrete stand-alone artifact while the display or exhibit is intended to convey a collection of information bits and artifacts, arranged, and presented in a generally cohesive manner related to the subject and some general visual display design dynamics. Interestingly, exhibits and displays often use narrative structures to convey a cohesive whole and to engage and persuade viewers to keep looking for what comes next. However, in most of these cases, the narrative is being used as a device to give the whole exhibit a sense of general cohesion in the way that selections of typographic fonts and layout templates may convey general cohesion in an illustrated storybook (Lambert, 2009).  Another way to look at the differences between the picturebook format and an illustrated storybook is to look at the illustrated storybook as one in which the storyteller/narrator or 8  descriptive narration has a dominant verbal presence. In a picturebook, the visuals are not necessarily dominant but they are not sufficient without the deployment of some verbal resources such as the expression of spoken words, thoughts, and descriptions of critical changes in time. In the picturebook, the visuals replace the narrator?s task of describing the setting, characters, gestures, and actions. Also, the visuals can replace the tone of the language needed to set the mood in a highly verbalized narrative.  1.5 The Multimodality of the Picturebook During its evolution, the picturebook has adopted collaborative word and image relationships, often simultaneously telling and showing the narrative. The picturebook?s multimodal nature is one of the prime reasons it has been a favoured format for conveying narrative stories to young children. As Matthiessen (2007) noted,  acknowledging Halliday?s 1975 book, Learning to Mean: Explorations in the Development of Language, young children?s early communication is multimodal  Young children actively read facial expressions and body gestures and scan the environment for variations. The picturebook?s multimodal format with its integration of visual and verbal modes is particularly accessible to young children.    The format?s multimodality makes picturebooks accessible to a variety of older readers who are new to English and who need clear visual contextual clues to meaning to help them make their way through printed texts. Now too, the picturebook format, having so long been used primarily with young children in mind, is fast becoming a format chosen for narratives on sophisticated themes for teen, young adult, and adult readers (Beckett, 2012). In the picturebook format, the verbal mode is rendered visually on the page by typography and/or a variety of printed marks. These printed marks draw on the resources of the system of typography. Interestingly, typography and drawing have related roots because written verbal text evolved from drawn symbols and from social needs to maintain records (Anderson, 1969). In Chapter Two, I discuss the origins of the picturebook format in greater detail. The roots of the picturebook format and of written language begin with the first drawings by humans. Early drawings in conjunction with social changes eventually led to innovations 9  related to the development of written language. Historically, words and pictures have had a long and complex relationship. 1.6 Genre and Register in Picturebooks In relation to picturebooks, I use the term genre as a literary term to address variations in types of narratives and I use the term register in reference to variations in kinds of language used in particular narrative contexts and in reference to different kinds of language use associated with different contexts of socio/cultural activity (Halliday, 1978). Therefore, I see the picturebook format used for a range of genres from mysteries to historical biography. Many picturebooks for young children are written as naturalistic narratives using the homey language registers familiar to young children even though the stories may involve elements of fantasy and may use animals to stand in for young children. Registers vary not only in the kinds of language associated with different activities and doings, but also with variations in the socio/economic/cultural contexts of the narrative?s origins and with variations in the activities and settings of the characters and plot within the narrative.   While there is a consistent use of naturalistic narrative registers in books for very young children, it is not always the case in picturebooks for older readers. Because of the increase in sophisticated picturebooks for older readers, it is now possible to find unusual registers such as the street language of a homeless teen in Woolvs in the Sittee by Margaret Wild (2006). Readers of that book must be able to translate and make sense of misspellings because the text reflects the oral and written language of an unschooled homeless youth from a rough urban neighbourhood. 1.7 Defining Text Text is an essential notion and a term frequently used in my study. I use the term text to mean a message, communication, or narrative that is a cohesive, coherent whole. It may be the short text of a traffic sign, the long text of a novel, or the complex multimodal text of a picturebook.  10  Two other key terms, closely related to text are cohesion and coherence. What makes a text coherent is the deployment of cohesive resources (Halliday and Hasan, 1976). Cohesive resources involve relationships that connect elements in the text so that they all hang together as something coherent. Coherence is used here to mean something that makes sense because it expresses logical relationships. The terms text, cohesion, and coherence will be explicated in greater detail in Chapters Four and Five. 1.8 Outline of the Thesis Chapter Two begins with a review of the origins of the picturebook format and the social practices related to its creation, production, distribution, and use.  Reviewing the origins of this complex form reveals the important relationship between formats and their affordances, and how technological advances and sociocultural changes promoted evolutionary changes in social practices involving the creation, production, distribution, and use of picturebooks. The review of origins provides a background for understanding the multiple semiotic contexts of situation associated with this format. A review of origins is essential to a social semiotic study because the format?s evolution in form and use is directly and inextricably related to its social contexts.    In Chapter Three I review important scholarly works on the construction of meaning in the picturebook.  They are organized according to their key contributions to the literature. I conclude Chapter Three with a description of the social semiotic approach I use to address a critical gap in the scholarly work that became my research question. I began my studies interested in how words and pictures work together to construct meaning in a narrative picturebook. My review of studies on the subject and the limitations of a dissertation helped me narrow my cumbersome question to a focussed investigation on how the cohesive resource of colour and the dynamic of repetition construct a coherent picturebook. My application of a social semiotic lens comes from my belief that children?s literature and, in this case, picturebooks, are social artifacts that reflect their sociocultural context of creation and use (Alderson, 1986; Darton, 1932; Egoff, 1967; Nodelman, 1988; Schwarcz, 1982).  11  Two critical notions, not previously mentioned, are borrowed from Halliday?s Systemic Functional Linguistics (SFL) and make significant contributions to the systematic approach of this study and to my illustration work, academic study, and teaching: semiotic metafunctions and the stratified model of how semiotic systems work. These notions provide a framework for looking at a modality as a ?resource for making meaning? with each ?system in the network representing choice? (Halliday, 1994, p. xxvi). In Chapter Four I discuss these notions in further detail, explicate my key theoretical frameworks and Halliday?s SFL, followed by a review of several SFL-inspired theoretical frameworks that demonstrate the viability of adapting and applying key notions of SFL to the visual modality. In Chapter Four I also discuss cohesive resources in detail.  In Chapter Five I discuss my methodology, my rationale for the selection of two picturebooks, Where the Wild Things Are (Sendak, 1963) and How to Heal a Broken Wing (Graham, 2008), and explain the visual configurations of colour and repetition that I use for the close analysis of the two texts. Chapter Six describes the close analysis of each of the picturebooks. In Chapter Seven I discuss the findings of my analyses and suggest future analytic work. I envision using my findings to enhance my teaching of illustration and illustrated children?s literature in support of the education of future teachers, librarians, writers, and illustrators. 1.9 Significance of This Study As education deals with the growing importance of multimodal instruction, there is a great need for providing educators with resources, language, and explications of the key notions that support multimodality. I believe that understanding how multimodal meaning is constructed is fundamental to assessing multimodal materials and student responses across the curriculum. As we move into an increasingly participatory digital age, educators need foundational knowledge with which to provide multimodal instruction and to critically assess students? multimodal classroom work (Jenkins, 2007). Also, it is social semiotic knowledge that will help educators and librarians recognize and select the best resources with the greatest potential for achieving the purpose or intended meaning of a message in a specific social context for a particular audience. I hope my work on textual meaning construction in 12  picturebooks contributes to the resources available to educators and illustrators to deepen their knowledge and application of the semiotic potentials of both language and image. I believe my work will serve as resources for multimodal response assessment and understanding because the picturebook is an ideal format for demonstrating key notions of multi-semiotic text construction and because the picturebook is highly accessible and not dependent upon technology. Although it is low tech, the picturebook remains a highly sophisticated multimodal format that supports literacy and literary development and has the potential to provide vital foundational learning that immediately applies to digital multimodal construction and construal.  13  Chapter 2: A Brief History of the Picturebook Format 2.1 Introduction This chapter presents a historical review of the origins and evolution of the picturebook format and its social practices. I review the history of the picturebook to argue that it cannot be defined by its textual construction alone. As a literary artifact the picturebook evolved in response to a complex succession of technical innovations and a range of social practices related to its use. This historical review provides the background and context for understanding the continuing evolution of this format and its other relatives in the family of sequential visual multimodal narratives. In this review I aim to capture key events and innovations and their relationships to the social practices of creation, publication, distribution, and use of picturebooks and their antecedents.  First, I will trace the evolution of the picturebook format by reviewing technological innovations, social practices, and the role of key authors, illustrators, and designers in shaping it. The first part of the review highlights the technological innovations and related variations in social practices associated with visual narratives that materially and rhetorically change visual narrative formats and what they afford. Changes in format, whether additions or deletions to its affordances, influence the format?s resource potential for constructing meaning. Technological innovations in the past two centuries afforded numerous creative innovations in the picturebook format by individual author-illustrators.  In the second part of this chapter, I review the particular innovations that contributed to the evolution of the format. Those innovations had origins in social contexts and reflected the fact that individuals were allowed and encouraged to creatively use new technologies. Although the picturebook format is ultimately a collaborative art form, it affords individual creators potential to convey particular stylistic and expressive meanings. This review is intended to provide the contextual background essential to a social semiotic study. The importance of that accounting derives from the notion that semiotic systems are constructed of and by sociocultural circumstances and practices (Halliday, 1978). In the case of the evolving format of the narrative picturebook, a review of the contexts of its origins and 14  social uses addresses three key form-defining factors. They are (a) the evolution of a technology that afforded changes in format, (b) the evolution of social practices in relation to the growth of the middle class with consequent new views of children, childhood, and education; and (c) an increase in the illustrator?s and designer?s role in the creation and innovation of this format. 2.2 The Origins of the Picturebook 2.2.1 Early Visual Narrative Arts Due to its narrative content, the origins of the narrative picturebook stem more directly from the traditions and conventions of narrative painting than from those of the illustrated book (Lewis, 1996). This is because in narrative paintings, as well as in other narrative visual arts, the important visual dynamic is derived from the fact that the images show much of the story or relate directly to narratives known by the viewers. Tracing the origins of the illustrated book illuminates the role of illustration as primarily decorative, however elaborately and lavishly rendered.  So while the origins of the picturebook go back to the earliest illustrated books, the development of narrative arts accessible to the masses creates a branch in the origin tree. This is because the role of images in the narrative arts and picturebooks is to show a story. The narrative picturebook shares common ground with other narrative arts including tapestries such as the Bayeux Tapestry (1070), stained glass windows (500), the chalk drawings in Danish churches (1100s), illustrated chap books (1600s) and early comics from the 1800s (Alderson, 1986; Lewis, 1996).  While the picturebook?s origins of visual narrative conventions and rhetoric go back to sequential visual and narrative arts, its material origins of form, design, typography, and layout reach back into the development of written language and the evolution of the illustrated book. There is, therefore, an interweaving of roots and offshoots in the history of the picturebook.  To start the review, I look at the connection between picturebooks and the traditions of visual narratives. I want to acknowledge that this review deals solely with origins of visual narratives in Western European contexts. As previously noted, Alderson (1986) and Lewis (1996), in looking at the origins of the narrative picturebook, see the most significant direct 15  connection to narrative painting and its traditions of painting stories and history to convey content to a mass audience (Alderson, 1986). Narrative painting is intertextual, requiring an understanding of the story and its origins to make sense of it. In narrative paintings of the past, the construal of meaning depended upon the viewer?s knowledge of history, myth, and the bible. A similar expectation is associated with narrative stained glass windows. For example in the late twelfth century stained glass art in the Notre Dame cathedral in Chartres depicts stories from the Old and New Testament, lives of the Saints, the zodiac, and scenes associated with the four seasons. In the case of the Bayeaux Tapestry (1070), the depiction of 70 sequential scenes of the Norman Conquest with Latin captions not only visually represents the story, it also captions activities in much the same way captions support comics and titles label narrative paintings. Churches typically allow the general public access to the stained glass windows, tapestries, and narrative paintings, which in turn exposes viewers to visual narratives. In medieval times, there were also roving puppet shows and traveling drama troupes providing visual sequential narratives for the general public. These visual sequential narrative forms were multimodal and used the resources of facial expressions, body gestures, sounds, music, colour, and costume to convey meaning. The literate elite also enjoyed multimodal visual narratives in plays, dances, and opera. For a time prior to the invention and use of the printing press, handwritten books, illuminated manuscripts, and narrative paintings flourished for a small, literate segment of the population. This snapshot?admittedly a sweeping generalization of the origins of multimodal narratives in Western civilization?makes the point that, in reviewing the history of sequential visual narratives, their use goes back to the beginning of human life, and their existence illustrates longstanding human interest in sequential, visual, multimodal narrative forms?an interest occurring in all socio-economic groups across cultures.  2.2.2 Early Books The origins of written language and the illustrated book go back to handwritten books, manuscripts, scrolls, and the early handmade books of the 15th century with moveable parts and flaps predating the invention of the printing press. Most of the early handmade books 16  with moveable parts were the work of astronomers with the exception of a few made by children (Reid-Walsh, 2012). In terms of connections between words and images, it is interesting to note that the beginnings of written language and illustration share roots in early drawings and pictographs. Moreover, from a social semiotic and sociological perspective of the evolution of social practices, the development of written language is closely related to the development of ownership and the accumulation of authority and wealth (Anderson, 1969).   The moveable type press afforded mass printing, mass communication, and stimulated social and political changes that influenced religious education and inevitably secular education As such, written language first limited to the literate elite extended into other classes with the spread of printing presses. Printers and printing presses brought about political and religious changes and promoted broader literacy?literacy defined as the ability to read written language. In addition to its influence on social practices, the printing press initiated a long period during which written or printed language was privileged over visual images. It wasn?t until the use of lithography in the 19th century that the relationship between print and illustration began to change. While some books included woodcut images and later wood engraved images, the images were not essential to the narrative. Indeed, it is in this period of history that the printed book evolved and conventionalized into a predominantly written text format. However, as will be noted later in this chapter, at that time there were also crude, mass-printed, illustrated materials (the early pulp or pop paper ephemera) available to the general illiterate public. As such, while printed literary books privileged written text, printing for the masses privileged images.  The information age is a multimodal age and currently, the privileging of written text is fading in the face of the visual nature of most contemporary communication tools and practices. In children?s literature there is a reflection of this trend in the proliferation of illustrated novels, chapter books, graphic novels, and wordless books. Their popularity and preponderance also reflects the growing market for second language acquisition materials and the increasing use of picturebooks with spare verbal texts and wordless books in classrooms for all ages (Arizpe, 2009).   17  2.2.3 Early Books Created for Children   F. J. Harvey Darton?s history of English children?s books is titled, Children?s Books in England: Five Centuries of Social Life (1966). This title highlights the widely acknowledged view that children?s books express their social origins and contexts of use. Since the publication of Darton?s book there have been many changes in the forms and numbers of children?s book published, however it is still true that children?s books reflect children?s social lives and challenges. As Darton wrote in his preface to the first edition:  There is really only one ?text? in these pages, and that is, that children?s books were always the scene of a battle between instruction and amusement, between restraint and freedom, between hesitant morality and spontaneous happiness. (1932, p. vii) This battle began with the first books written specifically for children in the 17th century by religious groups such as the Puritans. Their goal was to educate their young to read the Bible and thus be saved and ready to die a good death. The Puritans, among others, acknowledged that children are drawn to reading if images are included, so they created illustrated books for young learners. Most likely the notion to include images was inspired by popular sequential visual narrative forms to entertain, as visual action more easily engages audiences of all ages and levels of literacy.  Darton states that by ?children?s books? he means works published for children to give them pleasure rather than to educate them (1966). His history, based on his definition of children?s books, thus excluded educational books. Darton made a clear distinction between narrative fiction and pedagogical or religious work written specifically for children. He noted that the first non-instructional books for children were printed in 1744 by John Newbery. Prior to that, children read books published for adults including Robinson Crusoe, Pilgrims Progress, and Gulliver?s Travels. However, in passing it is useful to note that, in looking at the origins of illustrated books for children there are schoolbooks, alphabet books, and religious and moral tales to consider. These illustrated educational and religious texts are relevant from the standpoint that they mark the beginning of the publication of books with images for children. 18  The history of illustrated books created for children begins with what is often cited as the first picture book for children, Orbis Pictus, published in 1658. However, this first picture book is not a picturebook according to the format?s contemporary definition, rather it is an illustrated book of the alphabet. While it is related in the sense of being a book for children, it does not possess the narrative link to the picturebook format that evolved in the 19th and 20th centuries, which is the focus of this thesis. 2.2.4 The Comic Divide  With the affordances of mass printing, a variety of printed narrative formats emerged. The narrative picturebook format derived from a variety of early popular forms that David Lewis (1996) considers part of a relatively recent ?prehistory of the picture book, chapbooks, caricature and comic strips, and novelty/toy books and theatres? (1996, p. 5). These formats along with printing technology continued to evolve through the twentieth century. Until the 1800s, most of the images in the early mass printed formats were produced by woodcuts. Comics, chapbooks, and picturebook formats share a link to early woodcut broadsheets. Broadsheets were popular illustrated souvenirs sold to huge crowds gathered at events such as executions in the 1600s. On these printed sheets, words are positioned next to images of speakers, an early version of speech balloons. Indeed, the practice of positioning speech next to a speaker without a formal speech balloon continues in contemporary graphic novels.  Broadsheets, with their lurid content, mark the divide between pop and literary literature. This divide continues to the present. Early in the 20th century this divide and the prejudice of an educated class who devalued comics, impeded what may have been a speedier evolution of the picturebook format. While the public remains cautious about the value of comics, librarians and classroom teacher now embrace the literary genres of comics, graphic novels, and modern chapbooks. The evidence for this is in the recent proliferation of books and articles on the merits and ways of using graphic novels in the classroom and library. Recognition of the values and uses of graphic novels, particularly literary and curriculum-related graphic novels, has generated a demand that has been met by widespread publication of graphic novels for readers across ages and school curricula (Frey and Fisher, 2008). 19  Broadsheets eventually evolved into chapbooks. Early chapbooks from the 1600s are small, crudely produced, folded sheets of paper, sometimes stitched into tiny books. The printing limitations of their time and the use of woodcuts did not afford fine detail. Furthermore, the printing of multiple copies tended to wear down the woodcuts yielding even cruder printed images. Chapbooks emerged as a pop culture form in the 1700s. As their images were often used decoratively rather than illustratively, the woodcuts were used over and over again in many different chapbooks with little regard to how closely the image related to the printed word (Sabin, 1996). Early chapbook content varied from practical information guides to fanciful stories for both children and adults accessible to the poor and barely literate . Today they remain popular and are produced in the underground comic world by poets and spoken word performers. These modern chapbooks may use woodcuts, but for the most part, their images are drawn or painted and then scanned and printed using modern photocopying technologies.      In terms of social practices, broadsheets, chapbooks, and early comics were mass-produced for a predominantly semiliterate audience. They were considered popular culture products, whereas printed books, including illustrated books, were considered literature and primarily published for a literate elite. Mass printed newspapers and magazines proliferated in the 19th century following innovations in printing, the industrial revolution, and the growth of the middleclass. Eventually these mass-produced publications crossed and blurred the boundaries between popular culture and literature as more people learned to read. Major innovations in printing images, such as the shift from woodcuts to printing from wood engravings in the late 1700s, followed by the invention of lithography in the 1800s, and then photolithography in the late 1800s, afforded mass printing with the resources to print better images in greater numbers, which in turn generated more publications. Advancements in printing images afforded formats such as the chapbook and magazines, the opportunity to visually depict items, characters, and situations in more precise detail and nuance than was previously possible. While the first shift from woodcuts to wood engraved images allowed for the reproduction of a finer, potentially livelier line, the invention of photolithography afforded even more detail and more accurate reproduction. This enabled the mass production of line-based art such cartoons, caricatures, and highly detailed illustrations in the magazines 20  and newspapers of the 19th century (Sabin, 1996). Simultaneously, the growth of newspaper publishing initiated the new narrative development of serialized stories by Charles Dickens and other literary writers of the time. Serialized stories in newspapers and magazines began to bridge the worlds of popular culture and literature. Indeed, some books first appeared in newspapers and magazines. Social and industrial changes produced a larger reading population who not only wanted to read the next chapter in the newspaper; they were interested in reading other illustrated materials and books. 2.2.5 The Nineteenth Century?s Middle Class and the Increase in the Number of Literate Readers and Illustrators In the 19th century, a growing middle class brought about significant social changes that influenced social practices related to the creation of children?s literature and inevitably generated some of the innovations that produced picturebooks. A new population of readers supported the market for secular content in books, magazines, and newspapers for adults and children. A playful literature emerged from this literate class that had access to books and other print materials, opportunities to learn to read and draw, and the leisure time for play and the development of artistic and literary skills (Garvey, 2013). An increase in literacy and the proliferation of newspapers and magazines created opportunities for illustrators. David Lewis (1996) also notes that in the 19th century many early children?s book illustrators were also caricaturists and comic strip creators. The close connection between children?s book creators and cartoonist/animators continues to this day. Caricature artists? abilities to deploy line to efficiently and sparely render character, gesture, and action are essential for visual storytelling in cartoons and comics as well as for rendering the visual narrative in picturebooks. The sequential narrative activity in comics and cartoons that relies on word/picture collaboration is another early connection between these forms and the modern narrative picturebook. Chapbooks and the early magazines of the 1800s contained serialized stories that generated illustration work as a narrative mode.  As noted previously, the early history of the social practice involving illustrated books for children was primarily related to their use in religious educational settings and in private 21  educational settings for the elite. In contrast, the working classes and their children had broadsheets, chapbooks, and the fiction magazines known as ?penny dreadfuls? which showed stories in pictures. There is a further interesting contrast between mass-produced items and earlier illustrated books for the elite. Books for the elite contained spot illustrations added for decorative interest that rarely depicted essential visual narrative information, whereas mass-produced materials tended to use illustration to carry a good portion of the narrative. Many of the first printed books appear to use small images for the purposes of visual embellishment. However, sometimes they visually depict a relationship to the text as in the case of alphabet books and other catalogues in which a picture of an apple appears with the word apple or a verse about an apple.  2.2.6 The Invention of Photography and Photolithography Initiate Developments in Narrative Art Formats The combination of a growing middle class, the invention of photography and photolithography, and the growing number of skilled writers and illustrators, initiated the development of the narrative picturebook format and accelerated the evolution of three other narrative art formats that share common ground with the picturebook: cinema, animated films, and comics/graphic novels. Nineteenth century technological changes enabled an even greater volume of mass printing and extensive use of colour and halftone illustration. This, in turn enabled more detailed, nuanced, and accurate colour reproduction of illustrations in books. Technological innovations contributed to critical changes in the way books were produced by printers and illustrators. A significant change was that photolithography allowed the direct reproduction of the artist/illustrators? drawings (Alderson, 1987). Previously, most printed illustrations had to be copied from artists? drawings onto plates or made into wood engravings by highly skilled craftsmen.  While most craftsmen created amazing facsimiles of the original work, they were nonetheless recreations that rarely possessed the freshness of the original. This change in the practice of the creation and reproduction of illustration altered the role of the illustrator, allowing the illustrator more artistic influence over the production of illustrated books. 22  Once the technology allowed the artist/illustrator greater control over the illustrations, illustrators achieved more power and recognition (Alderson, 1987). This power gave picturebook creators opportunities to innovate. Another important point is that the invention of photography afforded a new way of documenting the physical world that provided writers and illustrators with visual information and knowledge often not previously available to them. A case in point that I will revisit later is the influence of the photographic work of Eadweard Muybridge (1872) on drawing and illustration. Muybridge photographed humans and animals in motion in extensive series of still shots that when viewed sequentially revealed the precise movement of bodies in motion, details next to impossible to see unaided by photography. Photography and its influence on printing techniques changed the role of the image in illustration and in scientific enquiry because a photograph or a photographic lithograph could be repeated. Photography could capture and report on things invisible to the eye, and that report could be exactly replicated (Ivins, 1953). The opportunities presented by photography to a more expressive use of drawing beyond the representational reporting role it had had. While photography could report, drawing could express, abstract, and simplify.  As outlined above, both the evolution of the picturebook and the role of its creators as innovators have complex histories of social practice. Not only did innovations in printing technology contribute to the picturebook?s evolution, there were also larger, broader innovations in all technologies during the industrial revolution, which precipitated dramatic shifts and changes in socioeconomic structures, all of which eventually influenced the picturebook, its creators, publishers, and readers. 2.2.7 Playful Leisure Time Activities and Shifts in the Purposes and Formats of Illustrated Children?s Books The great change in 19th century children?s literature was that books for children were no longer primarily instructional; they contained entertaining, playful, and puzzling narratives (Darton, 1966). These books, as is the case with all books, reflected evolving social contexts. The context at that time among the middle class was one of narrative playfulness with both 23  language and image. That narrative playfulness is reflected in the depiction of a range of social activities such as scrap-booking, making homemade books, playing with toy theatres, board games, and enacting charades, tableaus, and skits (Lewis, 1996). Throughout the history of children?s literature, the key forces for change and innovation are rooted in social practices related to the creation, production, distribution, and use of books (Bader, 1976). Of course, as noted earlier, many of those changes are directly related to innovations in printing technology. This is clearly evident historically, in the flourishing of children?s literature that followed the industrial revolution and the growth of the middle class. While technology afforded mass printing, the expansion of industrialization produced a market that had the time and money for leisure activities. This coincides with significant increases in the market for children?s books (Darton, 1966). Furthermore, the growth of the middle class influenced changes in notions about children, childhood, and education. For one thing, children no longer had to work to financially support their families, so they had time for education and recreational activities. They had time to be playful. The growth of leisure time for children and a playful approach to literature lead to a wider range of genres?adventure stories, fantasies, nonsense, and rhyme?entering the nursery for the purpose of entertainment. While a good deal of instructional literature for children was still being published an increasing number of these more entertaining narratives were being published as well. Moreover, as photolithography evolved, books for children included increasing numbers of illustrations (Alderson, 1987). The growing middle class and the increase in playful leisure activities influenced three major changes in the role and use of illustrated books, which in turn contributed to the evolution of the picturebook format. First, there was the proliferation of toy books and novelties with interactive features; second, there was a significant increase in narrative pastimes and games; and third, there was a shift in the purpose of illustration (Lewis, 1996).   24  Printed toy books,1 toy theatres, and novelties contributed directly to the evolution of the picturebook in the 1800s because they contained entertaining interactive physical features (Lewis, 1996). These items flourished because printing technology allowed for the use of more colour and more illustration, hence, more resources for making meaning. During this time, mass-produced novelty books featured engineered pop-ups, flaps, and pockets. Fifteenth century book creators used paper engineering to illustrate science books; however, it was not until printing and book binding technologies evolved that novelty toy books could be mass produced. There are a few examples of printed flap books from the 1700s, but they were unusual and not accessible to most of the public (Montanaro, n.d.). The fanciful Victorian novelty books foreshadow a shift from the illustrated book to the picturebook form because their gaps and multimodal meaning making encourage reader interaction (Lewis, 1996).  The new abundance of printed materials supported the development of pastimes that contributed to the evolution of the picturebook. Mass printing enabled the proliferation of newspapers that included evolving narrative forms, comics, caricatures, and the enormously popular serialized stories. This abundance of printed forms promoted another popular narrative activity, scrapbooking, the creation of family picturebooks or memory books and the archiving of the abundance of print information (Alderson, 1986; Garvey, 2013). The Victorian love of collage and decoupage emerged from and supported the mass publication of ?pretty? images and die-cut items. Hans Christian Andersen, in addition to creating narrative papercuts as he told stories, illustrated homemade books and created elaborate collaged screens and hand drawn storybooks for friends (Brust, 1994). The connection to the evolution of the narrative picturebook can be seen in specific examples of writers and illustrators who participated in these popular leisure time narrative activities, both as children and then as adults, thus giving them hands-on experience composing integrated word and                                                  1 While books as toys are not directly relevant to my study, it is interesting to note that the elements of their form that inspired interactive handling directly relate to recent innovations in picturebook apps. According to Nugent, this is because most of their key dynamics such as actions that use sliders, rotators, and pop-ups are the same as those of 19th century novelty books (as cited in Technology Review, 2013). 25  image pages (Alderson, 1986). As Ellen Gruber Garvey (2013) writes, ?Scrapbooks allowed all types of readers to ?write? a book with scissors. Children were even encouraged to store up barely understood writings in their scrapbooks to enjoy when they could read them better? (p. 27). Some of the most widely acknowledged classic children?s books began as small homemade picturebooks such as Lewis Carroll?s Alice?s Adventures Under Ground, Edward Lear?s The Book of Nonsense, Beatrix Potter?s The Tale of Peter Rabbit, and William Nicolson?s two great innovative books, Clever Bill and The Pirate Twins. As well, other illustrators such as Richard Doyle and Walter Crane created small books for family and friends (Alderson, 1986).  The question of how narrative picturebooks, even amusing novelty ones, engage and instruct readers relates to my interest in how multimodal texts are constructed. The pendulum of dominant themes and goals in children?s books continues to swing back and forth from instruction to amusement and has done so since the publication of Orbis Pictus, the first illustrated book for children. Since the construction of written and visual texts is directly related to their purposes and since those usually reflect social practice, an illustrator?s conceptions of a text?s purpose are bound to be shaped by its verbal content and by the illustrator?s own social contexts and reading experiences. Therefore, it is not surprising that illustrators experiencing narrative playfulness in their leisure time tended to express that playfulness in their illustrations.  These social changes in the middle class also shift the purpose of pictures in early children?s books, from essentially a decorative role as enticements to read to the role of providing key visual narrative information (Alderson, 1987). The market for lavishly illustrated narrative works increased and influenced a further shift away from the use of illustration to decorate to its use in the service of content and tone (Alderson, 1987). To be clear, this was not yet a shift to illustration as a major narrative force, but it was part of the evolution of the illustrated book. In the mid nineteenth century, new printing options supported more precise image reproduction and the production of more highly detailed illustrations to represent visual 26  information. These advances became resources for all forms of books and particularly books intended to instruct and inspire scientific inquiry (Ivins, 1953). Accurate colour printing using a photographic process and opportunities to include illustrations with high detail and colour nuance served not only scientific and educational purposes but also narrative story telling. This is because nuanced colour, detail, and quantity of illustration afford the visual modality more ways of conveying descriptive narrative content in images. However, despite advances in colour reproduction, it is the line illustrations of Randolph Caldecott that foreshadow the picturebook because they vividly render narrative activity and events rather than descriptive details.  The opportunity for illustrators to draw directly for the page and to have it reproduced in a way that retains and precisely repeats the freshness of the original line supported the development of sequential visual narratives (Alderson, 1987). Mid eighteenth century books continued to use most of the traditional form features of the illustrated book, but it was the liveliness of their illustration, particularly in the line work related to the art of caricature, that signalled the coming emergence of the narrative picturebook.   2.2.8 Influential Illustrators and Creators  Randolph Caldecott can be considered the first picturebook illustrator because his drawings play an active and prominent role in his narrative books (Alderson, 1986; Lewis, 1996). Caldecott?s coloured line drawings and his composition took full advantage of the technological affordances. His illustrations carried more of the narrative story than had been seen before in illustrated books for children (Alderson, 1987). While his works were in the typical format of the time, the liveliness of his line, and the dominance of the illustration over the design format pushed his work toward what became the classic picturebook associated with Maurice Sendak.  During Caldecott?s time (1846-1886) lithography dramatically changed printing. The innovations in printing technologies went hand in hand with the development of photography and cinema. In fact, there is evidence that those developments influenced and informed 27  Caldecott?s work.2 It is a safe conclusion that photography and cinema influenced many other narrative artists as well. The increased power afforded to the illustrator/creator of books is one of the most important changes in the social practice associated with the creation and production of illustrated books in the mid-19th century. Illustrators? increased control over their artwork   along with printing advances ushered in what has been called a ?golden age? of illustration. Children?s books of that time were lavishly and beautifully illustrated by creators whose work continues to inspire contemporary illustrators (Saltman, 1985). The 19th and early 20th century middle class lifestyles in Europe and North America nurtured many of the best known and most influential English and American children?s illustrators of all time: Randolph Caldecott, Kate Greenaway, Walter Crane, Kaj Nielsen, Maxfield Parrish, Arthur Rackham, Jessie Wilcox Smith, N.C. Wyeth, Beatrix Potter, and many more. Most of their illustrated books are collections of stories or novels in which illustrations are ?tipped? in every 20 to 50 pages. Walter Crane is known for his elaborate book design, particularly his composite design of illustrations and typographic elements on the page. Crane?s integration of image and typography continues to influence book designers and children?s picturebook illustrators. For the most part, the aforementioned artists of the golden age of illustration rendered complex and descriptive visual enhancements to the text. However, the illustrations were not integral or essential to the narrative content. These kinds of illustration often pull and even hold the reader/viewer in the visualization of the text?s world and away from the written narrative (Nodelman, 1988). In fact, as mentioned earlier, illustrations often enthral the reader/viewer in the moment portrayed. Most of the books of the golden age demonstrate an important                                                  2 Leonard Marcus in his upcoming book on Caldecott believes he can see the influence of Muybridges?s photography in Caldecott?s drawing of horses because the horses are drawn with all their hoofs above ground, a physical fact that is almost impossible to see with the naked eye but a fact that Muybridge documented when he took photos of horses in action. As well, Marcus has evidence that puts Caldecott near an exhibit of the famous photographer?s work (conversation with L. M., October 12, 2012). It could be argued that Caldecott, just like a brilliant hockey player such as Gretzky, possessed remarkable vision, so acute as to catch literal split second action. My point is that some people may possess the ability to see what average eyesight cannot. 28  difference in the purpose of illustration in illustrated books compared to picturebooks. In Illustrated books, the images are meant to captivate the reader?s attention in the captured moment of the narrative versus motivating the reader to turn the page to see what happens next in the picturebook. 2.3 Changes in the Purposes of Illustration There are many kinds of illustration and as many purposes that they serve. Starting with Caldecott, children?s books have seen a change in the purpose of illustrations from that of enhancing and elaborating the verbal descriptions of settings and characters to the use of illustration to convey story events in some of the ways that other narrative sequential visual arts use. This distinctness in the purpose of illustration is fundamental to the development of the picturebook. It is the difference between using illustration to tell the story and using illustration to convey descriptive information, such as what characters look like and where they live. For example, in most of his works, Rackham illustrated fairy tales and fantasies rendering pictures of what the characters looked like in their narrative settings. He created a visualization of the story?s world as an art director visualizes the visual world for a film or a play. Many talented children?s book creators also designed theatre, film, ballet, and opera, such as Kaj Nielsen, Maurice Sendak, Beni Montressor, Edward Gory, and more recently Mo Willems, Shaun Tan, and Jon Klassen. 2.4 The Illustrated Storybook and the Picturebook The main difference between illustration-as-enhancement and illustration-as-storytelling continues to be realized in two distinct children?s book formats, the illustrated storybook and the picturebook. There are many acclaimed contemporary illustrators such as Lizbeth Zwerger, Edward Gory, Grenady Spirin, and Chris van Allsburg who work as enhancers most of the time. Some of them use both methods depending upon the nature of the text.   Some ?switch-hitting? illustrators such as van Allsburg and Maurice Sendak can both descriptively illustrate some of their books and then visually sequentially narrate others. They capture or freeze moments in the narration for the reader/viewer to contemplate, and use the written text to move the story forward. Of course, sequential visual narrators also 29  need to create a world and its characters, but the difference is that the actions depicted in their illustrations actively propel the narrative forward. Sendak is the perfect example of an illustrator who, while considered the master of the narrative picturebook, also created stand-alone illustrations for many books, such as Lore Segal?s retellings of Grimms? tales.  Fundamental to sequential narrative forms is the role of illustration to actively contribute to narrative progression, while in some other illustrated formats the visual is meant to conceptualize or visualize the narrative environment and its characters. Visual sequential narrative storytelling usually requires more illustrations and it needs logically sequenced renderings of actions that are legible. This legibility depends to some extent on the viewer/reader?s ability to recognize what characters are doing and to predict almost unconsciously what characters do in the spaces/time between renderings. Since prediction requires experience, picturebooks for the youngest readers usually contain more illustrations with smaller gaps between the illustrated actions. In these kinds of nearly wordless books picturebooks, the verbal text fills the gap in some other way such as through the repetition of verbal elements that directly connect to elements in the images. Although the difference between many picturebooks and illustrated books may initially seem straightforward, there can be complex variations in each, which make the distinction less clear. There are some virtuoso illustrators, such as van Allsburg, Anthony Browne, and Sendak, whose work can hold the reader in a contemplative, reflective space with rich, complex, visually-rendered content. In these works, instead of using lively lines to create a sense of action and movement to draw the viewer?s attention, the illustrators use highly patterned compositions, complex orchestrations of contrasts, framings, symbolic elements, and positions of the narrative elements. In these complexly composed works the verbal text is meant to draw the reader/viewer forward through the story. In illustrated books there are complex relationships between visual and verbal elements in a text even though the images are not moving the story forward. However, they are relationships that add layers of meaning to the story. Books by Nancy Ekholm Burkert and Lizbeth Zwerger clearly exemplify the general purpose of enhancement. They both render a poetic storyworld and primarily rely on the written text to progress the narrative. However, 30  the nature of Ekholm Burkert?s poetic expression in her illustrations comes from highly detailed, incredibly precise, and complexly patterned renderings of specific cultural references related to the verbal text. On the other hand, Zwerger?s illustrations use minimal details rendered in simple strong portrayals of specific characters with almost no specific settings or backgrounds. Yet, they too rely on some cultural references to elements in the verbal text. Some of the works of van Allsburg, Browne, and Sendak display complex kinds of visual enhancements to the verbal texts that are similar to the kinds of relationships characteristic of picturebooks. This is partially because each modality specifies or limits the other (Nodelman, 1988). While there may be many similar relationships such as the particularizing and enhancing of the verbal description of a character through illustrations, the cohesive relationships as well as the narrative relationships may significantly differ in each format. The layers and complexity of the meaning making in these books underscores the importance of sorting out the kind of format at hand and the general purpose of the illustrations in order to do a close analysis of how each modality?s resources support the narrative descriptively or actively move the plot forward. The potential range of differences between and within formats is part of their complexity, their literariness, and what makes them such a rich resource for developing literary understanding. Until the 1920s, the typical illustrated children?s book format was a page of text, usually on the left, with an illustration on the right page of a double spread. Typical of this format is the still popular series by Beatrix Potter, of which The Tale of Peter Rabbit (1902) is best known. In these books realistic gentle watercolour illustrations depict story events. The verbal story can be read aloud without the benefit of its illustrations. In fact, it is often read aloud in audio format. However, even though the verbal story can be fully understood without illustrations, Potter?s delicate naturalistic depiction of a rabbit family sets an important tone. In the small printed book version of the story, the written text for each page, the pagination of the text, and the dynamic of the turning page work in combination to propel the narrative forward more than the actual illustrations. As noted by Nodelman (1988), this is the kind of illustration that often pulls the reader/viewer backward rather than forward.  31  Although Potter?s books are not narrative picturebooks by virtue of the fact that their stories stand alone, their popular use supported the development of several specific illustration conventions that became and remain typical of the narrative picturebook format for very young children. These key visual conventions, which young children naturalize and use for making meaning are the repeated depiction of the main character on every illustrated page, the use of small animals in place of small children, the interactive look of Peter Rabbit eyeing the reader, and the cohesive use of a limited colour palette throughout the book. These conventions were further nudged into children?s naturalization of them during the early 20th century by the widespread reading of comic strips in newspapers, the viewing of silent movies, and by children?s reading of comic books created especially for them (Lewis, 1996).  Interestingly, in terms of popular forms, the cultural divide between the books for children purchased and read by the middle class, and the visual narratives in comic strips, cartoons, animated films, and comic books, persisted throughout most of the 20th century. At the same time, the proliferation of the narrative arts of comics and film began to influence writers and illustrators of picturebooks (Alderson, 1987, Lewis, 1996).  2.5 Additional Key Influences Promoting and Hindering the Evolution of the Picturebook  A mentioned earlier, the picturebook format developed more directly from chapbooks and serial comics than from the illustrated book. Throughout the 20th century, the picturebook continued to evolve along with arts of the cinema and graphic novel. However, in spite of the fact that sequential narrative is the common denominator between these arts, there was a literary divide between trade books and the pop culture forms of chapbooks and serial comics. Initially the divide was primarily between visual sequential narratives, that is, between comics, cinema, cartoons, and animated films and illustrated books. A major reason for the persistence of this literary divide was the power of the ?book women.? The book women were a group of trained librarians who were highly vocal in their views on children?s literature (Marcus, 2008). To their credit, they were also highly influential in the establishment of library rooms and departments for children and in the promotion of the value of literature in young children?s lives. However, the divide between comics and books 32  these book women perpetuated delayed early innovations in the picturebook format from taking hold, as they had the power to influence book publishing and book sales. They critically dismissed some of the first books that began to employ sequential narrative conventions (Marcus, 2008). Nevertheless, they could not control the fact that picturebook creators worked across a variety of forms nor could they control how children spent their money. Beginning in the late 1800s middleclass children had pocket money to purchase their own reading material. In the early part of the 20th century, these children purchased the new ?comic book.? Their desire for comic books and their buying power contributed to a comic book boom (Sabin, 1996). Among these increasing numbers of children reading and naturalizing comic conventions were future illustrators such as Sendak, who eventually used comic panelling to create his own picturebooks, such as In the Night Kitchen (1970).   The simultaneous development of the narrative forms of comics, ?big little books,? films, and animated films is significant to the picturebook because many illustrators worked with all these forms, as well as in graphic arts and design (Marcus, 2007). The illustrators? breadth of background across the narrative arts and graphic design influenced their work and contributed to the significant innovations they made in the form of the narrative picturebook. However, because of the literary divide between pop culture?s narrative forms and literary books for children, some of the early innovations by crossover illustrators failed to be recognized for their literary merit and then failed to sell. It was not until well into the 20th century that comics and graphic novels received literary recognition. Ironically, the recognition was due in part to the picturebook form?s increasing deployment of comic conventions from Edward Ardizonne?s work in 1930 to David Weisner?s books from the past 25 years. Up to this day, comics continue to influence the picturebook format, as in the recent work of Mo Willems and Jon Klassen. To conclude this section on the origins of the picturebook, I provide a snapshot of some key changes that took place during the 20th century in the children?s picturebook market. The book women played an enormously important role in establishing children?s library departments, supporting children?s book publishing, and encouraging the use of children?s books at home and at school (Marcus, 2008). In fact, in the 1930s, across North America, public librarians made visits to schools and invited school classes to visit libraries. They had 33  a hand in reviewing books in magazines, newspapers, and on the radio, and in establishing what is still considered one of the most important journals of children?s literature, The Hornbook. They also supported the development of preschools and kindergartens and championed the notion of a library in every school. By the 1960s, there were kindergartens, preschools, and school libraries in most public schools across North America. The book women and the public librarians who followed them were so successful in promoting the value of school libraries and school librarians that their influence ushered in a new generation of children?s book promoters. These were the public educators and school librarians who ultimately became the most influential buyers and reviewers of children?s books. Back then public educators and school librarians took on a good deal of power over and had great influence on children?s book publishing eventually taking it away from the book women. When that happened, the biggest children?s book market was made up of public schools and their libraries. The boom lasted into the late 1980s and early 1990s when budget cuts reduced the size of school library budgets and cut library staffing. 2.6 Innovations in the Picturebook Format  So far, in this review I have focused on technological advances and changes in social practices related to illustrated books. Now the focus shifts to the key innovations in the format by individual picturebook creators. Although for the past hundred years printing and reproduction advances along with shifts in social contexts changed production and distribution practices, individual picturebook creators have been the prime innovators of the picturebook format (Bader, 1976).  2.6.1 William Nicolson and the First Modern Picturebooks Several important books published in the 1920?s clearly established the modern picturebook format. Until the 1920s, the illustrated book format dominated the children?s book trade. Talented illustrators increased the number of illustrations in books revealing some of the form?s visual narrative possibilities. However, in the mid-1920s in England, William Nicolson, a noted designer and illustrator created two small but landmark picturebooks for his grandchildren. These books have a handmade look in the fashion of Victorian family 34  picture/storybook making of Nicolson?s youth (Alderson, 1987). The books also reflect the influence of Victorian social leisure time practices on the first picturebook creators Nicolson?s innovations in Clever Bill and The Pirate Twins initiated dramatic changes in the illustrated book format. Nicolson employed a very spare text that relied on the colour illustrations for descriptive information and he spread the words of sentences across multiple double spreads, inducing the reader to turn the page to read the next clause or word. The artwork is line-based, as is much of the work of the first picturebook illustrators (such as Caldecott), in contrast to the painterly work of Potter and many of the golden age illustrators. Nicolson?s work is also innovative in its use of a limited palette of colours, which gives the book an immediate general visual cohesion.  He also utilizes nearly full-page illustrations on every page of the book, breaking away from the tradition of a page of verbal text adjacent to a page of illustration. To construct and pace the visual and verbal narrative text Nicolson uses the material features of the printed book. His innovations inspired other creators such as Wanda Gag and Marjorie Flack in the United States to play with the form (Bader, 1976). From the 1920s to the 1960s, the picturebook form continued to evolve into the classic form exemplified by Sendak?s 1963 publication, Where the Wild Things Are (Bader, 1976).  2.6.2 The Evolving Picturebook Format from 1929?1963  The most significant changes in format were the conventionalization of composite illustrations; that is, the integration of text and image on the page, the increased use of cinematic perspective, the use of comic panels and speech balloons, the dominant use of line-based art, the increased use of full colour printing, and the deployment of limited colour palettes.  2.6.3 Integrating Text and Image   In 1929 in the United States Wanda Gag?s Millions of Cats, a book influenced by Nicolson?s work and her own work as a graphic designer of posters and editorial art, attracted critical attention for its integration of image and written text. Gag uses a horizontal or landscape 35  format and the expanse of the double spread to depict a pastoral environment. The narrative is a traditional folk tale with a verbal text that can tell the story without images. However, Gag?s innovative winding of the text around illustrations and across pages?in other words, her conceptualization of words and text into a composite whole?make it an important book in the evolution of the picturebook format.    Gag?s composition set the precedent for integrating verbal and visual modes, the hallmark of the narrative picturebook format. Composite texts of words and images were used before in illuminated manuscripts, illustrated books, different kinds of graphics, and even by the gifted illustrators Walter Crane and Randolph Caldecott. However, Gag?s integration of verbal and visual components actively supports the sequencing of the narrative. The undulating flow of hills across the pages of double spreads articulates the journey of an old man across hill after hill and bleeds off the page inviting the reader to turn the page to see where he is going. In Millions of Cats, Gag uses the horizontal expanse to support the notion of a large world of hills. As Barbara Bader notes, the success of Millions of Cats has to do with text, drawing, and format (Bader, 1976). Gag?s contribution to the evolution of the picturebook is in her use of the physical horizontal format of the double page spread to serve the visual sequential storytelling. In another book, Snippy and Snappy, she creates the contrast of a tiny enclosed world that flows under a larger landscape, again taking advantage of the full width of two pages.  2.6.4 Refining Composite Illustration in Picturebook Formats and Adding Colour  Drawing on the use of the horizontal and the integration of text and image, Marjorie Flack in Angus and the Ducks (1930) added the use of a limited colour palette to the picturebook format. Flack uses her greatest asset, what Barbara Bader (1976) called ?a feel for stories?situations, for the most part?that would tell well in words and pictures and a knack for dramatizing them: a true picturebook sense? (p. 61). Although Flack is not considered one of the great writer/illustrators, it is her feel for stories that attracts readers to her books to this day. It is also her keen sense of the sequential visual possibilities of the picturebook format. Oral storytelling relies on repetition, timing, contrast, and surprise. Flack applies precisely 36  those dynamics to the format and design of her books and uses the drama of the page turn to provide a sense of time elapsing and to develop and convey tension and anticipation. Flack?s use of oral storytelling dynamics reflects the social practices of reading aloud. The patterns in the images and verbal text are realized in the oral sounds and pace of the text as it is read aloud. Furthermore, Flack?s understanding of her audience of young children is evident in the clarity of her images and easy-to-follow sequences of visualized narrative events.  2.6.5 Mechanical Colour Separation  Flack?s use of a limited colour palette is directly related to the technique of mechanical colour separation. The technique involves creating a master drawing and then four overlays, or three if the master drawing is done in black line. Each of those drawings must be perfectly registered in a manner similar to the registration procedure used for the printing of silk screens and colour woodcuts. For illustrators who worked in this manner the great challenge was to retain the liveliness of the line and the colour. That was always difficult after drawing the same illustration or parts of it at least four times. Some of the best work was done by illustrators who had been printmakers, such as Taro Yashima in the 1950s, because they were used to the way printing inks combine to create colours, and they possessed the essential skill of leaving places for the various colours to stand alone and convey information without appearing over drawn. Until the late 1980s, the use of mechanical colour separation was a cost saving measure for publishers. Not all colour printing in children?s books was done by mechanical colour separation, but it was a frequent choice since it reduced production costs. Publishers asked illustrators to do the separation so that they could publish more books than if they had to pay to have artwork photographically separated. Variations in the number of colours mechanically added can be seen in books printed in one or two colours such as many of Dr. Seuss?s works. As well, many books were printed in one colour only, such as McCloskey?s books Blueberries for Sal which uses a dark blue colour reminiscent of blueberries, and Make Way for Ducklings done in sepia ink. Most mechanically separated colour art for books from the 1920s to 1980s involved line-based art with the addition of flat colour or colours. There were a few highly skilled 37  illustrators who created separated half tone art but they were the exceptions. Line art dominates the picturebook until the 1940s. There are some highly colourful picturebooks such as Bemelman?s Madeline books, but they are still based upon line. It is illustrators such as Leonard Weisgard and Clement Hurd in the 1940s who worked in a painterly fashion that no longer relied on line. Their work was not easily mechanically colour-separated, so the colours appear naturalistic and not so process-like. The process colours can be quite different from the primary colours. For instance, the blue is a cyan blue, which is closer to turquoise than to primary blue, and instead of red, there is a magenta that requires the addition of yellow to create a primary red colour. The yellow and the black are close to what most people associate with yellow and black. The colours in Marjorie Flack?s books look very much like unmixed process colours, which are very bright and stark against the white of the paper they are printed on.   The increased use of black line art and hand separated colours influenced both the evolution of the picturebook format and the evolution of its creators? grasp of the format. For the first half of the 20th century, children?s book publishers chose to publish most of their picturebooks using line art with mechanically separated colour art. They used full colour (photographically separated) only for books by best-selling illustrators and writers. Illustrators such as Sendak began their careers doing line art. Next, they were given projects that allowed them to add one or two colours. By the time Sendak created Where the Wild Things Are (1963), he had illustrated well over 50 books (Bader, 1980). Although Where the Wild Things Are is line-based art, Sendak?s extensive use of cross-hatching and full colour verges on the painterly. The book feels entirely of one piece, it is a fully realized multimodal world, widely acclaimed as a literary work of art. Ezra Jack Keats? books from the 1960s also possess the kind of wholeness that emerges from the rendering of a fully realized visual environment. Just as the narrative picturebook evolved from its origins of lively narrative line work into a full colour 32-page world, many emerging creators learn to do picturebooks by first learning to show a narrative in a sequence of line drawings, then adding colour, and then, sometimes adding a painterly effect with colour or with cross-hatching.   38  In terms of innovations to the format, the key issues were the use of line-based art, the increased use of colour in response to market demand, and the two kinds of colour separation (mechanical and photographic). Flat, black line-based art was favoured for many years for stylistic and economic reasons. The addition of mechanically separated colour was popular for its economy as a step up. The processes of both mechanical and photographic colour separation influenced both the illustrators? choices in media and the art directors? preferences and expectations for illustration media. The same influences affected illustrators working for newspapers, magazines, and on other kinds of illustrated literature. For example, the use of certain kinds of media such as oil paints, pastels, and coloured pencils, were cautioned against because they made the final artwork more difficult to handle and separate. Also, certain colours, such as human skin tones, did not separate well. The limitations on media and particular colours and the cost of colour separation were partially responsible for the dominance of black line art and watercolour or gouache (opaque watercolour). There are other factors that have and continue to support the ongoing use of line and watercolour in illustration. Among those factors are the potential liveliness of line work, the speed at which it can be rendered, and now the ease with which it can be scanned and adjusted digitally. 2.6.6 Edward Ardizonne: Using Comic Conventions The work of Edward Ardizonne, particularly in Little Tim and the Brave Sea Captain (1936) and other Little Tim books, demonstrates the influence and adoption of comic conventions in the picturebook format (Lewis, 1996). Ardizone?s work was done in black line drawing with the addition of touches of watery watercolour. He is one of the first literary picturebook illustrators to use the convention of speech balloons. The drawn lines delineating the balloon and his hand lettering are in the same broken pen-stroke style as the rest of the drawn images. In fact, his subtle use of the convention may have escaped the book women?s notice because of the overall appearance of his work with its brilliant line drawings in the style of fine British literary illustration. Interestingly, in tracing the history leading up to the recent acknowledgment and increased use of graphic novels in classrooms and for circulation in public libraries, it was the gradual adoption of comic conventions by picturebook creators, beginning with Ardizonne, that gently opened the door to both the acknowledgment of the graphic novel?s literary genre as a literary format and to the practice of using literary graphic 39  novels in the classroom. Prior to the mid 1990s the only comic-like books found in libraries and classrooms were Tin Tin and Asterix along with hybrid picturebooks such as the Little Tim books (1936-1977), Raymond Briggs? Father Christmas (1973), and Sendak?s In the Night Kitchen (1970).   2.6.7 Comic Books and Mass Trade Books for Children Throughout the history of the picturebook in the 20th century there are ongoing connections to comics, animated films, and the growth of mass production printing. The Western Publishing Company in Racine, Wisconsin, began as a printing firm. A large portion of their trade during the 1930s came from printing and reprinting comic books based on Disney, Warner Bros., and Dell characters, to name a few. As the Western Publishing Company grew, they began to publish their own books. One of their popular products was the Big Little Books imprint, books of reprinted comic strips in a small form (4? x 4?). In the 1940s, in partnership with Simon and Schuster, they began publishing the Little Golden Books (Marcus, 2007). Since they were mass produced, cheaply priced at an affordable 25 cents each, and related to the film and comics businesses, much of the children?s literary community undervalued their contribution to children?s literature. Little Golden Books were sold in grocery stores and in the five-and-dime chains. They cost about a tenth of the price of a hardback children?s picturebook, making them affordable to millions of people. Indeed, millions of them were sold across North America.  In fact, one favourite, The Pokey Little Puppy, has sold over 15 million copies. Although the Little Golden Books list included a number of Disney-inspired stories, it also included works by many of the most talented children?s book writers and illustrators from the 1930s onward, such as Margaret Wise Brown, Garth Williams, and Gustav Tenggren. Many of their works were well written and beautifully illustrated. The Little Golden Books had a specific format so they tended to be illustrated storybooks, but among them were some picturebook gems such as the favourite, The Pokey Little Puppy. The popularity of the Little Golden Books supported the development of the Giant Golden Books and other mass published trade books illustrated by the likes of Leonard Weisgard, the Provensens, Richard Scarry, and many other gifted creators. 40  The Big Little Books, comic book series fiction such as Nancy Drew and the Hardy Boys, and the Little Golden Books were the so-called ?children?s pulp fiction? of the 1930s until recently. Now, with the popularization of the graphic novel, versions of the children?s series literature, such as Nancy Drew and the Babysitters Club, have been turned into graphic novels and added to children?s pulp fiction. However, a shift in values has influenced an interest in literary graphic novels for young readers along with a renewed appreciation of comics. However, sales of comics, while making a slight recovery, are dramatically short of the annual sales in the 1950s which were well over a billion copies a year. Today only 60 million copies are sold annually (Tilley, 2013). As has often been the case with popular forms of literature, academic literary scholarship has elevated the status of the Little Golden Books and now recognizes them for their contributions to children?s literacy and literary development (Marcus, 2007). I mention these books to note that just as reading comics influenced many children?s book illustrators, the Little Golden Books and the large format Golden Books also influenced generations of illustrators and animators. 2.6.8 The Visualizers and Picturebook Worlds In the world of animation, there are two kinds of creators, the visualizers and the animators. They do two separate and very different kinds of work. The visualizers imagine or visualize the look of the story to be animated, rendering visual representations of the characters, props, scenes, settings, and everything that the animators need to animate. The animators then take the visual representations and animate them; that is, they render the sequence of stills that will create the illusion of movement when screened at 22 stills or frames per second.   The connection between animated films, comics, and picturebook illustration is evident in the roster of Disney visualizers, which includes Mary Blair, Gustav Tenggren, Alice and Martin Provensen, and Kay Nielsen. These people were not the animators, rather they were the artists who conceptualized the overall look of the animated films and who also worked as children?s book illustrators. This tradition of animators and visualizers continues today in the animated film world. Sometimes, exceptionally talented animators and visualizers move into children?s book illustration so that they can have more control over their work. Film animators and visualizers may belong to a cast of hundreds with few opportunities to tell 41  their own stories. The visualizers? contributions to the picturebook came from their practice of conceptualizing, usually in a painterly way, the whole world of a story, its setting, characters, and tone. As the picturebook format evolved, many of the visualizers were particularly gifted at conceptualizing a story into a picturebook world, such as Tenggren who also created art for the Little Golden Books. Animation artists brought two kinds of expertise to picturebooks. The visualizers promoted a painterly conceptualization of the story and the animators could draw characters in action and bring them to life on the page. 2.6.9 Animators and the Illusion of Movement  In all sequential narrative arts, (especially in the picturebook with its limited length), it is always challenging to depict action and to create the illusion of motion on a static page. The invention of animated film particularly appealed to illustrators because the projection of many still images gave life to their drawings and created the visual illusion of movement, the very thing most illustrators desire. Canadian filmmaker Norman McClaren was especially attracted to animated film for that very reason. The potential to create the illusion of movement is one of the reasons for so much back and forth migration by illustrators from animated films to picturebooks from the early days of each art form. Animators and sequential narrative illustrators must be able to create the illusion of movement and break it down into its fundamental actions to catch the split second gesture and render it in a lively line. Animated film appealed to and drew on the talents of two kinds of illustrators: those who liked to design the sets, costumes, props, and the overall look of things, and those who wanted to draw sequential narrative stories and animate the action. Of course, there is never such a precise division, but what has been true is that both kinds of illustrators have been drawn to animated films and picturebooks. The crossover from animation to children?s book illustration continues today with, for example, one of Canada?s bright new picturebook creators, Jon Klassen, winner of the 2013 Caldecott Medal and 2013 Caldecott Honor Book, who works as a visualizer for Dreamworks. Another example is Mo Willems, who wrote and illustrated Don?t Let the Pigeon Drive the Bus and the Knuffle Bunny after winning an Emmy Award and  having a career as an animator on Sesame Street.   42  2.6.10 Cinematic Points of View  In the 1930s, live action cinema influenced the picturebook as can be seen in Robert Lawson?s illustrations for Ferdinand by Munro Leaf (1936). Lawson, a good example of a picturebook illustrator who utilized a cinematic point of view in his books, wrote and illustrated his own work, and illustrated many books by other writers. The Newbery and Caldecott awards he won for his work acknowledge his verbal and visual narrative skills and talent. Ferdinand is one of the first picturebooks to reflect the direct influence of cinema. It is therefore not surprising that soon after its publication Ferdinand was made into a Disney animated short. Lawson draws in the tradition of Caldecott, rendering his illustrations with lively line, excellent draftsmanship, and strong composition. His work is highly realistic but with comic touches. For example, he renders a cork tree with bundles of corks hanging from the branches. His compositions vary from close shots to broad establishing shots. His series of 36 illustrations read like a film storyboard. Anita Silvey (1995) noted that Lawson was primarily a storyteller since in his written work he ?created character through dialogue rather than description? (p. 396). In view of his visual narrative skills, it seems that Lawson had a keen sense of what words do best. Of course, in Ferdinand Lawson illustrates another author?s tale, but nonetheless he uses images to enhance the descriptive work, showing and setting the scene, creating the particular look of the characters and their actions, and adding page turning rhythm to the pace of the storytelling. He achieves this by the careful selection and conceptualization of the actions best suited to move the narrative forward. Ferdinand is a good example of the cinematic action possible with 72 pages, a relatively spare verbal text, and black and white illustrations. The number of illustrations and the number of quick page turns creates a sense of movement and animation. Today?s picturebook tends to be 32 to 40 pictures, which is less than half as long as Ferdinand. The length of picturebooks fluctuated throughout the early to mid-twentieth century. Robert McKloskey?s Make Way for Ducklings and Blueberries for Sal, to name only two examples, are 70 and 56 pages respectively. Over the past 60 years, for reasons of economy, the typical number of pages in a picturebook became fewer. To compensate for having less space to tell the story, many illustrators used comic conventions such as paneling to create additional scenes and the dynamic of ?continuous narrative,? as used by H. A. Rey in Curious George.  43  2.6.11 Continuous Narrative In all sequential narrative arts, and especially in the picturebook with its limited length, there is the challenge to portray action and to create the illusion of motion on a static page. The dynamic of continuous narrative is one way to represent a sequence of activity by illustrating a character in a quick succession of actions within a single or double page composition (Schwarcz, 1982). H. A. Rey makes frequent use of this dynamic in his popular Curious George book. For many years, children?s book illustrators, from Ernest H. Shephard to Maurice Sendak to Oliver Jeffers, have used this ancient illustration dynamic to convey movement and change (Schwartz, 1982). It is among the many kinds of illustration conventions children learn through reading picturebooks and other illustrated materials (Meek, 1988). In terms of picturebook dynamics, the use of continuous narrative and paneling is not only a means of creating the illusion of motion, it is also a resource for creating the cohesive rhythm of a text and for controlling the speed of the page turning, and hence the pace of the overall narrative. It is a resource for creating salient patterns of activity because it draws the viewer into a repeated focus by following a character?s or set of characters? activities.  Ludwig Bemelmans, the creator or the Madeline books, believed that clarity trumped beauty and that children?s book illustration ?must be absolutely clear for children. A flower must be a flower, the sun must be the sun. It can be stylized, it can be an image of a thing, but it must be simple, clear and at once understood? (Marciano, 1999, p. 130). Ideally, clarity and simplicity go together with great legibility and great line, as is always the case in the work of McCloskey. Rey?s work has great legibility and supports an amusing story. His work is a fine example of how great illustration is not always beautiful but it always contributes significantly to the narrative. In books for children, great illustration can rarely save a weak narrative, but a great narrative can survive weak illustration. The illustrations in Rey?s Curious George support the verbal text and progress the narrative. Furthermore, with its use of continuous narrative and multiple scenes, it works like a comic book without panel lines and speech balloons. McCloskey?s books work in a similar way although the illustrations are on a larger scale than Rey?s. The drawings by McCloskey and Rey sequentially progress the 44  narrative. In doing so, their work contributes to the establishment of key conventions related to modern narrative picturebooks. McCloskey contributed the convention of creating picturebooks writ large in a scale that supports group read alouds. The use of over-sized pages became a significant picturebook asset during the years when McCloskey began illustrating because public libraries, classrooms, and the growing number of nursery schools needed books with pictures that could be read from five to ten feet away. Later, in the mid-20th century, even larger format versions of popular picturebooks were published for use in groups, particularly for reading to entire classes.  For groups regular picturebooks and large format versions required highly legible illustrations that could be enlarged such as McCloskey?s in Make Way for Ducklings and Blueberries for Sal. His strong animated line work is ideal for printing in a variety of sizes. Of course his brilliant verbal storytelling makes them successful read alouds. McCloskey created strong patterns in the language of the narrative that are supported by strong patterns in the drawing and by the composition of the illustrations. While these two books are not deeply layered with meanings, they show the potential of words and pictures to work together in a multimodal literary artifact and to convey to young viewers/listeners the sights and sounds of fine literary patterning.        2.6.12 What Comes First in a Picturebook: Words or Pictures?  In books written and illustrated by two people the written manuscript almost always comes first. Even among illustrators who write their own stories, many claim to write the story first. Robert McCloskey asserts that the process of conceptualizing a picturebook began in his head where he visualized the whole scope and sequence of the story before drawing anything on paper (McCloskey, 1963). His process was rather like visualizing a small film or play. After he worked things out in his head, McCloskey wrote out the verbal text and then began to conceptualize the images. He claimed that the images in his head changed over a period of time but that the verbal text did not.  There is always a kind of chicken and egg debate over what comes first in picturebooks. As mentioned above until recently, a majority of picturebook creators who both write and illustrate claimed they wrote the story first and then illustrated it. From a practical point of 45  view, it is in fact much easier to change the words and rewrite a text than it is to draw all the pictures and then add the text. However, Virginia Lee Burton one of the most influential early picturebook creators and a contemporary of McCloskey, preferred to begin with sketches of all the pictures. Barbara Bader notes in describing Burton?s process that Disney cartoons were conceptualized in a similar fashion. Burton was a natural visual storyteller. She pasted words onto sketches and claimed, ?Whenever I can substitute picture for word I do.? (Bader, 1976, p. 199). Burton also wound the printed words around images creating composite word and picture texts. In her classics The Little House and Mike Mulligan?s Steam Shovel, as well as in many of her other books for boys, each page or double spread is a composite text. She often incorporated ?hypertext? in the margins of her books much the way Ernest Seton Thompson did in his 19th century illustrated novels for boys. Burton?s earlier contribution to the evolution of the picturebook was the further integration of text and image with the addition of soft colour into composite illustrations. Later she created sequential line illustrations inspired by both comic book art and the drawing of Caran d?Ache, for which the brand of drawing materials is named. Burton continued the gentle adoption of comic conventions begun by other picturebook creators, such as Ardizonne. In light of new drawing and animation programs available to illustrators for visually conceptualizing stories, there may be a shift from primarily narrative-first texts to image-first texts. The proliferation of opportunities for very young children to draw online and to learn to crafts and manipulate images will likely influence coming generations and their work with visual sequential narrative forms including the picturebook. 2.6.13 Style  As mentioned previously, both the cost of colour separation and specific limitations of its technology influenced illustrators? choices of media. Other limitations associated with color separation and printing technology, such as the maximum and minimum sizes allowed for the final art were also key factors in picturebook design and production. This particular limitation has a bearing on the size and the medium an illustrator chooses for the final art. Furthermore, these choices influenced the appearance of picturebooks at particular periods of time. The size of the final art is not necessarily the same size as the art as it appears in the 46  book. In pure black line drawing there is little problem in substantially reducing or substantially enlarging the original drawing because black line will hold its character and crispness. This is one of the reasons it has been used so extensively in book illustration. Black line can be drawn on paper with black ink and a nib or brush.  Black line art is also created using scratchboard (a clay-coated paper or board that is fully or partially covered with blank ink) as a medium. A scratchboard tool, a tempered steel point, is used to scratch through the ink revealing a white line. Thus the illustration can look like a drawing in reverse and is produced by drawing the light rather than the dark. Scratchboard, like black line ink work, can be successfully reduced or enlarged. Just as a cross-hatching of fine black lines can create a dimensional look, so can the cross-hatching of fine white lines on scratchboard. Scratchboard has been a popular children?s book medium since the early 20th century and was the medium of choice for a number of notable illustrators such as Don Freeman and Barbara Cooney. Today it is still used for chapter books and other formats using spot illustration, as well as for picturebooks such as the recent 2009 Caldecott winning, The House in the Night by Susan Marie Swanson and illustrated by Beth Krommes.  Barbara Cooney, the American illustrator, used scratchboard to illustrate of her books for the first half of her career. In the second half of her six decade long career she used gouache (opaque watercolour). In doing so, she created picturebooks rather than illustrated books. This change of medium in Cooney?s work?the change to filling the pages of a book with a full colour realization of a narrative place was a change in her narrative conceptualization and realization of the story. Scratchboard does not preclude being used in a picturebook format, but for the most part, it has been used for illustrated books. Yet, a few contemporary illustrators, such as Brian Pinkney and Jan Thornhill, have added colour to their scratchboard work to render full-colour picturebook environments.  In terms of the picturebook format, the use of black line and scratchboard requires illustrators with strong drawing skills and a talent for suggesting action. From the first picturebooks to the present, black line and line-based illustration remains a frequently used medium. It is a medium well suited to the use of comic styles and conventions and to the playful use of 47  typography. Black line and scratchboard tend to be used for spot or framed illustrations and rarely for full bleeds.  The term full bleeds refers to art that extends beyond the edge on all sides of the page and its use tends to bring the viewer into the illustrated world. A prime example is the set of three wordless double spreads in the middle of Where the Wild Things Are depicting the rumpus. This wordless sequence highlighted the meaning-making potential of full bleeds and inspired a proliferation of full-colour picturebooks. Other illustrators such as Ezra Jack Keats also exploited the potential of full-colour bleeds and inspired other illustrators to do the same. And, as mentioned earlier, illustrators such as Barbara Cooney and Don Freeman, who had spent half their careers doing black and white line work, began to illustrate in full colour creating more of their books in picturebook format.  2.6.14 The Full-Colour Picturebook As the use of colour became economically attractive to publishers, and as reproduction technologies improved, there was a gradual evolution of the picturebook conceptualized as a multimodal full-colour rendering of a narrative world. Improvements in full-colour printing allowed for the rendering of illustrations of high modality. This notion of high modality draws on Kress and van Leuween?s (2006) use of the term modality to indicate degrees of realistic representation. Realistic representation is a kind of highly true to life rendering that invites readers to easily identify and enter the illusion of a ?real? narrative world. Not all illustrators use this potential for creating a ?real? world. Some choose to use a limited palette that uses the resources of full-colour printing, but with fewer colours in the service of constructing colour cohesion and possibly rendering the narrative in an abstract fashion. The different choices lead to the immersion or distancing of the reader/viewer. The decision to immerse or distance the reader/viewer not only reflects the use of new meaning-making resources, it also reflects social practices and values in terms of what is considered acceptable or appropriate content experiences for the intended audience.   Illustration styles are socially and semiotically created because they are responses to social practices that are in turn responses to, and uses of, technological inventions and innovations. Like picturebooks, they are instances of cultural meaning making. Thus, in reviewing the 48  evolution of the picturebook format, it becomes apparent that stylistic choices in media and content reflect ever-shifting social contexts. By 1960, the picturebook format drew on four primary styles related to the choice of medium and media. First, there was the line-based style that at times included the addition of colour from another medium. Next, there was line-based art with or without colour that extensively used the resource of typography for the visual design and to convey verbal content. A third style, which is particularly popular again in this digital age, was based on flat shapes rendered in black and white or in colour. Throughout the history of the picturebook there have been illustrators working in this third style. It was particularly popular in the 1950s and 1960s with advertising artists, some of whom went on to become children?s book illustrators, such as Eric Carle, Leo Lionni, Bruno Manari, and more recently Lois Ehlert. Filmmakers and animators working in this flat style, such as Mary Blair and Saul Bass, also went on to create children?s books. The fourth style uses full colour and tends to be used to realize a whole narrative world. All four styles may use a full range of colours or a limited palette. Although there have been picturebooks that use photography, it is not a frequent style choice, thus I have not considered it here. The infrequent use of photography is primarily due to the fact that a book done entirely in staged photography can be very expensive. However, there are some interesting hybrid media styles that use bits of photography collaged into compositions.  While there are four dominant styles related to media, there is also variation within these four styles. There are also what I call hybrid formats. For example, there is the hybrid comic/graphic novel/picturebook, in which most of the comic conventions are deployed, including the extensive use of panels and speech balloons. The work of Mo Willems, Raymond Briggs, and David Weisner are hybrid comic/picturebooks. There are picturebook hybrids that include significant wordless sections such as Weisner?s Tuesday and Sendak?s Where the Wild Things Are. Currently, one of the most popular picturebook styles is cheeky and self-consciously metafictive. The metafictive styled book is an important picturebook hybrid because of the embedded discussion of its features of form. All of these hybrid forms are often used for the new and fast-growing category of picturebooks for young adults and adults. These books tend to deal with serious themes, employ multiple narratives, and contain 49  multiple layers of meaning. They may use a variety of media, styles of rendering (including the use of editorial art styles) metafictive elements, and be produced as very long books, sometime in unusual shapes and sizes.  Between the 1963 publication of Where the Wild Things Are and now, the picturebook format has firmly established itself in the literary world. The format and the use of media in it have changed to take advantage of some of the best affordances of digital progress. In addition, the content has changed to reflect successive generations of children and their concerns.  The contemporary picturebook continues to evolve in response to innovations in technology and social use. In the next chapter, I provide a critical review of scholarly work on the picturebook with a focus on those studies that look at how meaning is constructed in this literary format. 50  Chapter 3: A Review of the Academic Search for How Meaning is Constructed in Picturebooks 3.1 A Review of Scholarship on the Forms and Meanings in Picturebooks This chapter reviews the achievements of scholarship devoted to exploring relationships between form and meaning in picturebooks, and delineates a crucial gap in the research. I present this discussion of the academic studies on the picturebook with a profound respect and appreciation for the deep knowledge of children?s literature evident in the work of these scholars. Years ago, when asked what gave her the right to make a pronouncement about a number of books, Sheila Egoff, a well-known children?s literature scholar, responded that many years (at that time it would have been about 40) of reading, rereading, and writing about children?s literature gave her the right.3 This is the case with the scholars reviewed in this chapter. My criticism of some of their findings does not negate the benefits of their collective insights that continue to inform and shape the growing body of scholarship on picturebooks.   A great deal of scholarship has been produced on illustrated children?s literature and on the picturebook in particular. Much of that work pertains to how meaning is construed; however, this study explores how meaning is constructed. The distinction is between the meaning readers construe and the meaning authors and illustrators construct. Much of the existing research focuses on the kinds of meaning construed by young readers in particular, and how the reading of these books supports literacy and literary development (Appleyard, 1990; Chambers, 1991, 1995; Day, 1996; Graham, 1990; Heilman, 2005; Lysaker, 2006; Mackey, 1993; Meek, 1991, 1992; Nodelman, 1988; Pantaleo, 2004; Paris, 2007; Paris & Paris, 2001; Stephens, 1992, 2003; Walsh, 2003; White & Low, 2002; Williams, 1998). Since the mid-1950s, there has been considerable interest in the art of the picturebook, as well as in how readers construe meanings in them (Azripe & Styles, 2003; Klemin, 1966; Pantaleo, 2004,                                                  3 Sheila Egoff, a well-known children?s literature scholar and personal friend related this anecdote in several conversations with me and other children?s literature scholars including Professor Judith Saltman (Shoemaker, personal communication, 2002). 51  2008; Watson & Styles, 1996). These two interests reflect a significant shift in the context of picturebook use, from primarily home/library to educational settings, and is manifest in enormous increases in classroom use in support of literacy and aesthetic development (Doonan, 1992; Keifer, 1995; Schwarcz, 1982; Sipe, 1992, 2001; Styles, 2001). Up until 30 years ago, only scant attention was paid to the meaning-making relationships between language and illustration. This may be due in part to the brilliance of the best picturebooks in that, as is often the case with great art, the construction or technique is invisible. An exception to this has been the more recent proliferation of the metafictive picturebook, a narrative form that actively promotes textual metacognition and exploration. Metafictive picturebooks make textual narrative meaning making highly visible.  As the narrative picturebook form evolves and is acknowledged as a literary art form, theoretical and methodological approaches to social research are also evolving. These developments are reflected in some of the new theoretical perspectives used in critical studies of picturebooks (Dresang, 1999; Pantaleo, 2008; Painter 2011; Unsworth, 2006;). For instance, there is a good deal of work on how picturebook art can be used as a resource for aesthetic and artistic development (Albers, 2007; Dank-McGhee & Shaffer, 2003; Doonan, 1992; Gomez-Reino, House & Rule, 2005; Keifer, 1995; Leddy, 2002; Marantz & Marantz, 1992; Schwarcz, 1982; Sipe, 1995, 2001; Styles, 1996). Other work focuses on how children construe the narrative conventions in narrative picturebook texts, and on features of form responsible for constructing narrative sequence even though the intent of the studies is primarily to describe conventions that support children?s literacy and literary development (Chambers, 1991, 2008; Graham, 1990; Mackey 1993; Meek, 1991, 1992). While the scholarly interest and research on picturebooks increases, as does the publication and use of them, only a few studies focus specifically on how words and images textually construct meaning in narrative picturebooks. Research on language, thought, and culture from various academic disciplines had provided literature studies new ideas and social research strategies to draw upon. For instance, in the 1970s, literary theorists began to embrace the theory of reader response, such as the work of Wolfgang Iser and Louise Rosenblatt, and the study of semiotics, which typically focused on meanings construed. Interest in semiotics and its potential as a resource for close picturebook 52  analysis along with changes in academic perspectives and new theoretical resources, set the stage for scholarship in the 1980s to begin to look at how meaning is constructed in the picturebook. During this period, studies by academics such as Alderson (1986), Schwartz (1982), Moebius (1986), and Nodelman (1988) acknowledged the sequential nature of the picturebook format and the importance of considering each illustration in terms of its sequential position. This important acknowledgment supports a focus on picturebook art that factors in its sequential nature, particularly in the narrative picturebook.  While many picturebook studies have looked at the meaning construed, in what follows I focus on how meaning is constructed by textual resources within the context of the other semiotic systems in play in the picturebook format.  Beginning with important scholarship of the 1980s to the present, this discussion looks at the studies that make the greatest strides in furthering the understanding of the picturebook format, its meaning-making features, and its significance as a new literary format. While these studies make an important contribution to picturebook scholarship, it is important to note that the perspectives they bring reflect their historical context. As such, many of their limitations are due to the paucity of theoretical resources available at the time they were undertaken.  3.2 Form and Meaning in Sequential Narrative Picturebooks  In Ways of the Illustrator: Visual Communication in Children?s Literature (1982), Schwarcz clearly states his intentions to ?explore how aesthetic configuration expresses content and meanings and to explore how illustration relates to verbal texts? (p. 4). Schwarcz studies the work of illustration without defining varieties of formats. His interest is in how illustration communicates symbolically, the ways illustrators use pictures to express content and meaning, and how they relate words and images to each other. Schwarcz acknowledges that he applies no theory of illustration. His work does not cite particular theories of language or image, nor does he offer a systematic approach or definition of the picturebook. However, he makes important contributions to picturebook scholarship with his observations on meaning construction related to time, motion, and typography.  53  Schwarcz addresses the sequential nature of illustrations, symbolic meanings, visual perception, several kinds of word/picture relationships, and the dynamic of continuous narrative. He does not discuss the dynamics and relationships within and between modes that construct cohesion, such as the roles of colour and rhythm in a work. Rather, he emphasises the notion that metaphors can provide a cohesive framework at a semantic level.  Instead of explicating resources for textual cohesion, using the term cohesion Schwarcz looks at how illustrators achieve a unity of expression. He observes that unity involves time and requires a continuity of expression from message to message or from page to page. Schwarcz also regards tone, texture, and design as unifying factors. Further, he describes how the frequent reiteration of the character in both verbal and visual texts contributes to textual unity. Although he uses a different vocabulary, he essentially describes several kinds of cohesive work. His analysis, together with his acknowledgement of the cohesive work of metaphors, relates directly to the semantic nature of cohesive relations between elements.   Schwarcz?s insights and keen observations of the properties of illustrations have influenced the definition of the picturebook. His major contribution is the promotion of the value of close looking and his examination of the complexities of the picturebook?s intermodal and sequential relationships. 3.3 Codifying Principles of Visual Construction A frequently cited classic, the Moebius study (1986), ?Introduction to Picturebook Codes,? published in Word & Image, presents a systematic framework for assessing how visual systems construct meaning. It does not precisely define the picturebook nor does it employ a theory of language or any other theory that might be relevant in both modalities. Rather, Moebius? work can be viewed as a hermeneutic study. It demonstrates that careful looking reveals principles and dynamics of visual communication that can be clearly articulated. He defines codes or conventions of visual practice in light of how they construct meaning (for example the code of left and right). His research describes the key visual dynamics and outlines the start of a system for understanding visual meaning construction. Although it predates the work of Kress and van Leeuwen, Moebius?s discussion of the role of position on the page is similar to Arnheim?s principles of position, which Kress and van Leeuwen 54  adopted and extended. Moebius only briefly acknowledges semiotics, however his study demonstrates the value of a systematic approach and is in fact a fine semiotic approach to the analysis of image. 3.4 Pioneering Semiotic Analysis of Picturebooks In 1988, Perry Nodelman published the first comprehensive semiotic approach to analyzing picturebooks, Words about Pictures: The Narrative Art of Children?s Picture Books. Nodelman used close observation and a mixture of psychology of art and visual perception research to discuss how images contribute meaning to the narrative. His primary observation about how words and pictures work together is that they particularize and limit each other.    Nodelman?s work demonstrates, just as Moebius? study does, that a strategy of close observation and the articulation of visual meaning making is productive, but ultimately limited without the support of a theoretical framework. Nodelman understands and recognizes the need for a form of visual grammar to organize the resources and varieties of visual meaning construction. Nodelman?s call for a grammar was partially answered a few years later by Kress and van Leeuwen?s work on the semiotics of images, Reading Images: A Grammar of Visual Design (1996). Their work will be discussed in detail in Chapter Four. Although Nodelman?s work makes an important contribution to picturebook discourse, it nevertheless has a number of limitations. First of all, he does not clearly define the differences between illustrated books and picturebooks. Indeed, his discussion appears to conflate the formats. In fairness, the book is titled Words about Pictures, and pictures are primarily discussed. However, little attention is paid to the work of words and no theory of language is referenced. While Nodelman may draw on his deep knowledge of English grammar to frame his analysis, his theory of language remains unspecified. Moreover, his study lacks the systematic approach typical of the picturebook scholarship reviewed in this chapter.  Another weakness in Nodelman?s study arises in the application of several complex notions of narrative rhythm to the general category of illustrated books and picturebooks.  This part of his discussion takes a broad approach to the consideration of both the complexity of 55  rhythms in each modality and the intermodal rhythms of particular works. While Nodelman?s book opens the door to an important exploration of how pictures work in a multimodal text, without the resource of a theory of language he had no way of looking at the rhythms of emphasis that words can convey. To state as he does that ?told in words alone, every action described is of equal significance as part of a whole? (1988, p. 255) underestimates language?s resources for establishing variations in verbal significance. Patterns of significance are constructed by specific resource deployments in both verbal and visual modalities and are fundamental to constructing meaning (Halliday, 1994). Nodelman writes about the rhythms of the picturebook narrative. In describing his sense of the rhythms, he considers how the material sequencing of pictures from page to page affects the narrative rhythm. However, in making his case, he concludes that ?the rhythm of pictures in sequence is regular: one strong beat following an equally strong beat? (1988, p. 245). The assumption that all pictures have the same strong beat is contradicted by almost every picturebook extant. The beat of a picture can be strong, stronger, and every degree of less strong. In other words, there is an enormous range in the beat of any sequence of pictures. Furthermore, the beat of the picture in relation to the surrounding pictures is dramatically affected by its size, scale, framing, how it bleeds off the page, or by its colours and tones. Nodelman also introduces the notion that pictures may interrupt the acceleration of the verbal narrative. Clearly he notices how each modality affects the other and he alludes to the role of rhythm as a cohesive resource for a multimodal text. As will be discussed further in Chapter Four, Kress and van Leeuwen (1996) concluded that rhythm is the prime resource for constructing cohesive sequential visual narratives. This is a conclusion also investigated in this dissertation. Nodelman discusses colour in his chapter on format and design. The discussion is primarily about the use of colour to evoke moods and about the social semiotic meanings associated with specific colours. Nodelman?s discussion of colour is on meanings construed. It does not offer insight into how colour constructs specific meanings nor how it is a resource for constructing cohesion across the pages of the text.  56  Nodelman attempts to unravel the complexities of picturebook meaning construction but falters by making unwarranted generalizations. It is true that many great picturebooks mask complexity with their dominant visual texts and deceptively simple tales. They are wholes that yield whole meanings, but their constructions are complicated and not well served by generalized observations of verbal or visual dynamics. However, Nodelman?s landmark work is important because it unravels the complexities of the picturebook text and it continues to generate dialogue about how pictures work. Even with its limitations Nodelman?s study continues to lead the way to further scholarship because his ideas inspire questions that can now be explored using the resources  of social semiotic theories of language and image.    In short, Nodelman makes a major contribution to scholarly discussions of the picturebook. Most of the limitations in his work are due to the fact that the theoretical resources he acknowledges are needed were not yet available to him. His work continues to be cited and is used as a theoretical framework for picturebook explorations. He discusses well-known picturebooks in ways few adult picturebook readers have encountered and introduces readers to what may be their first social semiotic analysis of illustration. While there is a growing body of social semiotic scholarship (for example Halliday (1994), Kress and van Leeuwen (2006), and O?Toole (1994)) it is not widely known to the children?s literature academic community. 3.5 The Application of a Theory of Language to Picturebook Analysis Kiefer?s well-organized 1995 study, The Potential of Picturebooks: From Visual Literacy to Aesthetic Understanding, begins with a clear definition of the picturebook that draws  on the frequently quoted one by Barbara Bader (1976). Kiefer observes that that the picturebook is an art object that is more than the sum of its parts. While Kiefer?s work is typically used as a resource for practicing teachers, it presents a systematic description of the picturebook, its origins, and its social practices. She also demonstrates the application of Halliday?s functions of language to a systematic approach to reviewing children?s verbal responses to picturebooks.  Kiefer?s view of critical picturebook discourse is based on the notion of the picturebook as an art form. She agrees with S. K. Langer (1995) that a theory of criticism must describe how art 57  conveys meaning, which is foundational to work on how meaning is constructed in the picturebook. Kiefer systematically looks at the sets of resources for meaning making in the picturebook as a way to critically viewing them. Combining this approach with a hermeneutic circle of reflection and discussion provides teachers with a rich set of evaluation resources for construing meaning. However, Kiefer?s work does not provide particular insights into how relations between features of forms in the picturebook construct cohesive texts. 3.6 A Systematic Approach Through Social Semiotics  Chapter Four discusses the pioneering work of Kress and van Leeuwen in greater detail. However, mention of their research is included here because it has become an important resource for social semiotic studies on multimodal texts and, to some extent, on picturebooks. Although their book, Reading Images: A Grammar of Visual Design was first published in 1996, and their framework was used extensively in teaching in university courses in children?s literature in Australia (e.g. Williams, 1998), it wasn?t until 2001 that their work was introduced to published picturebook scholarship. Their visual grammar provides an excellent theoretical framework for exploring how visual meaning is constructed by elements of design. Their work draws inspiration from Halliday?s social semiotic theory of language, Systemic Functional Linguistics (SFL). Kress and van Leeuwen primarily analyse non-fiction and single page texts rather than sequential narrative picturebooks. Nonetheless, their work pioneered the application of SFL to multimodal analysis, making it an important model for further use in general and particularly in relation to picture book analysis.  3.7 The Ecology of the Picturebook?s Multimodal Text In 2001 David Lewis proposed an ecological approach to picturebook scholarship in his work, Reading Contemporary Picturebooks: Picturing Text. Extending Nodelman?s semiotic work, Lewis had the benefit of drawing on work by Kress and van Leeuwen (1996). Lewis defines the modern picturebook, describes its origins, and reviews a range of major contributions to picturebook discourse, making the point that while there have been good descriptions of relationships between words and pictures and attempts to categorize 58  picturebooks, every book has its own ecological system. He notes that rarely, if ever, does a book rely on only one kind of word/image relationship. In embracing Kress and van Leeuwen?s work, Lewis (2001) acknowledges a functional approach to semiotic analysis and posits that relationships between modes will vary within a picturebook text. Lewis asserts that the types of relationships between words and pictures will change and reflect the particular content and context of the book at hand.    Drawing on Kress and van Leeuwen, Lewis introduces the notions of offer and demand, as they pertain to visual analysis as well as the concept of variations of social distance between image and viewer. These important concepts are usually understood immediately and instantly applied as a viewing lens. There is great satisfaction in being able to immediately recognize a visual dynamic that makes a certain kind of meaning and conveys a particular kind of ideological view. Because Kress and van Leeuwen focused on single images rather than sequential ones, few of their notions specifically address a sequence of images. As I will argue in Chapter Four, Kress and van Leeuwen provide potential theoretical resources for a particularly savvy analyzer to discern patterns of social distance, salience, framing, and position within a composition that may ultimately reveal some kind of cohesive activity. However, reading both Lewis and Kress and van Leeuwen without the knowledge of SFL makes application of their insights challenging.  3.8 Categorizing Picturebooks by Word/Picture Relationships At the beginning of How Picturebooks Work, Nikolajeva and Scott (2001) state that pictures describe or represent and words narrate. This is both an oversimplification and an incorrect statement about how words work. They view picturebook reading from a hermeneutic perspective and recognize reader response theory in their general review of picturebook literature. Two major concerns about their work relate to their categorization of word/picture relationships and their use of ideas from Jacques Lacan to describe picturebooks for very young children.  First of all, their taxonomy of word/picture relationships only works for individual books that employ only one kind of word/picture relationship. In fact, there is a continuum of word/picture relationships. At one end of their continuum is the wordless book, which is a 59  book with pictures that can only relate to a title and an implied narrative. At the other end of their continuum is a book with all words and no pictures, which by extension contains no relationships between words and images. However, a pictureless book is still visual because the words arranged on the turning pages create visible printed texts and implied images. A major problem with the rest of Nikolajeva and Scott?s taxonomy is that it oversimplifies the range of complex word/picture relationships that exist within any picturebook text, even those that appear to have a dominant modality. One way to consider their taxonomy is as a possible literary framework, but even then it does not account for the range of variations in the relationships between words and pictures. Based on broad literary meanings, their notion applies to only a few books. In their study, particularly in its ambitions to create a taxonomy of word/picture relationships, the absence of theories of language and image prevents the creation of a useful classification system. They explore in detail a variety of word/picture relationships as they pertain to their continuum and then offer particular insights into how those relationships construct meaning. However, their discussion, while connected to their categorizations, otherwise lacks connection to a systematic organization of meaning making resources.  Although the taxonomy has limited application, Nikolajeva and Scott bring their considerable breadth of knowledge of children?s narrative literature to bear on the verbal modality as well as on elements such as setting, characters, narrative perspectives, time, motion, modality, metafiction, intertextuality, and paratext. Their major contribution, then, is to research on the picturebook narrative.  Nokolajeva and Scott?s application of Lacan?s theory of the imaginary and the symbolic as it pertains to language acquisition is at odds with Halliday?s theory of language development. Furthermore, they apply Lacan?s theory to the multimodal relationships in the picturebook and claim that the picturebook is the only literary form that has successfully combined ?the imaginary and the symbolic, the iconic and the conventional? (Nikolajeva & Scott, 2001, p. 262). Their claim does not account for the imaginary and the symbolic in graphic novels, films, and other narrative art forms. However, their ideas about the relationship between picturebook modalities are provocative and worth discussing as they demonstrate that 60  picturebook scholarship by academics with deep literary knowledge contributes and inspires further explication of the format. Nikolajeva and Scott?s work also raises a basic question about the feasibility of creating any taxonomy to classify picturebook texts by word/picture relationships. Indeed, a fundamental problem with their taxonomy is that, contrary to Nikolajeva and Scott?s initial description of the functions of words and pictures, words do far more than represent. In fact, this is only one of their functions.  As such, the central flaw in their taxonomy is that it has not addressed the complex range of functions of both words and pictures. Nor has the study acknowledged the variations in relationships within any given text. 3.9 Relationships Between Words and Pictures Lawrence Sipe?s contributions to picturebook scholarship over the past 20 years sadly came to an end with his death in 2011. However, his work has inspired and continues to inspire further study into how picturebooks work and how children construe meaning from them.    Sipe?s work reinforced the notion of the picturebook as a literary art object. He applied a hermeneutic approach to studying children?s meaning-making responses to reading picturebooks. His posthumous study, ?Revisiting the Relationships Between Text and Pictures? (2012), was to have been a chapter in a book on picturebooks. It is a thorough literature review of all the important scholarly work on picturebooks written over the past 32 years. Its bibliography constitutes the critical reading list for any aspiring picturebook scholar. The article is focussed on reviewing the ways academics have defined the relationships between words and pictures. Much of the discussion deals with typologies and taxonomies, some of which I have discussed in this chapter. While those frameworks provide excellent insights into the ways words and pictures relate, they rarely provide detailed explanations for the construction of those relationships. Nonetheless, to quote from one of his students, Sipe?s deep knowledge of picturebooks and curiosity clearly ?open[ed] up unforgettable spaces for our own readerly selves to dwell.? The indelible  ?experience of ?story??(Sipe 2008, p. 247) that Dr. Sipe made possible for us, contains a profound and enduring story about his teaching, as well, with so much left to learn, remember, and retell (Giola, p.73  2012).  61  3.10 Review Summary As stated from the outset, this review focuses on academic advancements in understanding how meaning is constructed in picturebooks. The reviewed works contain a rich body of insight, which is why they remain actively cited resources for picturebook analysis. However, I argue they would all have benefitted from the application of a systematic definition of the picturebook and a systematic consideration of the picturebook?s semiotic systems of meaning-making resources. Furthermore, each study needed the application of a theory of language, a theory of image, and a systematic analytic approach. Interestingly, the failure to apply a theory of language may reflect the emphasis in literature studies on how readers construe meaning and on reading strategies that address the comprehension of meaning rather than on the understanding of meaning construction.   My general criticism of picturebook scholarship to date is that attempts to label or categorize the way words and pictures interact do not provide sufficient delicacy of definition to apply to specific texts. In general, the terms reiterate what is easily visible in very broad ways. These broad ways include observations that some illustrations present parallel narrative information, some present contrary ironic information, and some particularize the narrative. Close observation of almost any picturebook demonstrates that images and words usually have most or all of these relationships. However, those relationships are challenging to describe without a framework for understanding how each modality works. Scholarly work on how meaning is constructed in picturebooks requires theories of language and image in order to sort out the complex ways that words and pictures work together.   The need for theories of language and image has been reinforced by this review of scholarship. Therefore, in this study I propose to draw on a social semiotic theory of language and several applications of it to images, to analyze my particular question. The general question of how words and pictures work is far too cumbersome; therefore, the question I will explore is how the cohesive resources of colour and rhythm contribute to the construction of coherent texts. This question addresses the sequential nature of narrative picturebooks, which is a vital but rarely addressed feature of the picturebook. Understanding 62  how visual cohesive resources work across a sequence of images and pages is foundational to ascertaining how meaning is constructed in the picturebook.    The next chapter provides a detailed description of the theoretical frameworks for this study, a discussion of SFL (Halliday, 1978, 1994), and further review of research by Kress and van Leeuwen (1996) and O?Toole (1994).  63  Chapter 4: A Theory of Language as Social Semiotic and Its Application to the Analysis of Multisemiosis in Picturebooks 4.1 Introduction The purpose of this chapter is to broadly describe Systemic Functional Linguistics (SFL), the theory of language that provides the primary theoretical foundation for my study. I also review several SFL inspired analytic works that set the precedent for its applicability to the analysis of images in multimodal texts. SFL offers a thorough accounting of meaning construction in language. My particular interest is in how it conceptualizes the flow of information in discourse to create ?text that ?hangs together? with itself and with its context of situation? (Halliday & Matthiessen, 2004, p. 87). The thrust of my discussion of SFL will be on its theories of information flow, information focus, and cohesion. For this I will also draw on the work of Halliday?s SFL colleagues. I refer to Matthiessen (2007) for his contributions to the notions of strata and language as multimodal, and I draw on the collaborative work of Hasan and Halliday (1976, 1985) for their work on cohesion because it is the most comprehensive SFL explication of the topic.  I will begin with the central notion of SFL?its stratified model of language?with special attention to the social semiotics of context and the metafunctions. These ideas are frequently cited as the source of SFL?s adaptability to other semiotic modes. Those discussions are followed by the notions that deal with information flow and information focus, the text-forming structural resources of Theme and Rheme, the information unit of New and Given, and the non-structural cohesive resources, for establish[ing] additional [semantic] relations within the text that are not subject to these limitations; relations that may involve elements of any extent, both small and larger than clauses, from single words to lengthy passages of text; and that may hold across gaps of any extent, both within the clause and beyond it, without regard to the nature of whatever intervenes. (Halliday, 1994, p. 309)  64  4.2 Stratified Model of Language SFL is a complex theory comprised of multiple systems that contribute to a comprehensive framework for analyzing language in many social contexts and across many different types of text. Its stratified model of language provides an important framework for exploring the relationships between meaning and expression. Here the term expression is related to the meaning ?expressed? in language by words (written or spoken) and to expression in other semiotic modes (such as dance) in which meaning is ?expressed? by the body in motion, in different positions, in relation to other bodies in motion, and to the facial and body gestures used to express meaning.  Halliday (1994) describes three strata of language. The first is the semantic stratum?the meaning. The second stratum is lexicogrammar?the wording which is expressed by the third stratum, graphology. Graphology is expressed by written wording or by phonology, which is the spoken wording. There is a fourth stratum that can be included if Halliday?s strata are viewed from a descriptive perspective. This fourth stratum is context, and while it is not regarded as a language stratum, it is nonetheless important in terms of looking at how Halliday?s theory accounts for language use in context. The social semiotic perspective of Halliday?s theory of language explains how expression relates to contexts of origin. Halliday wrote,  A social reality (for a ?culture?) is itself an edifice of meanings?a semiotic construct. In this perspective, language is one of the semiotic systems that constitute a culture; one that is distinctive in that it also serves as an encoding system for many (though not all) of the others. (Halliday, 1978, p. 2) 4.2.1 Context in Halliday?s Theory of Language The role of context is central to Halliday?s theory of language. When he describes his theory as social, he means ?social? in terms of the social system and as approximately synonymous with culture (Halliday & Hasan, 1985). His intention has been to ?relate language primarily to one particular aspect of human experience, namely that of social structure? (Halliday & Hasan, 1985, p. 4).   65  The meaning of context in Halliday?s theory is not only the physical or material context, however relevant to the situation that might be. For him, context is the semiotic context; that is, the social activities and relationships that influence the exchange of meanings. Whether real or fictional, meaning exchanges between people that involve language occur in a physical setting or context, and all those exchanges also have a particular social semiotic context. So, for example, in a school classroom during an instructional session the teacher and a group of students may engage in instructional activities led by the teacher and involving student/teacher relationships and the use of language for educational instruction. Within the same physical setting, and simultaneously, there might be a number of entirely different language exchanges going on between students that might be construed as informal. Each might also involve a range of different relationships between the students, running the gamut from that of a bully picking on a victim, to bosom buddies exchanging personal anecdotes, or to an older student helping a younger student. The language used in each of these relationships represents a different social semiotic context.    By describing ?language in a social semiotic perspective,? Halliday modifies the definition of semiotics from ?the general study of signs? to a ?study of sign systems,? for he sees semiotics as ?the study of meaning in its most general sense? (Halliday & Hasan, 1985, p. 4). From this perspective, in Halliday?s view, language or linguistics is one kind of study of one kind of meaning among many other modes of meaning in a culture. He goes on to define culture as ?a set of semiotic systems, a set of systems of meaning, all of which interrelate? (Halliday & Hasan, 1985, p. 4).  Halliday?s social semiotic theory of language describes the complex relationships between social context, meaning, and expression. One of the reasons Halliday?s work has been adapted to other semiotic modalities is because of his perspective on, and analytic use of, social context as central to meaning. Halliday believes that the notion of language use in context?the importance and influence of context on language use?is shared by other semiotic modes. The significance of the social context of language in use is woven through Halliday?s theory of language and SFL and its metafunctions. He argues that the study of texts goes hand in hand with the consideration of context and places emphasis ?on the situation, as the context in which texts unfold and in which they are to be interpreted? 66  (Halliday & Hasan, 1985, p. 5). Using this view of texts and my interest in multimodal texts, the descriptive model with the fourth stratum of context provides a way of theorizing other semiotic modes and multimodality, as Matthiessen (2007) has done in ?The Multimodal Page.? (Royce & Bowcher, 2007)  4.2.2 Strata and Context as Central Resources for Adaptations of SFL by Other Authors The notions of strata and context, central to Halliday?s theory of language and SFL, are the features that suggest and inspire the application of his work to multisemiotic analysis. In the following discussions, Matthiessen, Kress and van Leeuwen, and O?Toole each note that SFL?s social semiotic foundation is the key to their use and adaptation of SFL to other semiotic modes.  In Matthiessen?s multisemiotic model the three strata are (a) the overarching context of experience; (b) the semantic stratum, which is the stratum of meaning construed from the context above; and (c) the expressive stratum where meaning is realized by different modes such as language, image, musical sound, and dance, to name a few. Matthiessen?s model addresses the relationships between linguistic meaning-making and meaning constructed by other modalities. The model also illustrates the importance of context, a key notion in adapting SFL to a visual semiotic analysis (Matthiessen, 2007).  In Reading Images, Kress and van Leeuwen (2006), the pioneers in adapting SFL to other modalities, explain that one of the key notions they have taken from SFL is its social semiotic foundation and emphasis on the socio-cultural context of communication. Similarly, O?Toole (1994), another pioneer in adapting SFL to other modes, describes Halliday?s SFL as theorizing ?the dynamic interplay between the code of language and the social situations in which it is used? (p. 216). O?Toole (1994) goes on to point out that although Halliday?s work concerns language, Halliday ?makes clear that the same dynamic relation between the code and its instances of use, or texts, obtains for any semiotic system? (p. 216). 67  Citing Halliday?s work as the inspiration for their social semiotic visual grammar, Kress and van Leeuwen (2006) state that their work is about signs, and particularly about ?sign-making.? In the context of this work, they write:  we see representation as a process in which the makers of signs, whether child or adult, seek to make a representation of some object or entity, whether physical or semiotic, and in which their interest in the object, at the point of making the representation, is a complex one, arising out of the cultural, social and psychological history of the signmaker, and focused by the specific context  in which the sign maker produces the sign. (p. 7)       I find that for this research project SFL and some of its theoretical adaptations to other modes useful resources for the analysis of multimodal picturebooks. SFL is especially helpful in theorizing the relationships between language strata and for illuminating the functional work of language and other modes in a social semiotic context.  4.2.3 Realization and Instantiation In SFL two key ideas related to the notion of strata?realization and instantiation?are used to describe the relationships between the language strata. Realization is the term for the particular work that lexicogrammar or wording does to translate meaning into wording. It is also the term for the additional work that phonology or graphology do to put that wording into sound or graphic form. The realization of meaning as wording is realized, in turn, by the expressive stratum, the stratum of phonology, or graphology. Hasan describes the complex notion of realization as being characterized by two directions of relationships: meanings activate grammatical features, and certain configurations of grammatical features in a particular context will cause us to construe certain meanings (G. Williams, personal communication, April 20, 2013). At a more abstract level, meanings realize contexts or contexts of situation as they are technically called in SFL. Informally, this claim means that we ?construe? a context from the meanings and wordings being used, and that we can participate in contexts to the extent that we can ?activate? relevant meaning-wordings. Stated in this informal way the claim appears quite simple, but in Halliday?s theory the relationship of realization is extensively elaborated and nuanced to become a powerful resource for 68  predicting and systematically analyzing relationships between contexts, meanings, wordings, and sounds/written symbols. The second term, instantiation, refers to the particular instances of a selection of features from language?s potential to form a text. A text is a specific configuration of wordings and sounds or graphic elements selected to realize meaning in a context of situation. Halliday proposed instantiation as a concept to model the relationship between a particular text (?parole? in Saussure?s terms) and the semiotic system(s) on which the text draws (?langue? in Saussure?s terms). The importance of this concept for my study is that it provides a clear basis for a systematic analysis of individual picturebook texts because the analysis can be grounded in a detailed, fairly comprehensive description of the semiotic systems, both verbal and visual, of which the texts may be interesting instances.   Importantly for my study, in Halliday?s model between the system of language and the instance lies the concept of register. The key idea is that meanings cluster over time because of the way people do things in and through language in various types of context: these habitual ?clusters? of meanings are called registers. Contexts include, as previously mentioned, both the social activities and the relationships that are implicated in the exchange of meanings. Contexts, in the sense I use here, can be identified from the language used (technically, ?construed? from the language). In turn, contexts are brought into existence through the selection of language features and other semiotic modalities such as visual images and gestures. To illustrate informally, at a baseball game the following utterances might be heard: ?Let?s see that southpaw find the sweetspot? or ?Oh no, this rubber is done with that shoestring catch, they needed a tater!? These wordings make sense in the context of a baseball game but outside of that social context they sound like nonsense. The connection between registers and contexts of situation is important to understanding the shared semiotic labour of language and images in picturebooks. 4.2.4 Metafunctions  Halliday?s notion of metafunctions addresses the ways language means and defines the functional nature of SFL (Halliday, 1976, 1994; Halliday & Matthiessen, 2004). The three 69  metafunctions are: the ideational (the logical and the representational), which has to do with such activities as doing and saying, and the language of reflection; the interpersonal, which addresses ways of meaning that enact relationships such as turn taking and ways of addressing others in discourse or expressions of a speaker?s certainty about propositions; and, the textual metafunction, which Halliday describes as the weaving of disparate language features into coherent text (Halliday & Hasan, 1985). Halliday (2004) explains that language provides a theory of human experience because it literally can ?construe human experience? (p. 29). The social semiotic nature of communication, particularly language, is realized by the simultaneous work of these metafunctions. As a preface to describing Halliday?s foundational notion of text, I want to acknowledge the complexity of his theory and note that here I am only able to give a very broad characterization of some key features from what is now a large body of theoretical and descriptive literature.4  4.3 The Nature of Text According to Halliday and Hasan (1985), a text, whatever its length, has unity. It can be as short as a word or as long as the longest novel, play, or serial narrative (Halliday, 1978). Text is a unit of meaning expressed in spoken or written word. Its structure is strongly influenced by its context of situation, which as Halliday explains, is: the contextual configuration of field, tenor and mode. Field refers to the nature of the social action: what it is the interactants are about. Tenor refers to the statuses and the role relationships: who is taking part in the interaction. Mode refers to the rhetorical channel and the function of the discourse: what part the text is playing. (Halliday, 1994, p. 390)  The context of situation has a critical bearing on the perception of coherence, since knowledge of a context provides the resource for recognizing and predicting patterns of                                                  4 One wide-ranging but accessible overview of SFL is available in Geoff Williams? PhD dissertation (1995) ( 70  language. For example, there is the language associated with shopping, which is often brief and elliptical. In the situation between a customer and a clerk in a shop with a particular selection of goods, a typical exchange might consist of ?How much?? and a reply of ?Two-fifty? followed by ?And this?? To most western shoppers the clipped speech is immediately understood. If that verbal text was on a picturebook page along with an illustration of a small child in a shop with a caregiver, then the image conveys the material situational setting while the printed words convey the social context of situation. The example of shopping talk as a context of situation associated with a register of language can illustrate how the visual depiction of the shopping material context can support the verbal ?shopping? register talk in the book. The reader can construe the material situational setting from the visual, while simultaneously construing the register of the language being read aloud to her, and thus the social semiotic context of situation.        A picturebook text, in the semiotic sense, is an instance of cultural meaning making. It is an example of a particular context of situation, one in which an author and/or illustrator addresses an audience of young children and their caregivers about a topic within a child?s range of experience. Within such a text, there is typically a representation of several further contexts of situation, built up around specific social activities and inter-related as a narrative sequence (such as finding an injured bird, transporting it home, caring for it, transporting it back, releasing it). Joint reading of the picturebook by a child and a caregiver will form a further, quite different, context of situation, a context for the exchange of meanings, which might take place in a wide variety of material contexts (e.g. in bed prior to sleeping or on a lounge after the midday meal). Material contexts or settings, called by Hasan (1996) ?material situational settings? are the material environments in which meanings are exchanged but they are distinct from semiotic contexts of situation (p. 214). The unity of text is produced by the combination of its structure, which is related to its context of situation, and by its texture, that is, its meaning relations realized by cohesive ties. As previously explained, text is brought about by two processes, realization and instantiation (Halliday, 1994). Realization expresses the relationship between strata. Thus, a meaning from the semantic stratum, content, is realized by instantiations of lexicogrammar. Halliday describes grammar as a system for making meaning, a network of interrelated meaningful 71  choices that ?lies behind the text? (Halliday & Matthiessen, 2004, p. 33). One sub-system of grammar, together with certain other resources of wording, is particularly important for weaving a text together to form a unity. These are known as textual resources, and realized in those instantiations by drawing on textual resources. In SFL, those resources are Theme and Rheme, and the Information Unit.     4.4 Textual Resources of Theme and Rheme and the Information Unit of New and Given Halliday?s notion of a system of information provides a theoretical resource for analyzing the coherence of the narrative information flow that is constructed by structural and non-structural resources. The structural system consists of Theme and Rheme and the Information Unit. According to Halliday (1994), the system of Theme contributes to textual coherence as follows:  The Theme functions in the structure of the clause as a message. A clause has meaning as a message, a quantum of information; the Theme is the point of departure for the message. It is the element the speaker selects for ?grounding? what he is going to say. (p. 34)  The Theme as a point of departure is a point of significance, while the Rheme is everything that follows, or the rest of the message in the clause. The patterning of Theme and Rheme carries a text forward, creating information flow.  The Information Unit contributes information focus. Halliday (1994) describes the Information Unit as ?made up of two functions, the New and the Given? (p. 89). He writes: information in this sense . . . is the tension between what is already known or predictable and what is new and unpredictable. . . . It is the interplay of new and not new that generates information in the linguistic sense. Hence the information unit is a structure made up of the two functions, the New and the Given. (p. 89)  72  He further states, ?[s]imilarly what is treated as non-recoverable (New) may be something that has not been mentioned; but it may be something unexpected, whether previously mentioned or not. The meaning is: attend to this, this is news? (p. 91). With a view to text construction, the important work that New and Given does is to structurally position information in a way that the New, its variation and/or its contrast, is brought to the reader?s attention by its having been given information focus. In a spoken text the position of New in the linear sequence is marked by tonic prominence; in a visual text it can be marked by its visual position, salience, or framing. Information focus is essential for calling attention to the New whether in a spoken or written language or in a sequential visual text. As Halliday (1994) writes: At any point of the discourse process, there will have been built up a rich verbal and non-verbal environment for whatever is to follow; the speaker?s choices are made against the background of what has been said and what has happened before. The environment will often create local conditions which override the globally unmarked pattern of Theme with Given, New within Rheme. (p. 300) In some sequential visual texts the New is a variation in the previously visually depicted environment. Similarly, as we physically move through space and function in the visual physical world, we generally take in the overall environment and the position of each item in it (Gibson, 1986). What we notice are the salient variations, especially the unexpected variations, such as the New element or activity in the environment (Gibson, 1986). The flow of information and the work of information focus are complicated in picturebook texts due to the different affordances of language and image, and by a picturebook?s materiality. In the picturebook, the verbal narrative will have information flow with a patterning of Theme and Rheme, and information focus through the patterning of New and Given characteristic of its particular register. However, in the picturebook, the spatial nature of images changes the patternings of Theme and Rheme and New and Given. The linear nature of language affords particular positions for Theme and Rheme and New and Given. However, the spatial nature of the visual requires that Theme and Rheme and New and Given must be highlighted in some manner to draw attention and give prominence to them. In some 73  kinds of images, particularly in single images, some elements can be given focus by position, salience, and/or framing in the conventionalized ways Kress and van Leeuwen (2006) noted in their work on advertising and text book illustrations. However, their concepts of New and Given do not always apply to sequential visual narrative texts, particularly those narratives in picturebook formats. 4.4.1 Challenges to the Use of Structural Resources in Visual Modalities Theme is usually the first element in a language clause and indicates the main orienting idea; but, the spatiality of images makes it more challenging to position and attract attention. Theme is more challenging to express in a single image than in, for example, a three dimensional installation or in a sequential visual text where Theme may be established in the entrance to an exhibit or on the cover and title pages of a book (Martin & Stenglin, 2007). In a single image it can be difficult to visually isolate Theme. However, in a single multimodal page the printed verbal text may express the Theme with the image, thus enhancing the verbal Theme and Rheme. In these cases, the Theme can be give prominence by any or all of the resources of salience, colour, and framing. Book covers and title pages often offer good examples of this kind of multisemiotic realization of the Theme. On the printed page, Theme and Rheme can be given prominence by the use of larger, bolder, or differently coloured typography. Furthermore, the use of graphic prominence is a potential resource for constructing some of the written text?s rhythm.  The spatial nature of visual texts challenges the location of Theme and Rheme. In language, Theme may also be marked verbally and when spoken given some form of tonal variation. The linearity of a verbal clause or sentence affords a clear prominent initial location and speech affords tonic prominence. However, in visual texts, although it is possible to give an element visual salience and prominence in the composition, it is challenging to position an element or activity such as the Theme or the New so that it is the first one seen by viewers. Another particular challenge within a visual text is that, for example, in narrative picturebooks, the visual text usually depicts a material environment, such as a room within a home or building, in which it may be physically and or logically impossible to place the Theme in the position as the first visible element.   74  Another challenge to giving Theme and New a prominence in the narrative picturebook for young children is created by the convention of depicting the main character in almost every image. However, this very convention can sometimes offer opportunities to render the Theme, if Theme is the main character, as an immediately recognizable feature. This is because the main character is given a high degree of salience to facilitate easy following of the character across the pages of the text. A good example of this is the presence of Max in every image except the cover and endpapers of Where the Wild Things Are (Sendak, 1963).  The sequential nature of narrative picturebooks sometimes offers opportunities for Theme to be expressed in the first image or panel of a sequence. For example, in Where the Wild Things Are, in the first narrative illustration in which Max is constructing a tent, he is given salience by his position, brightness, and size in the tightly framed image. The image conveys Max as the main feature. However, the Theme is more complex and requires the collaborative work of both verbal text and image. The Theme is ?The night Max wore his wolf suit and made mischief of one kind and another? (Sendak, 1963). Sendak uses two double spreads to convey the verbal and visual Theme. The verbal Theme instantiates the visual activity as something specific that occurs at a specific time. In this example, both verbal and visual modes articulate the Theme. Although a character may be depicted with one kind of high salience throughout the visual text, Themes involving other narrative elements and actions can be given prominence by the use of framing, size, scale, and colour. Matthiessen (2007) notes the use of colour in rendering thematic prominence in images. Chapter Five explores the role of colour in cohesion and as a resource for constructing thematic prominence and discusses the patterning of colour deployments and how they support both Theme and the integrating code of rhythm. In sequential visual texts, the Theme may be located or framed as the first in a sequence of images. Also, in a multimodal picturebook, the image appears with a printed text or title, which can carry a Theme so that Theme is collaboratively conveyed by both image and written text and its typographic affordances. An example of this can be seen in Sendak?s In The Night Kitchen (1970) in which the protagonist, Mickey, first appears framed on the cover under the title. Next he is framed in a circle on the first half title page, again with the title so 75  that in both cases the Theme of ?Mickey in the night? is articulated in both verbal and visual modalities.  Changes in the tradition of designing the cover and front matter of a printed book in support of the book?s major theme present a challenge to the placement of Theme. The change is due to a progressive reduction in the page length of typical picturebooks and is manifest in the frequent use of the cover, endpapers, and title pages to wordlessly introduce the narrative by visually establishing the setting and some key characters.  Due to the challenges of depicting Theme visually in a picturebook, it is my view that in looking at the picturebook as an integrated whole multimodal text, the collaborative work of verbal and visual semiotic systems conveys Theme. An example of that kind of collaborative work can be seen in the first spread of How to Heal a Broken Wing (Graham, 2008). In the verbal text, ?high above the city, no one heard the soft thud of feathers against glass,? the verbal Theme is ?high above the city? (Graham, 2008). The 2-page illustration is a full colour 3-sided bleed, bird?s eye view of a cityscape. Most of the image is rendered in soft greys except for the tallest skyscraper that has a reflection of white clouds against a light blue sky. The falling bird, a tiny speck of grey, is positioned just below the top of the highest building. The narrative activity of the bird falling ?high above? together with the prominence of scale created by the stark contrast of its tiny size to the huge cityscape attracts attention to the building as being high above the city. Attention is drawn to the visual rendering of the Theme ?high above the city? (Graham, 2008). The building is given added salience by the bright blue sky reflected in its top floors. The rendering of the falling bird is part of the multisemiotic Rheme. It would be difficult to understand what the story concerns without the collaborative work of the verbal text because the illustration is rendered with a limited colour palette of soft greys, beiges, and only the tiniest bit of a bright light blue that altogether conveys an expansive city scape. The next page of illustrations also shows the verbal text providing the Theme for three illustrations of the bird falling. Since the bird is the only item in each panel without the verbal text, the Theme might be construed to be the bird when the verbal text instantiates ?No one,? as the Theme for the line is ?No one saw the bird fall? (Graham, 2008). 76  Matthiessen (2007) sees the integrated work as:  coordinated in the process of making meanings and integrated (more or less tightly) into a Gesamtkunstwerk. The nature of the division of semiotic labour depends on the context: Different recurrent settings of the field, tenor, and mode parameters of context (i.e., different situation types) are associated with different registers, each of which is characterized by some particular division of labour among the semiotic systems listed earlier. (p. 25) The division of labour is highly visible in some narrative picturebooks with spare verbal texts, where the material contexts of situations can be visually elaborated in anticipation of young readers? abilities to construe the register and context of situation from the cohesive multimodal whole. For example, in Raschka?s Caldecott Honour book Yo! Yes? (1993), the very spare text of 34 words (20 different ones) is visually elaborated by the expressive body and facial characters of the two main characters, two shy young boys attempting to get acquainted. So, while there are challenges to visually depicting Theme in multimodal picturebooks, the division of semiotic labour characteristic of the format provides the affordance of using verbal and visual resources to express Theme.     4.5 Cohesive Resources for Constructing Coherent Texts The rationale for looking at the theory of cohesion in verbal texts to explore cohesion in the visual modality is based on the notion that in a multimodal text it is important to understand how cohesion is constructed in each of its modalities as a background for understanding if there are ways that the modalities share cohesive work. The reason for borrowing notions of cohesion from language is inspired by the comments of Halliday and Matthiessen (2004) on the essentially multimodal nature of language. Further rationale for claiming that modalities share cohesive work is provided by Matthiessen?s (2007) work on the multimodal page and his view of the semiotic division of labour characteristic of multimodal texts.  A key resource for the analysis of text is the notion of texture, which provides another source of textual unity (Halliday & Hasan, 1985). Texture is constructed, inter alia, by the meaning relations created by cohesive ties between messages. Cohesive ties are resources for realizing 77  semantic relations between individual elements and messages. Clear semantic ties are critical to a reader?s perception of coherence and recognition of the text as a unified whole.   Semantic relations are the basis of cohesion (Halliday & Hasan, 1985). Cohesive devices can be lexical and/or grammatical. In fact, lexical cohesion supports grammatical cohesion and visa versa. In considering cohesive resources, it is helpful to sort them as types of devices and by their tie relations. There are two key ways of meaning connected by a link between elements: by the identity or by the similarity of the elements.  Table 1 Visual Cohesive Resources Colour as a Cohesive Resource  ? Colour as a structural cohesive resource ? Limited colour palettes of two or more colours ? Single colour as an addition to the black of the print over multiple pages ? Single colour configurations ? flows ? ribbons ? Colour as a non-structural resource ? Colour associations ? Sense relations of similarity, hyponymy, and meronymy ? Colour identities ? Colour environments function as identities and as associations and references  ? Backgrounds ? embedded material situational settings ? embedded semiotic contexts of situation Repetition as a cohesive resource ? Repetition is a dynamic that must be realized by its being made visible by colour, line, shape, texture  78  Table 2 Verbal Cohesive Resources  Co-referential ? Devices of reference: pronominals, definite articles ? kinds of reference: ? endophoric: refers to something within the text ? exophoric: refers to something outside the text but within the context ? anaphoric: refers to the preceding text ? cataphoric: refers to the following text Co-classification: identity by class  ? Use of substitution or ellipsis Co-extension: the sense relations of ? Synonymy: almost identical in meaning, e.g. woman and lady ? Antonymy: opposites such as black and white, polite and rude ? Hyponymy: relationship between a general class and its sub-classes ? Meronymy: a part of a whole such as a leaf and a branch, a flower and a stem ? Repetition: each iteration builds the ties and chains. Parallelism Note. Adapted from Cohesion in English, by M.A.K. Halliday and R. Hasan, 1976, and Language, context, and text: Aspects of language in a social-semiotic perspective, by M.A.K. Halliday and R. Hasan, 1985.    Cohesion also depends upon context. If the reference is to something within the text, it is an endophoric reference; however, if it is to something outside the text, it is exophoric. The exophoric tie relies on an understanding of the context.    The issue of context of reference in or outside a text is interesting in regard to whether or not a semantic cohesive relation, such as between a verbal item and something in the visual text, is exophoric. The question about the wholly integrated multimodal narrative picturebook text is whether or not verbal and visual texts should be regarded as one text. If it is seen to be one whole text then cohesive relations that cross modalities should be considered endophoric. Repetition of the names and visual depictions of characters in children's picturebooks, when considered together, construct a sequence of cohesive relationships in both modalities and across modalities. A child?s multimodal experience of these picturebook texts generally involves following the characters and visually reading the physical gestures and contexts while listening to the verbal text that is naming the characters and particularizing the depicted activities. 79  Matthiessen (2007) argues that language is multimodal, citing young children?s multimodal protolanguage, and discussing Halliday?s (1978) similar assertions as evidence of language?s essentially multimodal nature. Young children?s protolanguage is multimodal; that is, they communicate multimodally with the spoken word, body gestures, facial gestures, and with things at hand.  Both Matthiessen (2004) and Halliday (2004) argue that the multimodal nature of language in use--particularly in young children?s early communication. Not only do children draw from both verbal and visual modes in their everyday discourse they do so as they view the images in books while being read to. Young children experience cross-modal ties constructed by images of the characters they see on the page with the names and pronouns of the characters they hear read to them. Similarly, they may construe cross-modal cohesive relationships between images on a screen and the sounds of language. Related work has been done on cross-modal ties in film by Bateman (2008). Film is a sequential multimodal form and therefore offers a good precedent for the claim that cross modal ties exist. Also, in my view, Matthiessen?s (2007) case for the division of semiotic labour in multimodal texts applies to cross modal cohesive ties in picturebook texts. Thus, returning to the issue of endophoric cross modal ties, a cohesive tie crossing modalities can be considered endophoric as long as it is referring to something in the other modality. This is not to exclude the possibility of exophoric ties in a picturebook, particularly those in highly metafictive and intertextual works.   Halliday and Hasan?s (1985) work on cohesive sparely worded poetry texts makes the point that those minimal texts often rely on co-extensional ties. Halliday and Hasan use the example of silver and gold. What they have is a tie that has a sense relationship to something outside the text, such as metal. However, since they are both within the text, the tie between them is endophoric. Another example is a branch and a leaf, both are parts of a tree or bush. Picturebook texts are often spare and poetic and frequently contain cohesive ties that have sense relationships, which are the relations of synonymy (identical or near-identical), antonymy (opposite), hyponymy (in a class), and meronymy (part/whole) (Halliday & Hasan, 1985).  80  In addition to the aforementioned kinds of cohesive ties, there is also basic repetition: the repetition of the same lexical unit, which, as Halliday (1994) noted, is the most direct form of lexical cohesion. Repetition contributes to the construction of continuity, cohesion, and coherence. It also constructs emphasis, foregrounds the unexpected, and develops narrative elements. As will be discussed later, repetition is frequently used across modalities in picturebooks for young children as a means of helping them identify with and follow characters through the narrative. Visual repetition utilizing a range of text-forming features will be described in detail in Chapter Five. In section 4.7 I return to the question of cohesion and discuss the specific issue of how cohesion and coherence are achieved in a multimodal text. However, I first address the issue of modelling analytic frameworks for multimodal texts on the systemic functional linguistic model. 4.6 Adaptations of SFL to the Analysis of Multimodal Texts The challenges in adapting SFL to other modes are, first, in conceptualizing how the linguistic metafunctions represent things and the social activities and the social relationships related to them and; second, to factoring in the particular textual affordances of another semiotic mode.  For my work I considered the precedents set by Kress and van Leeuwen?s adaptation of SFL?s textual metafunction to the analysis of image, O?Toole?s creation of a framework for the analysis of painting, sculpture and architecture, and, as indicated above, Matthiessen?s work on multimodal pages addressing the division of semiotic work among semiotic systems. Kress and van Leeuwen (2006) named the analogue to the textual function of language the ?compositional? function of images. For example, in their work on advertising they see the compositional metafunction drawing together the simultaneous work of the ideational and the interpersonal, as the textual metafunction does in language. Matthiessen (2007) sees the organization or compositional work varying according to registerial nature, ?which means that the variation correlates with different contextual settings? (p. 27). This has to do with the fact that multimodal texts,  81  appear in a wide range of registers that occur in different situation types characterized by different combinations of field, tenor and mode settings. These situation types in which the registers of printed pages occur obviously fall within the overall context of culture of a society, with certain general restraints placed on field, tenor and field. (Matthiessen, 2007, p. 270)   Further, Matthiessen explains that ?the overall organization of the page is paralinguistic rather than linguistic: the layout frames the texts and the images? (p. 25). He is referring to a single multimodal page. Moreover Matthiessen (2007) describes the ?visual paralanguage? system as comprised of typography and layout.  Matthiessen?s use of the paralinguistic and his recognition of the influence of registerial natures inspires me to view the picturebook format as having a similar kind of paralinguistic organization. Earlier, in Chapter One, I described typography as one of the systems at work in constructing a picturebook. I find Matthiessen?s model a helpful construct for considering how evolving conventions related to the system of typography and layout in a picturebook influence its registerial nature as a format. Looking at the textual systems of a format offers a way of understanding its organization and semiotic division of labour. I believe that Matthiessen?s notion of the semiotic division of labour particularly supports my analytic approach. Although I prefer to consider typography as a separate system from layout, the overall notion of the paralinguistic organization of the multimodal picturebook format works well and is vital to the consideration of its registerial nature, which evolves out of both technological and social changes but also in relation to its family of ever evolving narrative sequential arts. The picturebook format has great kinship with theatre and opera (among the many other previously mentioned members of the sequential narrative visual arts family). Organization of the picturebook as a sequence of scenes is much like the organization of a play or opera. I believe the registerial nature of the picturebook format is close to the registerial nature of the play format. Of course, within the narrative of a picturebook or a play there are other social registers, various contexts of situation, and material contexts that influence the organization of both formats. The conventions associated with the registerial nature of a scene-by-scene 82  format such as the picturebook or a drama, have a vital influence on how the sequential narrative is organized to unfold before a viewer or viewers. Kress and van Leeuwen provided key contributions to visual textual analysis with their notion of composition as the organizing dynamic for single composition images, and their view of rhythm as the major organizing dynamic for integrating sequential visual texts. Both Kress and van Leeuwen?s Reading Images (2006) and O?Toole?s The Language of Displayed Art (1994) contributed important pioneering insights into the grammar of image. However, neither of these works addressed sequential visual narrative texts in depth. O?Toole looked at episodes in narrative paintings, but not the nature of sequential visual narrative typical of books or films.   A recurring critical concern about using a linguistic theory for the analysis of another modality stems from the recognition that language is predominantly linear while image is spatial. Kress and van Leeuwen (2006) believe that semiotic systems, and in fact all communication, stem from social environments and that they express ?first and foremost social meanings? (p. 20). Also key is their acknowledgement that a primary SFL resource is the model of metafunctions. In their use of metafunctions they also ?borrow? what they term ??an attitude? which assumes that, as a resource for representation, images, like language, will display regularities, which can be made the subject of relatively formal description? (p. 21). Further, they state ?that a semiotic mode must . . . be able to represent aspects of the world as it is experienced by humans . . . be able to project the relations between the producer and the receiver of a sign . . . and have a capacity to form texts? (p. 43). In a language text, there are particular resources for constructing that unity and coherence of the verbal text, namely, the cohesive resources. In multimodal texts, Kress and van Leeuwen (2006) make the point that it is the visual modality that provides the cohesion and gives the multimodal text unity. Looking at a variety of multimodal print advertising texts, Kress and van Leeuwen concluded that the general cohesion of the text was the work of visual composition/layout and colour. However, I argue that these visual qualities are not always the primary cohesive resources in play in a multimodal sequential text because both the 83  meaning and the selected multimodal form of expression influence the choice of cohesive resources for a particular text.   O?Toole (1994) observed that different kinds of texts have distinctive kinds of unity and cohesion, a view relevant to my study. Creators of multimodal texts draw on their knowledge of each modality?s resources, utilizing the potential for selecting the best resource from each modality for a particular meaning. For example, in a picturebook, the best resource for conveying what a character looks like is visual. Whereas the best resource for conveying what that character is saying or thinking will likely be selected from the verbal systems. Kress and van Leeuwen (2006) express a similar point of view and look for regularities in the meaning making particular to specific multimodal formats. O?Toole (1994) and Kress and van Leeuwen (2006) acknowledge the foundational notion that there is variation in the relationships between forms and functions and highlight the potential for differences in relationships between metafunctions and various semiotic modes and their forms and/or formats. Matthiessen (2007) additionally notes that variations in these relationships will also be associated with registerial differences. For example, within the category of picturebooks, a number of narratives draw on the traditions of nursery tales, fairy tales, naturalistic stories, homely tales, and futuristic stories, to name a few.   The fundamental relationship between form and function is one of the prime reasons for having defined the picturebook format in detail early in this study. Analysis of a picturebook?s features of format must account for the relationships between its material and semantic literary properties. The textual function integrates the material format with the semantic content as it draws on cohesive resources from the modalities involved. Variations in these integrative relationships will occur in response to registerial differences, as Matthiessen noted (2007). 4.7 The Cohesive Work of Composition The resource systems of composition are central to discussions of the textual construction of image because they are most often the key cohesive resources at work. Kress and van Leeuwen (2006) discuss image composition in terms of how it textually relates the 84  representational or ideational meanings to the interactive or interpersonal. They define the following systems: (1) Information value. The placement of elements (participants and syntagms that relate them to each other and to the viewer) endows them with the specific informational values attached to the various ?zones of the image?: left and right, top and bottom, centre and margin.  (2) Salience. The elements (participants as well as representational and interactive syntagms) are made to attract the viewer?s attention to different degrees, as realized by such factors as placement in the foreground or background, relative size, contrasts in tonal value (or colour), differences in sharpness, etc. (3) Framing. The presence or absence of framing devices (realized by elements which create dividing lines, or by actual frame lines) disconnects or connects elements of the image, signifying that they belong or do not belong together in some sense. (p. 177) 4.7.1 Information Value  Kress and van Leeuwen (2006) discuss information value, the visual corollary to the information units New and Given, in terms of the position of elements in a composition. They acknowledge that their ideas pertain to Western cultures with languages that are read from left to right. They also acknowledge use of Arnheim?s (2004) principles of design as precedents for their conclusions about the information value of position within a composition. For the most part, their ideas pertain to single page compositions and not to other multimodal forms such as a sequential visual narrative. This is because within a single multimodal composition the information value, the New and Given?what is new to the viewer and what is known?can be designated by a resources such as salience and position within the composition of that the single entity. However, in sequential visual narrative forms most often everything in the first illustration is New. The second illustration in the sequence might be all Given because it might be an exact repetition of everything in the first scene. In this situation the New will be in the verbal text.  85  There are further limitations in the application of Kress and van Leeuwen?s notions of information value that Thomas (2009) notes when he describes locating those values on a toothpaste package and the restrictions imposed by specific forms. Two key issues arise about information value. First, the value of particular positions varies in different forms and formats. Second, and this is especially true in sequential narrative arts, the position of elements must also be determined by the physical location that makes sense in the context of the sequence of narrative events. For example, if the entrance to a room is on the left side of the page and the narrative calls for someone inside to answer the door to a New visitor, that New is located on the left side of the page. Of course, it would be possible to set up the scene from an outdoor perspective so that the door can open on the right side of the page with the New guest appearing on the right side of the page, which is the conventional location for the New.     When Kress and van Leeuwen (2006) looked at advertisements they focused on the point of view and meanings in a particular genre. They looked at the New and Given in the positions of left and right and observed:  For something to be Given means that it is presented as something the viewer already knows, as a familiar and agreed upon departure for the message. For something to be New means that it is presented as something which is not yet known or agreed upon by the viewer, hence as something to which the viewer must pay special attention. (p. 181)  They concluded that the New is usually realized on the right side of a page or on the right page of a double spread, while the Given is realized on the left side or left page. Most of their work looked at magazine advertisements and charts where that principle applies fairly predictably. As previously noted, Kress and van Leeuwen?s findings do not hold true consistently for the location of New and Given in picturebooks. A key difference stems from the fact that composition constructs cohesion in a single page multimodal text, while rhythm, as Kress and van Leeuwen acknowledge, constructs visual cohesion in sequential visual texts (2006). In the narrative sequence of the picturebook, the placement of key characters, elements, or actions may need to follow a physical logic realized on previous pages 86  necessitating the New to be placed in another location. To compensate for a different location, the system of salience may sometimes be a resource for constructing New to attract attention.   The picturebook has the physical form of bound pages that are read and turned one at a time and thus present challenges to the determination of information value. This determination must factor in the point of view of the image and the visual environment that has been rendered both on other pages and possibly created in the viewer?s mind. In the case of spare backgrounds, there is the physical logic of the placement, previous narrative content, and the most likely narrative event to take place next. Patterns of information value position may be found within specific texts and within specific narrative genres, but in spite of the array of variables in narrative visual texts, there appears to be a recurring principle in play. There is frequent use of the left side when all the variables allow for that position of Given and for New to be positioned on the right. In conclusion, in single multimodal compositions, patterned expressions bearing information value appear to play an important structural and textual role. However, they do not as predictably play these roles in forms that contain multiple or sequential sets of compositions, such as in the sequential narrative picturebook in which the integrating code of rhythm is at work (Kress & van Leeuwen, 2006). 4.7.2 Salience and Framing  Kress and van Leeuwen (2006) discussed the system of salience as a resource for   composition in visual texts. They defined salience as a dynamic constructed by the positioning, size, degree of prominence, and relationships between elements or of an element in the page composition that attracts the viewer?s immediate attention. What is salient may be the thing that is New or the thing that has the highest information value, but it is almost always the thing first noticed.  In images, salience is a powerful resource for attracting attention and constructing cohesion. In language, information value is a key structural dynamic in messaging because it is coding what is New, or what should be attended to. As a reminder, New is coded by tonic 87  prominence, which attracts a listener?s attention. However, in the case of images, highlighting is not necessarily always a matter of drawing attention to New information. Rather, attracting attention may be a means of creating a salient beat as a part of a pattern of salient beats that serves an integrating function over a sequence of pages. Salience may be a resource for drawing attention to the iteration of a key narrative character or element across the sequence of the visual narrative that constructs continuity and cohesion.     In addition to position in the composition, salience and framing are resources for rendering New and Given in picturebook texts. The page turning nature, along with the integrating code of rhythm, may shift the patterning of New and Given locations. Also, there may be multiple expressions of New related to different characters, elements, episodes, and even secondary narratives rendered on the page. Some of these expressions of New can be constructed by the resource of salience and some by framing. As Kress and van Leeuwen (2006) noted, framing is a resource for organizing composition. It can be either formal or informal. In picturebooks, formal framing is rendered with lines or borders creating individual separate panels or framed illustrations. Informal framing takes place within a composition utilizing elements in the composition such as windows, doorways, or tree trunks. Informal framing may be used to foreground a character or action, to isolate, to foreshadow, to emphasize, or to do other work in service of the textual content.  4.7.3 Typography and Colour as Additional Integrating Systems I consider typography and colour to each be integrating systems. However, colour together with information value, salience, and framing provide the resources for the textual integration work of composition and rhythm in both single integrated texts and sequential texts. In Chapter Five I present a detailed case for colour as an integrating system. I concur with Thomas (2009) that typography has been overlooked in multimodal studies as a key integrating resource. He noted that more attention must be given to the role of typography in constructing cohesive multimodal texts. Typography historian, Bringhurst (2005) also acknowledges typography as a cohesive, longstanding resource for page and book design. Van Leeuwen (2006) also saw the integrating cohesive role of typography and proposed that it is a semiotic mode and needs a detailed grammar.  88  4.7.4 Visual Cohesion and Composition as the Integrating Code of Image In picturebooks, the cohesive visual work is done by the integrating code of composition, which utilizes five resources, as well as the integrating code of rhythm, to construct the sequential narrative. Those resources are information value, salience, framing, colour, and typography. To return briefly to colour, I note that it is also an integrating system in O?Toole?s (1994) SFL inspired framework for analyzing images. He suggests that the use of SFL?s metafunctions makes it possible to delicately describe and discuss the rhythms, patterns, line, volume, and other materially accessible aspects of the work. O?Toole?s framework, like Kress and van Leeuwen?s, draws on composition as the analogue of the textual metafunction that constructs paintings and some other form of art. The integrating systems O?Toole uses are similar to those of Kress and van Leeuwen and both note colour as a resource for compositional work.  Almost every study of multimodality using SFL draws on the work of both Kress and van Leeuwen (2006) and O?Toole (1994), citing them for their important pioneering work. The two studies reinforce each other and foreground the complex simultaneous functioning of dynamics in visual semiotic modes. It should also be noted that subsequent studies, as Thomas (2009) remarked, reproduce some of the oversights of Kress and van Leeuwen. However, the work of Kress and van Leeuwen and O?Toole still provides the foundation for critical analysis of the construction of multimodal formats. Their oversights should be inspiration for further explorations into the textual cohesive work in the visual modality. Their work continues to provide an important precedent for using SFL and supplies key ideas for visual analysis. 4.8 Applications of SFL to Sequential Narrative Forms The notions central to SFL, the three metafunctions and the ?three corresponding situational variables [of field, tenor and mode] that operate in all communicative contexts? (Unsworth, 2006, p. 57), are central to Unsworth?s systematic functional semiotic perspective. Those ideas structure his review of the range of verbal/visual interactions in multimodal texts. 89  Unsworth is particularly interested in developing a metalanguage for language-image interactions, especially one that can be used by teachers. Because his work uses picturebooks to illustrate the language-image interactions, his explanations and findings have been helpful to me in demonstrating ways to describe those relationships with a vigilant view to the situational variables.     The articulation of Halliday?s three metafunctions and their close relations with the three situational variables of field, tenor, and mode as resources for the analysis of interaction of verbal text and image is a foundational contribution to the creation of visual grammars and metalanguage (Kress & van Leeuwen, 2006; Lemke, 1998; Martin, 2003; O?Toole, 1994). Although my focus is on cohesion, essential to my study has been the understanding of the relations between the metafunctions and the situational variables in order to relate the particular structural work of information flow and information focus with the non-structural work of cohesive resources. The systematic approaches of Halliday and those who adapt his work have inspired my approach to looking at colour and repetition as cohesive resources. SFL and its related frameworks have contributed a rich system of theories and terms for multimodal analysis. Opportunities for further investigation arise from omissions, such as questions about the purpose of illustration or image in the case of narrative forms and formats, the narrative structure, and in the area of typography. As previously mentioned, Thomas?s work on typography as a cohesive resource addresses Kress and van Leeuwen?s oversight of typography as a possible integrative system or resource for composition.  I want to acknowledge recent important applications of SFL to other multimodal and visual forms and formats that do not bear directly on my focus (e.g. Astorga, 2009; Guijarro & Pinar Sanz, 2009; Painter, Martin & Unsworth, 2011; Unsworth, 2001).  4.9 SFL Resources Supporting My Study In conclusion, SFL has inspired my application of a systematic approach to the investigation of cohesion in picturebooks. In particular, the notions I find most important are the three metafunctions of language that apply to multisemiotic analysis and the three situational variables that address the critical issue of context in that work. Equally important is the 90  comprehensive work on cohesion by Halliday and Hasan (1976) and Hasan?s notion of ?cohesive harmony.? In the next two chapters on colour and repetition I have applied these central notions to the analysis of cohesive resources in picturebooks with special attention to the issue of context, which is central to the multiple contexts of situation in picturebook narratives. SFL?s textual metafunction and the visual variations of it developed by Kress and van Leuween and O?Toole as ?composition? along with Halliday and Hasan?s comprehensive work on cohesion are key analytic resources.  Since the work of cohesive resources such as colour and repetition is work that relates to information flow and information focus, I am also drawing on SFL?s notions of Theme and Rheme and New and Given. 4.10 Cohesion  Earlier in this chapter, I discussed some key aspects of cohesion derived from SFL. Here, I will extend the discussion to describe how the visual cohesive resources work and what they look like when constructing coherence in the picturebook format. As O?Toole (1994) and Kress and van Leeuwen (2006) noted in their work on multimodal texts, what integrates a text varies from form to form. Also, they note that in multimodal texts the general sense of cohesion is highly visible. The picturebook, as a visual narrative, is a particular form that due to its sequential multimodal nature involves several integrating codes and generally achieves its cohesive unity visually.   The picturebook is a composite of turning illustrated pages, organized by rhythm, and by the repetition and patterning of elements across all its pages. Integrating sequential forms involves two kinds of cohesion with a difference ?between cohesion as a relation in the system, and cohesion as a process in the text. ?Cohesion? is defined as the set of possibilities that exist in the language for making text hang together? (Halliday & Hasan, 1976, p. 18). The relevance here in terms of the visual cohesion of a text is that in looking at general visual cohesion, it looks like a directional process in a sequential narrative. Halliday and Hasan explain that ?cohesion as a process always involves one item pointing to another; whereas the significant property of the cohesive relation, . . . is the fact that one item provides the source for the interpretation of another? (1976, p. 19). 91  The construction of physical continuity is usually a directional cohesive process. The repeated iterations of colour in a limited colour palette, combined with physical continuity, construct a general visual cohesion that constitutes a cohesive process. Earlier, text was defined as a unified entity that can be of any length. A coherent text, no matter how long or how many pages or scenes, is one that hangs together. To use Hasan?s (1984) clear and direct definition: ?I used the word [coherence] to refer to the property of ?unity,? of ?hanging together.? By this definition, any object is coherent to the extent that its parts hang together? (p. 181). The coherence of a sequential multimodal form depends on the logic of the visual sequence, the continuity of rendered images, and cohesive support from the verbal text. In addition, the coherence of the picturebook text requires continuity of the verbal cohesive resources as well as the visual ones (Halliday & Hasan, 1976).  There are visual corollaries to the verbal cohesive relations constructed by the deployment of the resources of lexicogrammar, particularly the resources of what Halliday (1985) terms the sense relations of synonymy (identical meaning), antonymy (opposite meaning), hyponymy (relationship between class and sub-class), and meronymy (part-whole). In visuals, cohesive relations can also be constructed by some of the sense relations, as can be seen in I Went Walking (Williams & Vivas, 1989), in which, on each double spread with the question ?What did you see?? a portion of the animal is partially visible and then featured as a full rendering on the next double spread. In this book the sense relationship of meronymy is used throughout the visual text. The basic repetition of a lexical unit is also a resource for cohesion (Halliday & Hasan, 1985). It may be the simple direct repetition of a name or term. This kind of cohesive work is used frequently in picturebooks for children under five years old both in the verbal text and in the visual where it is characteristic to repeatedly depict visual iterations of the main character on almost every page or double spread. As mentioned earlier, this kind of repetitive cohesive work tends to be part of a continuing cohesive process, since it is a directional progress of iterations rather then relational cohesive work.  92  Cohesive relations can be constructed cross-modally by referring verbally to visually depicted elements. Coherent texts can contain chains of cohesive relationships, such as in I Went Walking (Williams & Vivas, 1989), where there are long cohesive chains of repetition in each modality, and long chains of cross-modal cohesive relations between the animals mentioned in the verbal text and then pictured in the visual text.  In long, complex texts, cohesive ties alone are not sufficient to construct a coherent text. A coherent text is formed by a number of interactive relationships occurring across strings of clauses. Hasan (1984) calls them chain interactions because of the semantic connections between the chains of cohesive ties that make them ?hang together? in a logical manner. ?Cohesive harmony? is Hasan?s notion of how the coherence in text is related to interactions between cohesive chains. In her discussion there are two categories of cohesive ties: identity chains and similarity chains (Halliday & Hasan, 1985). Identity chains are made of things that are identical, whereas similarity chains are chains of similar things. For example, John?s hat is identical to his hat, while his kids are similar but not identical to my children. Nor is an oak tree identical to a fir tree, but they are both similar because they are trees. Identity chains are text-bound while similarity chains are not (Hasan, 1984). Hasan (1984) posits that verbal textual coherence is constructed by the degree of interrelations between the chains of identity and the chains of similarity.  Semantic interaction between chains is criterial to Hasan?s notion of ?cohesive harmony?; thus an abundance of interactions between the chains increases textual coherence. It seems reasonable to posit that interrelations exist between verbal chains of cohesive ties and sequences of visual instantiations.  Visual iterations of identity and similarity items that relate to items in verbal cohesive ties appear to strengthen the cohesion and coherence of the multimodal text. In verbal texts functional grammatical connections between members in the two chains, that is, semantic relations between an item in the identity chain and a member in the similarity chair, strengthen the coherence of texts. It appears that the coherence of a multimodal text is strengthened by interactions between cohesive chains and further strengthened by the additional interaction between verbal chains and visual cohesive chains or configurations. 93  However, the interactions between the verbal and visual chains are semantic. Again, a simple example appears in the book I Went Walking (Williams, 1989), in which the illustrator Julie Vivas renders visual instantiations of a number of animals on most of the pages of the book, which form numerous visual chains. One chain of visual depictions of a black cat can be likened to an identity chain. At one point in the verbal text the term ?black cat? is written, and I believe it adds coherence to the multimodal text with the cross-modal interaction between the verbal cohesive item and the visual renderings. Also, the identity chain of black cat images visually interacts with the multipage colour configuration of black typography. It could be viewed that the repeating black colour configuration of the typography of the written words ?black cat? interacts with the visual black cat when they appear on the page together. This question would take me beyond the scope of my study but it is one that needs to be investigated in subsequent work on the nature of the interaction between verbal cohesive resources and visual cohesive resources.  Hasan?s description of the relations between chains of cohesive resources may apply to cross modal cohesion. She observes: The echoing of functional relations becomes a powerful device, but this happens only because the members of the chain echo one another in any event. Herein lies the rationale for insisting that the minimal condition for chain interaction is the ?echo? of a functional relation, in other words, chain interaction can be recognized only if at least two members of a chain stand in the same grammatical relation to at least two members of some other specific chain. (Hasan, 1984, p. 219) Clearly, Hasan is looking at the grammatical relation, which in the case of a cross-modal interaction may not involve a grammatical relation but more of a semantic-graphological one. In my analysis, what I see very clearly in the relations between the visual chains is colour cohesion, which is the focus of my next chapter. Halliday and Hasan (1976) posit that cohesion is semantic work and is the basis of a text?s coherence, although it is not the sole source of coherence. They argue that what makes a text coherent is its quality of wholeness. There is the question of whether or not a text can be constructed cohesively but still lack coherence. In fact, it is possible for a text to have general 94  cohesion, particularly a visual or multimodal text, because it may look cohesive due to deployment of colour and compositional elements. However, it may not have coherence because there is no interaction between long chains of ties. Some texts, often poetic ones, may require multiple readings and knowledge of esoteric references to ascertain what may appear to be a missing source of reference. In fact, as the picturebook format is now sometimes appropriated by creators for an adult audience, there is an increase in the number of books that appear to lack coherence because they require multiple re-readings and serious puzzling-out of their meanings. What is fascinating is that it is the poetic picturebooks of Maurice Sendak, Anthony Browne, and Ron Brooks, to name a few, that established this poetic format. Indeed, children, because they can be very keen re-readers, are often more patient and successful than adults in construing the meaning. In summary, my investigation into how cohesive resources construct sequential multimodal narrative texts has been narrowed to a focus on colour and repetition. Colour was selected because it is both a cohesive resource and a key integrating visual system that provides immediate and general visual cohesion to sequential texts. Colour constructs general and particular cohesion through the use of colour deployments. It is particularly good for visually demonstrating the cohesive role of repetition and the close relationship between the cohesive work of colour and the dynamic of repetition.  Repetition is selected as a focus for four reasons. First of all, repetition is a basic building block for patterning. Second, it is a key cohesive resource. Third, repetition is essential to the construction of visual textual continuity, which is integral to the logical reading of the visual narrative text. Fourth, repetition is an important dynamic for providing levels of invariance, or the Given within a context, which is essential to the foregrounding of the New.  In Chapter Five, I discuss the cohesive resources of colour and repetition to set the foundation for my analysis of them and their role in the textual construction of the two picturebooks reported in Chapter Six.  95  Chapter 5: How the Cohesive Resources of Colour and Repetition Contribute to the Construction of Coherent Narrative Picturebooks 5.1 Introduction This chapter begins with a description of my methodology for a close analysis of the cohesive work of colour and repetition in picturebooks. This is followed by a discussion  of rhythm, which is the integrating code in cohesive sequential visual texts. In the third part of this chapter I describe how colour constructs cohesive relations. Colour?s meaning-making, textual work is explicated in terms of how it produces continuity, variation, contrast, salience, and texture. Next I discuss the role of colour in the construction of general visual cohesion of the whole text. This is followed by definitions of the key colour configurations that create particular cohesive relations as well as textual rhythms.  I conclude with a discussion of the work of repetition. My discussion on colour precedes my discussion on repetition because colour creates a high degree of immediate general cohesion and its deployments are key resources for manifesting visual repetition in the creation of a book?s rhythm and pacing. Colour has a vital role in the rhythm of visual storytelling for as Matthiessen (2007) notes, ?while prosodies and waves can be manifested through intonation in spoken language, they can be manifested through color in images? (p. 24). 5.2 Methodology 5.2.1 Introduction My methodology evolved over the course of my exploration of the picturebook and its scholarship. Early in this research I used a close analysis of picturebook construction to isolate some key features. I selected this approach as it is commonly used in literary discourse studies and because I found it useful for deconstructing the textual makeup of picturebooks. Furthermore, my choice of methodology was reinforced by the work of O?Toole (1994) in The Language of Displayed Art, by Kress and van Leeuwen?s development of a grammar of visual image, and by the work of a growing number of academics (Hasan, 1985; Unsworth, 2007; Williams, 1988) using SFL inspired frameworks 96  to analyze specific multimodal texts. My analyses of colour and repetition use the same method, which involves a systematic reporting of the close, analytic reading of two narrative picturebooks for young children three to five years of age. My decision to use close reading was especially inspired by O?Toole (1994) whose particular interest is in making the analysis of visual construction accessible to general viewers. Like him, my aim is to provide visual knowledge in language accessible to educators and multimodal creators.  SFL led to my decision to use a systematic approach to analyzing text construction. This decision along with my observation that many academic picturebook studies in this area lack a systematic methodological lens inspired my conception of a set of six systems of resources as a way of looking at the textual construction of a picturebook (described in Chapter 1). As I progressed in my study of SFL and its application to sequential visuals as well as in my explorations of narrative conventions, I undertook close but informal readings of many picturebooks. This led to a formal study in 2010 using close analysis of the patterned verbal and visual instantiations of the main characters in 25 picturebooks for young children. These informal and formal analyses reinforced my choice of methodology. Furthermore, these studies led to my creation of the definitions of colour configurations and the intra-textual environment that became features of this analytic work. 5.2.2 Evolution of My Methodology My first analytic work focused on the compositional placement of characters and elements in sequences of scenes in several picturebooks including Where the Wild Things Are. I experimented by cutting out key narrative figures and items and then altering their placement on the page to determine how these alterations might change the sequential rhythms. The adjustments to Sendak?s original choreography demonstrated that my changes altered the narrative flow of characters and events and hence the meaning of the narrative. This close analytic work also confirmed the musicality of Sendak?s work, which is evident in the patterning of page composition, framing, colour, physical gestures, and in the narrative language. Focusing on the placement of key characters and elements foregrounded patterns and rhythms in the visual narration.. Further, reviews of the verbal text revealed frequent 97  verbal iterations of the main character as well. These findings motivated me to analyse a collection of 25 picturebooks for their patterns of main character iteration. My study of character iterations in 25 picturebooks was further inspired by Altman?s narrative theory for film and written fiction in which he notes that repeated instantiations of a character in text and image construct the character and that in film the visual iterations constitute the visual following required in order for the viewer to construct a character (2008). Altman (2008) posited a theory based on the key notions of Narrative Materials, Narrative Activity, and Narrative Drive. Within the notion of narrative activity are the two processes, Following and Framing. Framing is the process of contextualizing the narrative material/characters and the narrative activity. Following is a very good metaphor for the construal of cohesive process of visual repetition because it describes the viewer?s activity in reading a picturebook or film.  With the principle of following in mind I looked at how characters are iterated visually and verbally in picturebooks. This study underscored the quantity of visual character representations taking place in books for young children. It especially demonstrated the important role of visual repetition in constructing characters. It was revealing to see how words and images collaboratively iterated the character and to observe patterns of those incidents. For the formal study I selected 25 award-winning picturebooks published from 1960 to present. These books were selected because they represent outstanding achievement in the picturebook format and have been widely acclaimed for their excellence by awards and prizes in their countries of origin. My interest in naturalistic narrative picturebooks for two- to five-year-old children related to their widely acknowledged contribution to young children?s literacy and literary development.   My initial informal review of 25 books influenced the focus of my doctoral research because in the course of my close examination of the iterations of main characters, other patterns of cohesive resources emerged. In looking for the resources used to highlight the repetition and support the following of characters I noted that colour, salience, and framing emerged as key 98  resources for that work. Thus, the study was instrumental in my final choice to focus on colour and repetition. For my dissertation, I narrowed my selection to four books from the group of 25. They were, Where the Wild Things Are (1963) by Maurice Sendak,  I Went Walking (1989) by Sue Williams, illustrated by Julie Vivas, Beegu  (2003) by Alexis Deacon, and How to Heal a Broken Wing (2008) by Bob Graham. Although I analyzed four, in the interests of time and the length of my dissertation, I chose to report on two, Where the Wild Things Are (1963), acknowledged as the ?classic? picturebook of the 20th century, and Australian award-winning book, How to Heal a Broken Wing (2008), an exemplar of contemporary picturebook creation.  The two I chose to report on are the earliest (1963) and the most recent of the four publications, Graham?s book (2008). These two offered an opportunity to demonstrate several significant changes in picturebook conventions that have evolved over the forty year period between their publications. Where the Wild Things Are has a single protagonist and a narrative single focus whereas How to Heal a Broken Wing has a dual narrative focus with two protagonists. Each of the books features a young protagonist. The various settings include the home, the city, and a distant island. Each protagonist (a boy) is approximately four years old. The narratives and the language are naturalistic and deal with the emotional themes of a young child?s life.   Although there are a number of interesting differences between the two books most obvious is the near absence of a parent in Where the Wild Things Are compared to the active participation of two parents in How to Heal a Broken Wing.  The parent in Sendak?s book has an important role although she is never pictured. In How to Heal a Broken Wing, both parents are pictured and actively participating in the narrative. Graham?s use of the comic book paneling depicts 75 scenes in contrast to Sendak?s 18 scenes. Where the Wild Things Are has been widely acknowledged as the quintessential classic picturebook. It has been highly influential in the field of picturebook creation from the time of its publication to the present day.  For its time it represents a significant change in the 99  emotional content of the narrative and it represents the picturebook as a literary art object?a literary format that offers many levels of meaning to readers of all ages.  How to Heal a Broken Wing is a twenty-first-century book that is also a multiple award-winning book, one, in my view, that has the makings of a classic in its multiple levels of meaning. Its use of comic conventions and its narrative structure are significantly different from Where the Wild Things Are.  Furthermore, its dealing with a child?s context of living in an urban setting with access to global events, is in stark contrast to the internal story of Max. How to Heal a Broken Wing represents a shift toward stories dealing with social action and social justice. These concerns outside the child?s home are handled here in narrative shifts between the child?s home and the child?s interest in the small injured bird he finds on a visit to the city. 5.2.3 Colour Configurations that Construct Cohesive Relations In looking at one of the books, I Went Walking (1989) by Sue Williams and illustrated by Julie Vivas, I noticed that an outstanding and obvious cohesive feature is the colour palette and the configurations of colours across the pages. This made me look at colour as a powerful resource for general cohesion and then as a more subtle cohesive resource for cohesive relationships throughout the text. I began to see several different types of patterned configurations. Finding a variety of colour configuration, associations, and identities helped me to define ways that colour constructs cohesion. After looking for the incidences of them in many picturebooks I chose to use them as features in this study.  Later in this chapter I will describe them in detail and explain how they do cohesive work. 5.2.4 A Systematic Perspective  Because there is often complexity in what may appear to be a very simple narrative picturebook, it is important to initially analyze the systems one by one. Even a simple multimodal picturebook involves a complex orchestration of resource deployments from all the systems that construct its text. Each of those systems affords the expression of many patterns. Furthermore, there are patterns created by the interaction of systems and modalities. For example, in Blueberries for Sal, the narrative contains two stories of a mother bear and 100  her cub who feed on blueberries, while a human mother and her child collect berries to preserve for the coming months. The verbal text describes the parallel activities of the two pairs, the emotions of each pair, and then it realizes the sounds of blueberries being eaten in patterned language- ? Little bear, munch, munch Eat all you- gulp, you can possibly hold! Swallow. Little Sal said nothing. She picked three berries and dropped them, kuplink, kuplank, kuplunk?. The physical construction of the book, its horizontal layout and its depiction of a distant hill with each couple foraging through blueberry bushes, establishes the material situation setting. These are only a few of the many patterns in the story. Here, the point is that, even a simple narrative has complexity in the orchestration of all the systems implemented to express it. While my intention was to be systematic and consistent in the close analysis of each picturebook text, there are several differences between the two discussions of colour and repetition due to significant differences in the nature of each resource. Colour can construct general cohesion, which is immediately and clearly observable. Repetition is a dynamic that is cohesive over a sequence of images or scenes because it is a sequential process that must be made visible by the deployment of a number of other resources such as colour, salience, framing, size, scale, perspective, and position.  Colour can render repetition but the reverse is not so. The significant difference between colour and repetition is due to colour?s nature as a resource that can realize and express dynamics such as repetition and salience. Repetition cannot render colour. Colour is first of all a basic resource, while repetition is first of all a basic dynamic Iterations of varieties of salience, size, scale, and position can also render repetition.  The methodology used here includes the formulation and application of two systematic frameworks. The first of these, a framework of textual resources I introduced in Chapter One and the second, a model of the intra-textual environment I will now describe. 5.2.5 The Intra-textual Environment of a Text In the course of reviewing many picturebooks, I found and defined what I term the intra-textual environment of the text. As I looked many picturebooks I began to see that each one?s 101  intra-textual environment is comprised of a set of textual environments, each related directly to at least one of the six resource systems that I view as textually supporting the construction of a picturebook. Introduction  The notion of the composite intra-textual environment provides a framework for considering the systems in play in picturebook text. I introduced these sets in Chapter 1 for use in constructing, editing, and analyzing a picturebook.5 They are closely related to the intra-textual environments. For example, there is the typographic system, a set of resources that can be deployed to construct both the typographic environment of a text while they simultaneously express the narrative and collaborate with the resources that are rendering the visual environment. This framework for looking at a physical book is a practical model for looking at how the narrative is expressed visually by each intra-textual environment and for calling out the interactivity between the layers or environments. It is a model that works for wordless and worded narrative books as well as illustrated books, picturebooks, and other sequential narrative forms. The intra-textual environment of a picturebook is a collection of intra-textual layers, beginning with its narrative and followed by its topographic, typographic, and visual (including the colour environments) layers.6 These environments while closely related to the textual systems described earlier are not precisely the same. The verbal system is realized in the typographic, particularly the written words of the text.  A way of understanding the functions of multiple intra-textual environments in the picturebook is to consider how they construct the sensual environments of the text: the physical handling of the book is the sensual experience of its topographic environment; the visual experience is in the viewing of its topographic, typographic, compositional, colour, and rhythmic environments; the aural experience includes the sound of the turning pages and                                                  5 See also Table 3 Resource Systems in a Picturebook 6 See also Table 4 The Intra-textual Environment of the Text and Its Textual Environments 102  the sounds of the language; the rhythmic environment is experienced through  the touch, sight, and sound  of deployed resources in all the intra-textual environments. The narrative system is the armature for the textual systems, the narrative, and any embedded narratives. This system is often the inspiration for constructing all the other intra-textual systems. Similarly, the rhythms of the narrative construct a kind of tempo for the deployment of cohesive resources such as colour and repetition. In considering the sensual environments of the picturebook the overall rhythmic environment is often less obvious than other elements of the auditory and visual environments because it may be subtle visually and aurally (except in the case of rhyming texts and those with dominant repeating patterns). The Narrative Intra-textual Environment The narrative environment is expressed by some or all of the resource systems, and in some or all of the intra textual environments. Since stories often contain secondary narratives, there can be embedded narrative environments in the picturebook. Each of the environments collaborates to construct the overall intra-textual environment. Sometimes colour environments function as specific narrative environments.  A narrative environment is a semantic notion that is expressed in a picturebook by the rendering of the visual and the aural sensual environments. To describe the expressed content of any given narrative environment one must describe what is at hand and what can be seen, touched, or heard. Thus, to describe the expression of the narrative environment one must look at the topographic, typographic, and visual environments. I use the term narrative environment to refer to particular material situational settings and scenes that render the material contexts particular to the narrative, which may include embedded narratives that will also be rendered as material situational settings. An embedded environment is a scene or setting literally embedded in a larger scene or setting. It can be likened to a large stage set with a spotlight on an area of activity. Embedded environments could also be likened to framing devices. The framing device of a tab or flap such as those found in pop-up books, conceals another environment within the larger environment. In some cases, the embedded environment could be like a secret room behind a bookcase in which a tab or a flap lifts up to reveal another small setting. In a picturebook application for the iPad, 103  a tap on the bookcase would open it up to reveal the embedded environment. In a picturebook, the use of colour can highlight an embedded area of activity. The colour can highlight a particular material situational setting, as well as the narrative activity taking place in that location. I mention these variations in embedded narrative environments to suggest the range of possible embeddings in the narrative environments of the ever-evolving material formats for the picturebook. Topographic Intra-textual Environment Although creating the intra-textual environment of the work begins and is inspired by the narrative, the experience of reading a narrative picturebook begins in handling the topographic environment of the picturebook, its physical form. There is a challenge in selecting the sequential order for describing these intra-textual environments because the order in which the environments are experienced is different from the order in which they are created.  The overall intra-textual environment is made up of layers that begin with the physical topology of the book (what I have termed the topographical environment of the text). There is first the physical cover binding and the paper used for the printed pages. The colour of the paper may provide the background for the printed text and its images and as such becomes the first layer, the background layer of the intra-textual color environment. The physical book may be bound in a fabric of a particular colour and the pages of the book will have a particular colour, texture, size, and weight. If the book is not bound in fabric then it may have a printed cover, which will have colour of one sort or another.  The colour of the binding and the paper usually will be related to colour used in the book to construct its overall intra-textual colour narrative environment. These physical features can be resources to express narrative information or themes. For example, the colour of the cover binding and the endpapers can be selected to reflect a general theme from the narrative such as a red binding and red endpapers for a book on the history of Valentine?s Day cards. 104  The first or basic layer of the topographic environment is the paper colour or the rendered background colour, which may be the actual colour of the paper or a background colour printed onto every page. For example, in I Went Walking every illustration is rendered on a white background. The white is the colour of the paper. In some books, such as How To Heal a Broken Wing, there is a mixture of full colour bleeds and some pages with spot illustrations that use the colour of the paper for a background colour. The term bleed is used to indicate that the printed colour on a page extends beyond the edge of the page. So a full bleed is a page with printed colour that extends beyond the edges of every side of the page. In pages with four- sided colour bleeds, the background colour or colours are within the printed image. In non-bleed formats there may also be formal framings, which offer the possibility for multiple backgrounds in multiple colour palettes. Colour backgrounds can vary from framed scene to framed scene, from framed to unframed scene, and from unframed to unframed due to the rendering of representational content. Variations in backgrounds may be rendered to express not only ideational changes but also variations and changes in narrative content, mood, and interpersonal engagement. Figure 1 Example of full colour bleed, partial bleeds, and spots 105  Some texts use only one background colour selected from the work?s limited palette, but it is not uncommon for narrative picturebooks to use a number of background colours. The variations in colour backgrounds within one text can be used to convey changes in the circumstances of time, mood, and place. In a complex text, such as Where the Wild Things Are, the visual text uses different background colours to convey changes in time and place. Typographic Intra-textual Environment The typographic intra-textual environment includes all printed language, from that on the cover and book jacket to the printing on the bound pages of the book. Although the colour environment is often the first accessed, the intra?textual typographic environment can sometimes play a strong initial role. The typographic intra-textual environment in Anthony Browne?s Voices in the Park illustrates my point in his use of different fonts for each character?s versions of their trip to the park. There are four stories within the book and each has its own typographic environment. As well, each story within the picturebook has its own intra-textual colour environment. The typographic layer is closely related to the overall design and layout. Visual Intra-textual Environment This environment is composed of all visual elements. With the exception of wordless books the visual environment of a picturebook is  primarily composed of images and typography. Composition or the design and layout integrates the typographic and the visual. The visual environment includes aspects of the topographic, the typographic, and all the systems (framing, salience, position) that construct the page compositions and the text?s major rhythm. The overall visual textual environment may be further specialized into particular colour environments. These colour environments will be related to particular material situational settings and particular contexts of situation.  The colour environment of the picturebook can be constructed by a number of resources and their iterations (such as the limited colour palette), iterations of its individual colours, and subsets of its palette colours. I believe that selections of those colours can be used to distinguish a variety of embedded environments within the overall intra-textual environment 106  of the text. Browne?s Voices in the Park is a fine example of a text containing four distinct colour environments. As previously mentioned, each of those colour environments has its own typographic environment. There can be a number of embedded limited colour palettes, just as within a house there may be rooms that have distinct colour schemes made from some but not all of the colours from the overall limited colour scheme set for the home. Each embedded palette is cohesive with the overall palette and with the other embedded environments that feature colour(s) from the home?s overall colour plan. In a text with multiple embedded limited palettes, each of these palettes is based on the text?s overall limited palette, thus the general visual cohesion comes from this overall limited palette. Colour constructs relationships across all the pages and environments of the text. I believe it functions both as a structural and as a referencing cohesive resource. Some colour environments associate with certain colours so it could be argued that some rhythmic patterns associate with particular narrative environments. Rhythm and repetition, especially the repetition of colours from a limited palette, connect the environments, constructing beats through all the other textual environments. Just as colour and or tone is a powerful general cohesive resource, so is rhythm. Thus, the cohesive resources of colour, tone, and rhythm are key constructors of intra-textual environments and of the cohesion and coherence of picturebooks. Historical Context for Describing Intra-textual Environments Describing a book in terms of layers of environments fits nicely with the historical evolution of the book and printing but, more importantly,  the concept of layering, functionally separates out critical resources. For a good period in its history, the typographical intra-textual environment was the significant layer for analysis of the printed book.  In those times images such as a small etching or woodcut was added by physically glueing or, to use the correct term, tipping-it-in, into the printed and bound book. As printing technology advanced to its current digital state, the notion of layers accurately describes the digital process of designing and composing the typographic text. In a design program there will be a layer?the typographic layer?dealing with all the copy. Another layer has either full colour images or black line work. If it is black line work with one colour 107  then another layer will contain the additional colour. Digitally it breaks down into layers. Even within the layer of full colour work, it would be possible to separate the colours digitally into their constituent process colours of cyan, magenta, black, and yellow. The point of this description is to underscore the fact that using the framework of layers is a functional way to describe and define the resources and intra-textual resources of a picturebook. The Intra-textual Environment of a Text and the System of Cohesive Resources The framework of intra-textual environments is useful for separating out five of the systems of resources in order to isolate their deployments, but not for separating out the sixth system, the system of cohesive resources I define a system of visual and verbal cohesive resources but I do not describe cohesive activity as an environment because each of the other environments contains cohesive activity, and much of it relates to cohesive activity in other layers. In fact, the relationships between cohesive activities in one layer of the intra-textual environment and another and constitutes interactivity of the sort Hasan (1984) described as a resource for cohesive harmony. This interactivity between environments is often cross-modal and can be multimodally cohesive. These descriptions of the intra-textual environments conclude the description of my methodology. The systems of resources and the intra-textual environments they construct have been integral to my close analysis of cohesive work. Next, I describe the integrating code of rhythm as a foundation to the detailed discussions of how colour and repetition work cohesively.  5.3 The Integrating Code of Rhythm in Sequential Narratives Sequential visual narratives are integrated by the code of rhythm. This idea draws on Kress and van Leeuwen?s (2006) notion that ?the integration of different semiotic modes is the work of an overarching code whose rules and meanings provide the multimodal text with the logic of its integration. There are two such codes: the mode of spatial composition . . . and rhythm, the mode of temporal composition? (p. 177). A picturebook text unfolds over time as its sequence of page-turns constructs some of the rhythm that will contribute to the 108  overarching code that inevitably integrates the text. Composition is an integrating code for a book?s individual pages or the double-page spread compositions. However, it is the patterned deployment of resources over the sequence of pages that ultimately integrates the entire sequential narrative text.   Rhythm is fundamentally a construction of patterns, which may be comprised of sounds, marks, complex combinations of multiple media, and combinations of verbal and visual dynamics. In temporal media such as film and picturebooks, the rhythm is constructed by a combination of the rhythms of the medium?s intrinsic materiality (such as the turning pages of a printed book) and those created by the patterning of all the resources used to construct the semantic content of that book including the typography, images, and page formatting. In a picturebook text, the textual resources of the words?their sounds and written typographic iterations?are deployed in a range of patterns across the turning pages. Those patterns contribute to the textual relationships that construct a cohesive text, what Hasan (1984) describes as having the quality of ?hanging together? (p.181).  In most cases, a book?s narrative content and structure inspire and influence the construction of its major rhythm. In fact, from an illustrator?s perspective, I believe the picturebook?s rhythm and its textual cohesion should work in the service of its narrative content. Sometimes the narrative content contains features associated with particular visual and aural rhythms, such as a prancing horse, a train on a track, or the waves of the sea. In picturebook texts and films with spare use of words, the narrative unfolds primarily in the visual modality and so the patterned or repetitive deployment of colour, shape, salience, framing, and compositional positions expresses much of the text?s rhythm. Among visual resource deployments, colour is often the most visible, powerful, and cohesive in constructing a text?s rhythm(s).  The narrative?s meaning and structure provide the armature or scaffold for the other textual systems as well as semantic inspiration for the rhythms of resource deployments instantiated across the text. For example, multiple rhythms may be expressed in a text by narrative resources and features of form, such as the pacing and sequencing of the narrative content, by the amount of time taking place in the fabula or story events, and by the actual time spent 109  accounting for those narrative events (Bal, 1985; Verstraten, 2009). Next, the topographic or physical form of the picturebook, beginning with the pagination, establishes the text?s basic physical rhythm(s). Indeed, crafting the visual rhythm of a picturebook begins with fitting the narrative pacing into a topographic format, which is part of the process of paginating the story. Pagination is the work of dividing a story to fit the pages of the book. The physical form of the book, its size, weight, and bound paper pages is beginning to be acknowledged as an important factor in the reading experience. In a recent article in Scientific American, focusing on the comparison of reading paper books and reading screens, Ferris Jabr (2013) wrote: Turning the pages of a paper book is like leaving one footprint after another on the trail?there?s a rhythm to it and a visible record of how far one has traveled. All these features not only make text in a paper book easily navigable, they also make it easier to form a coherent mental map of the text. (p. 2) The rhythm of a well-crafted narrative picturebook whether in traditional book format or e-book usually reflects its story and its themes and is usually visible in the physical design and pacing of the page-by-page storytelling. As previously noted, paginating the story is an important first resource for reinforcing and visually constructing its rhythm and pacing. Framing is an important secondary resource for creating pacing variations by adding or deleting scenes. In addition, pacing variation can be created by changing the size, shape, and point of view of each scene or framed panel and by the complexity and number of key elements integrated into each scene or page. Pagination physically establishes the reading pace of the narrative by dividing up the narrative text and allocating its blocks of text to each of the pages or double spreads of a paper dummy. A paper dummy is a paper facsimile of the book that illustrators use to sketch and plan a book. Establishing and pasting the basic sequence of images and typographic elements onto the pages of a paper dummy provides a concrete model of the picturebook, which can then be fine-tuned by sketching and adjusting page layouts that render the desired visual rhythms. That fine-tuning involves drawing on the resources of salience, position, and formal and informal framing.  110  The nature of the transitions from what is depicted in one panel or page to the next affords another resource for pacing and rhythm. Scott McCloud (1993) notes the five key transitions in sequential visual narratives: moment-to-moment, action-to-action, aspect-to-aspect, subject-to-subject and anything to a non sequitur (p.70-72). Rhythms can be constructed by the patterning of different kinds of transitions. High action, fast-paced stories usually use many action-to-action and moment-to-moment transitions that are punctuated occasionally by subject-to-subject and/or aspect-to-aspect. The non sequitur transition is particularly useful for slowing things down, for pausing, or for giving the reader/viewer time and space to absorb an intense emotion.   To conclude this section and to situate the discussions of colour and repetition within the construction of textual rhythms I want to briefly review the systematic approach I use to create the rhythm of a picturebook text or to evaluate one. The narrative structure and content usually provide the foundation or armature for the book, along with the selection of the topographic-physical form the text will take. Next, there is the fine-tuning of the pagination of the verbal text followed by the sketching in of visual elements and actions using a selection of resources, such as colour, to construct the multimodal cohesion of the work. As previously noted, in picturebooks the work of creating textual rhythm usually begins by drawing on the rhythms introduced in the verbal narrative text. In texts with dominant verbal patterns, such as those by Dr. Seuss or adaptations of Mother Goose, the verbal system sets up the text?s overall rhythm(s), which are then reflected in the patterns constructed in the other systems that co-construct the picturebook text.  In complex sequential texts, there may be multiple rhythms but, in the case of narrative picturebooks for young readers, there are fewer, and usually only one dominant rhythm and theme. The dominant theme and rhythm may be reiterated by deployments of some or all of the resources of salience, information value, framing, colour, and repetition.  5.4 Colour  This section provides background on the properties of colour. I want to reiterate that because my investigation concerns the inter-textual functions of this resource I have not discussed its 111  interpersonal work nor its associations and identities outside each of the picturebooks being studied  5.4.1 Colour as an Integrating System  Colour integrates textual elements across time and within space, which leads to the question of whether it is a system and/or an integrating code. Rhythm integrates across time and within a space and is considered an integrating code (Kress & van Leeuwen, 2006). While colour can integrate a sequential narrative text, it simultaneously works as a resource for information value, salience, and framing. Kress and van Leeuwen?s (2006) rationale for positing composition and rhythm as integrating codes is that both codes put representational and interactive elements into relationships. Can colour do similar integrating work? Van Leeuwen (2011) argues that colour is a mode and therefore does do such work.  As for colour as a code, imagine a character in a red hat rendered on a page in a book that has a limited palette of red, black, and white. Colour can be a resource for representing the appearance of the hat and for associating it with a particular character. The red hat could be also the most salient item on the page, in which case the colour red is integrating the representational with the textual. As well, the red colour, while a resource for the salience of the hat, also serves an interpersonal function. In the situation at hand, colour, which is realizing the resource system of salience, puts representational and interactive/interpersonal elements into relationship. Textually the red hat may be functioning as an identifying element of a character and contributing to both the iteration of the character (which develops the character) and the continuity of the character across the narrative. Furthermore, the red hat along with other beats of red colour can construct patterns or rhythms across the pages of the narrative.   Colour is not only a resource for constructing visual continuity and general visual cohesion; it can also construct colour associations and identities with characters or events, particular backgrounds, narrative environments, embedded environments, reading paths, ribbons or flows of colour, and other patterns or rhythms of colour, all of which constitute integrating configurations across sequential texts.  112  In conclusion, since an integrating code is essentially a system for constructing coherent composition through the deployment of cohesive resources and cohesive processes, the general work of colour in constructing cohesion across the multimodal text constitutes the work of an integrating code. The integrating resource systems are the means of relating the representational elements with the interactive (Kress & van Leeuwen, 2006). Colour, both in its general cohesive work and in its configurations and support of the other systems of salience, information value, framing, and typography, works to integrate and relate the representational and the interactive. 5.4.2 Colour and Cartography  The field of cartography has a deep history of using colour to display information and provides theory and insight into the function of colour. Contemporary information display and cartography specialists such as Imhof (1965), Bertin (1967), Tufte (1990), and have drawn on the field?s extensive archives to illustrate and articulate important insights into how colour functions in graphics and illustration. According to Bertin (1967) colour is one of six visual variables that include:  size, value, texture, color, orientation and shape and their instantaneous perceptual spheres of influence (associative, selective, ordered and quantitative). Color is only seen as associative, which means that all signs of the same color and visibility can be perceived instantaneously, and selectively, meaning that all colored signs can be grouped into families. (p.12) Colour?s ability to create associations and other perceptual dynamics is key to how it affords immediate general cohesion. It can create cohesive associative relationships on its own which partly explains its frequent use in graphics as a resource for overall cohesion and harmony. Combined with other visual variables, colour becomes a resource for representing order (when combined with value), selectivity or particularity (when combined with texture, value, shape, or size), and quantity (especially when combined with shape).  While colour is a powerful cohesive resource, it also has the potential to create disturbing visual noise that disrupts cohesion and coherence. Loud colour visual noise can be deliberate 113  or unintentional. Sometimes colour noise serves narrative or graphic themes by using colours that seem to shout off the page or shout at a high pitch at other colours rendered on the page. In general, colour noise is distracting on a map and any kind of document or illustration in which attention is meant to be drawn to salient details such as location names or labels.  Both colour and shape can construct strong general visual cohesion and highly nuanced, specific kinds of cohesion. I return to cartography for additional insights on this. Cartographer Imhof?s (1965) first rule of colour use touches on a key way that colour constructs contrast in the service of cohesion and harmony when he notes that ?color spots against a light gray or muted field highlight and italicize data, and also help to weave an overall harmony? (p. 72). His remark touches on colour?s potential to provide association, variation, and contrast without creating noise. Cartographers have been using colour in maps for a long time because of its visual appeal, its influence on memory through association, and its power to particularize, quantify, and order information (Samson, 1985). Its power to particularize relies on the human eye?s ability to recognize thousands and thousands of subtle variations in colour (Tufte, 1990). 5.4.3 Colour and Memory   The frequent use of colour to integrate visual sequential texts likely reflects the illustrators? implicit knowledge of the strong associative nature of colour images and how they are retained as memory. As Arnheim (1969) wrote, ?mental images admit of selectivity. The thinker can focus on what is relevant and dismiss from visibility what is not? (p. 105). Arnheim?s (1969) notions are based on experiments that demonstrated that participants described having collage-like memory snippets of a visual experience. The snippets consisted of salient features or gestures selected from the experience rather than a whole image of everything present in the experience. Similarly, illustration abstracts and selects salient features by using particular colours and, especially, by using limited colour palettes. The limited colour palette and the use of salience influence the viewer to retain certain items because the memory image as Arnheim (1969) states ?is more evidently limited to a few salient features, which correspond perhaps to everything the original visual experience amounted to in the first place or which are the partial components the observer drew from a 114  more complete trace? (p. 105). Even without the benefit of great salience, the repetition of colours from a limited colour palette goes into the short-term memory of the reader/viewer (Malamed, 2009). The repetition of colours from a limited palette establishes the palette as emblematic, almost like a flag, which creates a vivid association that, when repeated over a number of pages, constructs a memorable chain of cohesive iterations.   While images can convey ?high cognitive load,? which can overload and impair the working memory, well-designed images help readers/viewers navigate a complex image and pay attention to the key elements construing the main idea. The act of focused attention on key elements in an image supports the transformation of that attention into meaning that will likely be transferred to the long-term memory (Malamed, 2009). Colour salient configurations promote focused attention, which make key narrative elements more memorable and aid in the information and meaning retention of those features in the long-term memory (Malamed, 2009)..  Focused attention and the capacity of mental images to ?admit of selectivity? (Arnheim, 1969, p.104) both aid in the conceptual thinking that leads to making the connections that are essential to construing cohesion and coherence. Arnheim is acknowledging the brain?s capacity to select and hold onto what is relevant and dismiss what is not from a memory trace.  Well-designed images and diagrams convey large amounts of information efficiently and often use colour as a vital organizing resource. This is partly because colour is one of the most powerful visual resources for constructing a hierarchy of values that will be quickly perceived. In particular, the use of salient colour in a visual diagram can construct levels of emphasis and visually order key bits or groups of information. A well-designed information visual provides a quickly perceived general context that supports easy retention and invites the viewer to consider its informational details. While the role of colour includes the work of representing things, at the same time it can be used to influence the viewer?s attitude and interaction with the image or graphic. It is often the designer?s task to edit the visual text for narrative flow or information flow and to construct something that can be perceived by the viewer/reader as having continuity and cohesion. This is observable in finely conceptualized maps, which contain enormous amounts of information bits while retaining high legibility. In these maps, the selectivity of colour is enhanced by subtle variations that order and quantify 115  information, together with shape and size, so that the viewer is directed to key areas and to important foregrounded information.   5.4.4 Colour as a Meaning Making Resource   Although this study looks primarily at the cohesive textual work of colour, it is important to acknowledge the range of colour?s work, since it is a resource for creating representational images and expressing many other kinds of cultural, symbolic, and interpersonal information that descriptively serve the text. As Tufte (1990) wrote, ?tying colour to information is as elementary and straightforward as colour techniques in art. Paul Klee offers an ironic prescription, claiming that ?to paint well is simply this, to put the right colour in the right place?? (p. 81).  Tufte also makes the point that ?colour is a natural quantifier, with a perceptually continuous (in value and saturation) span of incredible fineness of distinction, at a precision comparable to most measurement? (Tufte, 1990, p. 91). The point in quoting Tufte is to emphasise that every rendered instance of colour has the potential to contribute quantitative information in support of descriptive narrative content. This quantitative information may also serve to construct salience, contrast, focus, and continuity. 5.4.5 Colour and Continuity  Sequential visual narratives require visual logical continuity to support readers as they follow the visual progression of events. This requirement exerts pressure on the creators of these texts to carefully construct the logical progression of events and to attend to the continuity of the visual representations across the sequences. In picturebooks for young readers, who have had relatively few life and narrative experiences, there is additional pressure on creators to express narrative events with an emphasis on a clear chronological and physically legible and logical sequencing, and on the frequent repetition of narrative characters and elements. Associating and identifying characters with particular colours aids visual recognition and visual following.  116  Representation of certain elements can be achieved by using a full range of colours in a realistic or abstract style. In a sequential visual narrative, this representation work must be done with a view to continuity. Renderings with a high ?modality,? (Kress & van Leeuwen, 2006, p.155) to use Kress and van Leeuwen?s term, are renderings that look highly realistic and usually represent things using a full spectrum of colours. Whether or not a narrative element?s rendering appears realistic, it must have continuity with its other depictions in the text because it is essential that it can be easily recognized and followed from one page to the next and from one narrative event to the next. Hence, rendering for continuity supports cohesion across the pages of a picturebook Furthermore, the continuity of the rendering of the characters and settings is important to the construction of the information flow and the Given elements in the story, so that variation foregrounds an intentional story event or activity.   5.4.6 Colour Variation Variation is an essential dynamic for constructing differences and changes in any modality. Colour is one of the resources for constructing and highlighting variances and changes in visual modalities. For example, if a narrative requires the depiction of ten little bears of the same size, shape, and texture in order to establish some variation from one to the other, each could be a different colour or wear hats or other clothing items in a variety of colours. To create variations, particularly the kind that might also make one of the bears the most salient, the coloured hats could all be rendered in the same size and shape, but in different colours with one hat rendered in a more saturated colour and possibly with a different texture.   Within the background or rendered environment of an illustration, variations in colour may be used to foreground narrative changes. For example, a change in colour made within an otherwise unchanged scene shows how colour adds contrast to the cohesive resource of repetition being used to construct the continuity from background scene to background scene. In that context, it is the change in colour that draws attention to a key narrative change or event. 117  5.4.7 Colour and Contrast Colour can create contrast between similarly rendered and textured elements. This can be seen on the cover and on the endpapers of Where the Wild Things Are, where colour is used to differentiate leafy shapes rendered with a crosshatched line. Subtle contrast is created by variations in the colours of the textured leaves. Further contrast can be created by the lightness and darkness of the colour, its saturation level. As a design strategy, the use of contrast is a resource for creating a hierarchy of visual information.  5.4.8 Colour and Salience, Framing, and Texture  Colour?s potential for salience is related to its rendered saturation and volume. Increasing the saturation of a colour increases its potential for salience. In addition, a colour?s salience can be altered by its size and shape relative to other elements. As such, although two items are the same shape and are rendered in the same saturated red colour, the larger shape of two will have more salience because the volume of the saturated colour further increases its salience (Bang, 1991). Thus, a very large bright red triangle is more salient than a smaller red triangle. When colour is a resource for constructing salience it is often the most noticeable colour in the composition on the page or double spread. It might not necessarily be the most saturated or purest hue, but it will have one or more of these attention-drawing qualities: lightness, brightness, size, shape, and prominence. Salient colour deployments can construct salience, framing, and information value, as well as reading paths and some of the other configurations discussed below.  Figure 2 Cross-hatched leaves with variations of colour and density of cross-hatching 118  Colour is sometimes a resource for rendering framing more salient, and thus it may in turn support the construction of a pattern of framing that constructs cohesion throughout a text. Framing using colour may also serve to separate and/or to create contrast between elements and events within the frame and those outside the frame. The framing itself can bring a focus on either a close-up or a long shot of an aspect of the narrative. At the same time the colour of the frame, through its associating dynamic, can construct cohesive ties that connect narrative themes, characters, and/or their activities as they occur throughout the text.  In the example of two salient shapes described earlier, dominance can be shifted to the smaller triangle by the application of texture to the large red triangle, thus damping down the large triangle?s salience. This texture can be created by a cross hatching of lines, as is frequently the case in Sendak?s Where the Wild Things Are. Texture can also be created by the patterning of a variety of colour shapes rendered in a various image sizes. Finally, other illustration media and techniques can create textures. 5.4.9 Colour and SFL  Drawing on SFL and the notion of metafunctions, when colour is a resource for ideational/representational content, it simultaneously functions interpersonally and textually. However, as previously stated in my study, the focus is in how colour functions textually as a cohesive resource within the environment of a picturebook.  Colour renders the representation of things and actions, while simultaneously constructing composition and rhythm. Colour configurations contribute to the construction of the composition of individual pages or double spreads by their particular placement within the image. In addition to the placement of configurations, their size, shape, texture, and scale co-Figure 3 Textures created by different patterns, media, and techniques 119  construct the composition while simultaneously representing narrative elements and actions logically consistent with the style and nature of the story. Patterned instantiations of colour configurations across the pages of the entire text contribute to the construction of the text?s overall rhythm.  From an SFL perspective, colour configurations simultaneously function interpersonally in the text. For example, in Where the Wild Things Are, Max is often the most salient element in the composition, interpersonally engaging the reader/view. The bright whiteness of his wolf suit is salient while it also represents the wolf suit colour. Textually, the instance of Max?s white wolf suit is one of the patterned renderings of white elements across the pages of the text.  As a basic resource for representing narrative elements and settings, colour can particularize the look of components introduced in the verbal mode. For example, a verbal text might describe a bear wearing a hat. In the image of the bear wearing a hat, the colour of the hat makes it more particular. The shape and texture of the hat may further particularize it. Textually, colour deployments can extend, elaborate, and enhance the verbal mode. In many picturebooks, the visuals do most or even all of the descriptive work of the narrative by rendering the look of characters, settings, and actions. In these cases colour adds representational information. Over a set of pages, visual information must be rendered with continuity in order to support the following of sequential action and to construct textual cohesion. Colour can co-construct the New and Given of the Information Value system. First, colour can be used to create an association, so that an element with a particular colour association will be readily perceived as a Given in subsequent appearances across the narrative text. Colour and variations of it can also indicate the New (Kress & van Leeuwen, 2006).    120  Examples of several colour constructions are visible in the first illustration of Where the Wild Things Are. In this first scene, Max in his white wolf suit is the most salient element in the picture. Everything pictured is New, but in the next spread Max in his white wolf suit is a Given, although he remains salient. In addition, this second iteration of Max in his wolf suit is building a colour association with Max. The framing of both images is rendered by the creamy white colour of the paper, in contrast to the textured drawing of each scene.  Just as the verbal text must be cohesive and coherent, the visual text must also be cohesive from page to page as well as coherent and cohesive with the verbal text. Additionally, the verbal and visual texts together must possess semantic unity. By semantic unity, I mean that Figure 4 Max in his salient white wolf suit 121  it must be rendered in reference to logical physical sequencing and visual narrative continuity. The text?s semantic unity can be seen in how the characters, the settings, and the narrative activities relate to the narrative theme(s).  5.4.10 Colour and Typography as Integrating Codes In addition to Kress and van Leeuwen?s acknowledgement that both the integrating codes of rhythm and composition draw on the resources of salience, information value, and framing, I believe the integrating codes also draw on the system of typography and the resource of colour. This is particularly true of colour when it is iterated in a directional sequence. For example, in Anthony Browne?s Voices in the Park, each character?s speech is rendered in a special font throughout the text. The shape, style, size, tone and colour quality of the fonts selected for each character visually particularize each of their personalities. Van Leeuwen acknowledged the oversight in not including typography and colour as key resources (van Leeuwen, 2005). In fact, van Leeuwen subsequently argued that colour is a mode (van Leeuwen, 2011) and acknowledged the importance of typography in graphic design and composite images (2005). 5.4.11 Colour and Contexts Context is central to Halliday?s (1978) view of language as social semiotic and it is central to understanding the environments of the picturebook. Defining the intra-textual environment provides a context for describing colour environments and the work of colour configurations. As I wrote in Chapter Four, meaning exchanges between people that involve language, whether real or fictional, occur in a physical setting or context, and all these exchanges also have a particular social semiotic contexts. Earlier I also noted David Lewis? (2001) view of the picturebook as having its own ecology related to its narrative environment(s) rendered onto the pages of a physical book. It is a small world of thirty-two pages; yet that small world has an environment that I am naming the intra-textual environment because it is a composite of different textual resource deployments, each of which can potentially construct a particular kind of textual environment. Individual textual environments can construct and or highlight both material situational settings and semiotic contexts of situation. Since there are usually 122  several material situational settings and several semiotic contexts of situation within the small world of the narrative picturebook, I believe it is helpful to briefly review the notions of ?material situational setting? and the social ?context of situation,? in order to consider how they relate to the various textual environments and to the composite intra-textual environment.   In the overall intra-textual environment of the picturebook, there may be a number of different material situational settings rendered on the pages, depending upon the content of the narrative. The physical intra-textual narrative environments, such as the topographic and typographic, are themselves material situational settings because the material context is literally the physical setting of a situation. In the case of a hand-held book, the book itself with its turning pages is a key element of the material setting for presenting the narrative. It is part of the material situational setting whether or not it is fictional, realistic, or fantastical. It is the physical environment where an exchange is taking place, whereas the context of situation refers to the kind of semiotic activity taking place in that material situational setting. In a picturebook, there may be many material situational settings and many different semiotic contexts of situations and they can all be parts of the intra-textual narrative environment of the text. Within one material situational setting there may occur several different semiotic contexts of situations. For example, a classroom is a material situational setting that at one moment may host a semiotic context of situation that is an informal discourse between friends on the latest film or sports event. There may even be several different kinds of informal discourse between friends taking place in different areas of the material situational setting. A few minutes later, the same material situational setting may host an academic semiotic context of situation involving a professor delivering a lecture on film theory. Then, as part of the academic semiotic context of situation, the professor might screen a film. While watching the film members of the academic semiotic context of situation may view a narrative environment within the context of the film that presents a number of material situational settings that in turn have several semiotic contexts of situation taking place.  Just as there can be changes in the semiotic contexts of situation taking place within material situational settings, there can also be changes in the colours used in those material situational settings that signal changes in the semiotic context of situation. For example, in Anthony 123  Browne?s book, Gorilla, the colour of material situational settings is changed to reflect changes in the emotional relationships between a young girl and her father. It can be complex because within a semiotic context of situation there may be several embedded material situational settings with associated, embedded semiotic contexts of situation. In a visual text, colour is a particularly powerful resource for signaling and then associating a variety of contexts with particular colours or colours from a limited palette. As I will describe in greater detail later, Sendak uses colour to subtly associate and indicate changes in the semiotic context of situation occurring within the material situational settings. 5.4.12 Colour as a Cohesive Resource Introduction This study looks at colour as a cohesive resource but with the acknowledgement that colour is not always the source of cohesion in sequential visual narratives. In a black-and-white text, other resources construct the rhythm(s) that create a cohesive coherent text, such as the other visually-sensed resources of value or tone, shape, texture, line, and the general illustration style and medium used.  Tone or the degree of darkness or lightness makes it possible to visually sense the shape and discreteness of things. In some illustrations, especially black-and-white, the tone is replaced by line so that elements are defined by outlines and details drawn by lines within outlined areas. Tone is a feature of what is known as half-tone art, which is drawing or painting in shades of grey that render different levels of darkness and lightness. It is possible to use fine black lines to render textures that look like half-tones. An example of this can be found in scratchboard illustration. Whatever is done with colour in terms of limited palettes and colour configurations can be done with tone, and even with fine lined textures, thus most of what is described in this study can be applied to black-and-white sequential visual texts. Colour cohesive work can be either or both structural and referential. General colour cohesion is produced structurally while colour associations and identities are constructed by relationships of reference. 124 Constructing General Visual Cohesion  Black and white are sometimes considered neutrals instead of colours. However, in this discussion I consider them to be colours because in many books they construct flows, paths, and backgrounds, which are all activities that often require more dynamic colours than that of neutrals. Neutrals are often used for backgrounds and to construct contrasts, but in sequential visuals, they remain neutrals if they are not functionally used for ideational or interpersonal purposes. In fact, other colours can also be used as neutrals when their function is to provide a quiet, non-expressive background. Visual colour cohesion differs from verbal cohesion because there can be general compositional visual colour cohesion that has little if any semantic connection, whereas the nature of verbal cohesion is based on semantic relations. For example, the colour of the paper and the typography can provide a sense of immediate general cohesion. Books in a series are usually connected by a cohesive design of style, colour, typography, size, and shape. Limited Colour Palette  Selecting the limited palette is a first step in constructing strong cohesive colour deployments and general visual cohesion. While the limited palette can reduce the range of representational meaning-making potential, it concentrates the colour resources and intensifies their cohesive relationship potential.   A limited palette can be a specific set of colours in a range of saturations, shades, and tints, and may be many colours in specific saturations, shades, and tints. However, a colour palette limited in terms of the ranges of saturation, shade, and tints possesses less potential for creating dynamic backgrounds, flows, and reading paths. Although an extremely limited palette can reduce the range of variations, it does not necessarily eliminate the possibilities for subtlety and complexity, as the book, Where the Wild Things Are, exemplifies.  The limited colour palette is often called a colour scheme; however, I prefer to use the term ?palette? to describe colour choices for sequential visual narratives. The term colour scheme is well suited to discussing colour schemes for buildings, brochures, and websites, but my 125  choice reflects my perspective and experience as an illustrator. A painting or illustration may be created with a certain palette at hand. The idea goes back to a time when artists used a physical palette with dabs of paint on it. Today the palette will more likely be a collection of cans, tubs of paint, or a set of selections in a computer program. It can be a less tidy array than a colour scheme. Nonetheless, it is an intentional selection of colours specific to a particular work of art or literature.  A limited colour palette is a selection of colours chosen by the illustrator or designer to support the narrative themes of a story, the desired ambiance of a building, or the functions of a graphic site. In picturebooks, it is most frequently used to construct a particular narrative environment for a text?what can be referred to as the intra-textual colour environment. An illustrator?s palette choices intentionally relate to the content, themes, and emotional mood of the narratives. In most cases the illustrator selects the palette; however, sometimes book designers or art directors influence choices with a view to having the book conform to format requirements related to a series or their particular company?s design sensibilities. Colour choices may be further influenced by the demands of editors and production staff with regard to legibility, scheduling, and expense.  Colour constructs general visual cohesion by creating an immediate flow of associations through introducing and connecting a few limited colours to a narrative?s main characters and events. This approach makes those items more memorable, especially if they are also rendered with high salience. For example, in a limited palette of two colours, one colour might be used for most of the printed text/typography and drawn line. The second colour might create a pattern of highlights in each illustration by strongly associating the main, frequently appearing character with that colour. In a limited palette, one or two of the colours may be more salient than the others may, so that in rendering a sequence of illustrations, the salient colours construct a reading path of salient beats that encourage the reader/viewer to move forward through the visual text.  126 Colour Configurations Introduction  Colour can be deployed in a variety of visually cohesive configurations. I chose to call them configurations for two reasons; first, because their varied and patterned usage creates spatial forms and configurations, and second, to differentiate them from the term chain used by Halliday and Hasan (1976) to describe cohesive verbal chains. Furthermore, the term ?chain? usually refers to something linear, while the term configuration commonly refers to something spatial in arrangement. From time to time I refer to a chain of colour iterations because in some situations they can appear in strong linear, chain-like fashion rather than in spatial configurations. The reason for chain-like iterations is likely due to the linear nature of sequential narratives. The sequential narratives? linearity is a key reason notions of cohesion in language can be applied to visual storytelling in naturalistic picture books.  Potentially, colour deployment can be highly layered, since there can be particular patterns of colour use in the material situational settings and backgrounds of the paper page, and different patterns and colour choices used in any of the distinct intra-textual colour environments of the text as well as in colour reading paths and colour flows. Each kind of colour configuration deployment has potential for adding to the construction of the general cohesion, while at the same time constructing particular cohesive relations within the intra-textual environment of the text.  Figure 5 Reading path of salient red beats across a sequence of small images 127  Each colour configuration has the potential for variations in colour saturations, shades, tones, and contrasts. Those variations allow for the embedding of one environment within another. In addition, iterations of colour can construct reading paths and cohesive colour flows that wind through all the colour configurations, ultimately creating a complex orchestration of cohesive relations. Another point about colour configurations is that some suit particular texts better than others, so not all great books contain all possible colour configurations. As Lewis (2001) posited, each picturebook has a unique ecology; each is created using a unique blend of the expressive resources at hand for that format. A rich diversity of resources increases the potential complexity of picturebook construction. As mentioned above, narrative environments can contain additional embedded environments, and a variety of colour configurations can be embedded within colour configurations. Furthermore, the literariness of a picturebook can be constructed using any or all of the configurations, narrative devices, and environments.   My intention here is to describe commonly encountered colour configurations, the ones most often used to construct cohesion. However, I acknowledge the possibility of future innovations in this area in view of the ever-evolving picturebook format. Colour Flows Colour flows are resources for continuity and cohesion across the pages of a book that work by constructing activity in layers over or under backgrounds and environments. A colour flow is usually a general flow of large blocks or areas of the same colour that take up as much space as the background. However, they can be distinguished from backgrounds because they appear on a variety of narrative elements. They are waves of colour that flow across the pages of a book, visually connecting narrative elements by drawing on colour?s associative nature. These colours tie elements together so that they literally hang together visually. As mentioned earlier, the fact that they hang together is not sufficient to construct a coherent multimodal text. A coherent multimodal narrative text must possess semantic logic. Although colour flow will contribute a general visual cohesion that invites closer reading, this closer reading may or may not reveal clear semantic relationships between elements in 128  the flow. It will nonetheless show colour functioning textually and compositionally, more like an interactive reading path. Colour flows can be rendered as colour ribbons?continuous ribbon-like strips?that are sometimes used in highly graphic and abstract styles. There can be multiple colour ribbons deployed over the pages of a picturebook. That is, within the frame or frames, there may be ribbons of colour that connect one frame to another while simultaneously another or even several other colour ribbons connect items from page to page. These ribbons of colour will be in addition to the flow of the background colour(s) and possibly colour flows. Colour ribbons can construct reading paths. Sometimes colour ribbons contain embedded colour ribbons in variations of saturation and salience of the main colour, although that dynamic is more likely found in displays or diagrams than in a narrative sequence. A general colour flow is a sequential continuity of colour over a number of pages created by substantial areas of the colour on each page. This is different from colour ribbons that are literally ribbons of continuous colour across a spread and sometimes across many spreads. Colour flows are composed of areas of a particular colour frequently rendered in large shapes across a number of pages that reproduce in clear pattern of iteration. If one were to squint ones eyes at the sequence of pages, the pattern of colour iterations would look like a flow. On Figure 6 Storyboard showing a sequence of skyscraper grays and blues 129  a computer screen, pixilating the image will reveal the colour flow(s) in an image and a sequence of images. In the book I Went Walking, every colour in its limited palette is present on every page and double spread. Many of those iterations construct a number of clear colour flows established from the cover onward. There is a dominant flow of reddish brown colour on several of the large animals. In addition, a flow of black constructed by the lines of type and the flow of the white background appears on every page. These three flows construct general visual cohesion across the pages of the book. In How to Heal a Broken Wing, there are multiple background colours. First, there is the flow of white across pages with white backgrounds that connect with white clouds embedded in framed panels. The quantity of white across the spreads constitutes a general flow of white. In this book, there are also general flows of sky grays, skyscraper grays, and blues that construct a substantial flow of closely related grays and blues.  General flows are constructed by the deployment of substantial quantities of one or several colours. An example of this is in Sendak?s Where the Wild Things Are. The dominant general colour flow from cover to cover in this book is the creamy white of the pages under and around the framed images and printed text. Although there is one dominant flow of creamy white in this complex book, there are other colour continuities and particular colour paths. However, this particular colour flow is important because, as the narrative progresses, the creamy white background colour constructs frames, then recedes from view, and then returns in a patterned rhythmic flow across spreads. The rhythmic flow of colour across the pages underscores the protagonist?s journey because the flow takes him deep into ?where the wild things are,? out again, and back to his room.  Sometimes a colour flow may be rendered in the same or similar colour as the background creating a subtle contrast. Colour flows have fluidity and usually quantitatively take more space than colour-reading paths. Colour Reading Paths  Colour reading paths are usually created by a number of separate salient colour shapes that attract attention and thereby construct a path of salient spots. A colour-reading path, in addition to building cohesion, functions as a narrative support to help readers follow the protagonist and the narrative activities. Indeed, in books for a young child, there may be reading paths that help her follow the movement of a character through the scene and toward 130  the next narrative activity. Colour shapes constructing a colour-reading path are usually separate entities. A path can be formed by a number of items from a limited palette that are of a similar saturation and degree of salience. The colour reading path can be described as a set of visual stepping-stones for the reader/viewer. Reading paths are not necessarily as fluid as ribbons and flows and are often enhanced by salience and vectors. Their iterations may also participate in the construction of other patterns and in other colour configurations in the visual narrative sequence. Paths might also be constructions of bright spots or areas. They may be within the composition of a page or double spread and/or across a sequence of pages. In some complex compositions, there may be a number of reading paths. Other resources such as line, framing, and vectors can support the construction of colour-reading paths.  Colour-reading paths may also participate in the construction of cross-modal verbal cohesive chains, colour associations, identities, and flows and may appear to convey a very general sense of visual cohesion, particularly when they are primarily constructed by the salient use of a colour. In addition, they can construct and support the pace of visual narratives. Colour Configurations and the Construction of Contexts of Situation Each of the colour configurations just defined has the potential for contributing to the construction of varieties of narrative contexts. Some may draw on traditions of colour associations. However, whether or not they draw on socio-cultural associations, within the intra-textual environment of the text, these configurations help construct distinct narrative material contexts. They construct those contexts simultaneously as they construct cohesive relations. Each configuration may be constructed by patterns of colour iterations and by patterns in the variations of the size, shape, and texture of these colour iterations to help construct a range of contexts of situation. Sometimes the configurations may be associated with particular registers of language in the verbal mode. For an example of a colour flow that might weave across the text in a pattern that supports narrative material situational setting, imagine an illustrated scene of a mother otter swimming with a baby otter on her stomach with the words of a lyrical lullaby printed across the scene. The context of situation, a mother 131  singing to her child, is visually expressed in the soft flow of the sea-green water waving across the pages and by the words of the lullaby. However, the semiotic context of situation might change to one in which there is a threat to the otters in which case there might be a change in the colour flow;or the flow might remain the same colour and represent the same semiotic context of situation but it might flow through a number of different material situational settings, such as kelp beds and waves along the shore. In a multimodal text, especially a sequential narrative picturebook text, the material situational settings, and the semiotic contexts of situation, along with verbal language registers are critical to the construction and construal of cohesion. Associating or identifying material situational settings and semiotic contexts of situation with particular colours can help viewers recognize the material situational settings and the semiotic contexts of situation. Furthermore, establishing these associations helps viewers predict and construe the narrative, from the relatively spare bits of verbal text characteristic of picturebooks. Colour Associations and Identities Introduction Across sequential visual narrative, colour associations and identities can be resources for constructing patterns of familiar narrative elements to help young viewers follow them through the story. Additionally, the flow of associations and identities across the visual narrative sequence can add continuity. Furthermore, the use of colour, particularly salient colour, to construct association configurations increases their retention in the short-term memories of viewers, which aids in the construal of textual continuity and cohesion.  The cohesive work of colour associations and colour identities is like the cohesive work of similarity and identity chains in language. In Chapter Four, I discussed the role of similarity and identity chains because the flow of those chains and interactions between them is, as Hasan (1984) noted, a source of textual cohesion and harmony. In language, the lexicogrammar realizes an imagined visual world and its inhabitants. It does this work by using words to construct a cohesive world for the characters and the reader to inhabit. In language, particular words establish associations and identities, just as colour does in a visual 132  rendering. In a multimodal text, there is the bonus resource of cross modal association and identity chains that realize associations and identities in both modes. Ultimately, much of colour?s cohesive work stems from colour associations and/or identities rendered across the pages of the text because colour, by nature, has strong associative power. Colour associations and identities are visual cohesive colour configurations with some similarity to verbal cohesive chains in their linear progression as described by Halliday and Hasan (1976) and by Hasan (1984). For example, dressing a character in a red coat constructs an association between a particular character and a red coat. It can remain an association or become an identity. In a crowd scene, a reader/viewer will be able to find the red coat and the associated character. A character?s red coat supports the following of that character and the continuity of the story. Establishing this association is the basis for further cohesive work across the pages of the narrative. If the red coat is the only red element in the visual narrative, then the red functions as an identity element, and its relation to the character is one of identity. Even if the red coat is not the only red element in the narrative, if the character always wears a red coat, then the red coat is an identifying element for that character. It could also be an associating element, as well as an identifying one, if a number of other characters also wear something red. Of course, if that is the case, it can be argued that it is the coat that is identifying the character and the redness of the coat is associated with a group of redness. Alternatively, the red coat can be cohesive with other red elements thus establishing additional cohesive ties and possibly constructing colour configurations that are like cohesive chains of similarity as well as identity chains. The use of red in these situations is as an associating resource, which is different from the colour red as identification. The term colour identification refers to things that are not similar, but precisely the same. Hasan?s (1984) categories of cohesive chains, those of identity and those of similarity, inspire a way to distinguish between colour as association and colour as identity (pp. 201-202). Colour as identity uses colour in a representational way, for it is colour identified with a specific narrative element. Colour association is tied to the notion of similarity. For example, in the film written by M. Night Shyamalan, The Sixth Sense, (Kennedy, Marshall, Mendel, & Shyamalan 1999) the colour red is associated with the presence of dead people. In the context of the film, it can be considered symbolic, as well as metaphoric and 133  interpersonal in its attracting attention to look for and follow its iterations. It is an association established early in the film and becomes a part of what is sometimes called a film?s ?image system? (Mercado, 2011). Colour associations in this discussion are those established within the context of a particular picturebook text. For example, in How to Heal a Broken Wing, there is a good deal of blue associated with the young boy, Will, and his family. The colour blue is cohesive with the blue of the sky. There are general colour associations used by illustrators and writers for symbolic and emotional reasons. Even though the focus here is on how the colour associations construct cohesive visual relationships, it is clear that creators can have additional reasons for selecting specific colours. Although a reader may or may not construe all the meanings of particular colour selections, more than likely they will construe the general visual colour cohesion.  Colour associations can be established to foreground specific kinds of engagement between key narrative characters. A good example of this kind of colour association is found in the book, How to Heal a Broken Wing. On the cover, which has an overall blue background, a spotlight-like circle of yellow lightness surrounds a small child holding a bird. This particular use of colour is the first of several similar renderings that create strong colour associations across the pages and highlights items of salience that support young readers? narrative comprehension of the story.  The use of a spotlight colour on the cover foreshadows and begins the pattern of highlighting the child and bird in a bright circle of light yellow. This lightness is a resource for foregrounding the connection between the child and the found bird. It is introduced on the cover and on the sixth double spread, and then reinforced in the next seven panels. The yellow spotlight is specific to instances in the city when the child and bird first meet and bond. In the story, the child and his mother take the bird home. In the home scenes the child and his parents work together to rescue the pigeon. However, in these scenes other colour dynamics come into play and foreground key actions. When it is time to return the pigeon to where it was found, Will and the pigeon are spotlighted. In the next scene, which is a bird?s eye view of the city square where Will releases the bird, Will and his parents are spotlighted 134  on the ground and the bird flying high above them is similarly spotlighted creating a visual cohesive tie. Colour Associations Iterations of colour associations can occur within any of the colour configurations, from backgrounds to colour flows. In some cases, the iterations of colour associations may simultaneously construct a reading path. I distinguish between the notions of colour association and reading paths because they have different functions. The iteration of colours in a reading path can include some instances of colour that have colour associations along with many others that do not but that nonetheless functionally support a reading path. For example in Where the Wild Things Are, it is possible to argue that the white of Max?s wolf suit and his four sets of claws associates white with claws and with all the claws of the ?wild things.? In fact, the white colour of teeth is a similar association. Thus, it is possible to construct a similarity chain that begins with Max in his wolf suit, which is colour cohesive with the claws and teeth of the wild things. Colour associations connect items by using colour as a source of similarity. To conclude, colour associations refer to things that may be similar but are not identical. Colour Identities Colour identifications can be basic, such as in the example of the red hat or having a character wear a particular colour of garment throughout the narrative. This kind of colour identity is frequently used in picturebooks and is primarily done to facilitate the reader/viewer?s following of a major character across the visual text. For instance, there are a good many red headed protagonists in children?s literature. They occur in higher proportion to the incidence of red heads in the world, presumably to provide salience to the protagonist.  5.5 Repetition as a Cohesive Resource The following discussion focuses on how repetition constructs narrative environments, narrative activities, and elements as well as the information structure?s New and Given.  135  5.5.1 Introduction Repetition works as a narrative cohesive process when it involves repeated iterations of the same or very similar items and patterned iterations of things or parts of things within a text. These kinds of cohesive repetition are based on relationships of sameness and similarity and are foundational to patterning, since fundamentally, patterning involves repeated iterations of the same or similar configurations of things. In a picturebook, text repetition is a key cohesive relationship constructed by deployments, of, for example, particular colours, as well as other compositional resources, relations, and processes. 5.5.2 Repetition and Context Context is central to the investigation of the meaning-making construction and construal of picturebooks. As discussed previously, in the case of language there are both material situational settings (which are the physical circumstances of the discourse) and the semiotic contexts of situation. In picturebook narratives, there can be a variety of material situational settings as well as a variety of semiotic contexts of situation. Also, in the case of picturebooks and picturebook reading there are the contexts situation within the narrative as well as the context of situation of reading it. In this analysis, I am interested in the contexts within the narrative, which are constructed by both the verbal and visual modes. 5.5.3 Repetition and Contexts of Situation In language, the construal of cohesion and coherence is supported by the recognition of familiar registers of language, material situational settings, semiotic contexts of situation, and narrative conventions and forms. Similarly, the construal of narrative sequential visual texts benefits from the recognition of these same communication features. In narrative picturebooks for the young readers, it is especially important to support their construal by the continuity of the visual cohesive resources deployed. This continuity is most often constructed by the visual repetitive rendering of key narrative elements, activities, and by the clear depiction of narrative material situational settings, which are the visual narrative environments particular to the text. 136  In picturebook texts, the young readers? recognition of visually conveyed contexts and language registers helps them make sense of the texts. They draw on their knowledge of the particular contexts and registers associated with their own lives, and with familiar forms of literature to construe the meaning of them. They also rely on the continuity of the visual depiction of the narrative material contexts particular to a text. For example, in Where the Wild Things Are, the verbal text names Max?s room. Repetition of the features in that material situational setting gives it continuity as it is pictured repeatedly during its transformation to ?the world all around? (Sendak, 1963). In addition to multiple contexts of situation, a text may contain one or more particular material situational settings or narrative environments. Repetition constructs the continuity and cohesion essential to the coherence and construal of both material contexts and contexts of situation. In picturebooks, visual repetition is the primary constructor of cohesive material contexts of situation, while verbal repetition is often the prime constructor of contexts of situation. However, in picturebooks for the young, visually depicted body gestures, facial expressions, and social relationships can provide additional accessible information about contexts of situations. As mentioned previously in this chapter, the difference between the two kinds of contexts sometimes involves a complex distinction: A picture book will typically involve several different contexts of situation, e.g. a child playing alone with a pet, maternal control/punishment, emergence of a dream-like material setting, sailing a boat. The ?boundaries? are sometimes complex to determine. I always start with one of the dimensions of field?the social activity?to get a sense of the meaning-making context. (Williams, personal communication, March 20, 2013) Colour and repetition are often resources and co-constructors of material situational settings and semiotic contexts of situation. The patterned use of colour, salience, and variations in style and/or texture can graphically define particular contexts to support ongoing construal of the distinction between contexts in a visual sequential text. Special highlighting or shifts in the palette or style of drawing or painting can be used to construct the variations of contexts 137  that may occur and be embedded in the narratives? material situational settings  and semiotic contexts of situation. 5.5.4 Repetition and Key Narrative Elements and Activities As previously noted, the visual repetition of the character and key elements and settings supports young readers in the following and construal of the development of essential narrative components. Repetition, as Mieke Bal wrote, ?is thus an important principle of the construction of the image of the character? (1985, p. 85). While Bal is referring to constructing a character in a written story, her principle is visible in film and picturebooks, especially picturebooks for young children where the image of the character is primarily constructed in the visual mode. The main characters in the books in this study are frequently visually iterated and constructed by a high degree of repetition in both language and image. Across the 40 pages of Where the Wild Things Are, the main character, Max, is iterated visually 19 times in the 18 scenes in the book. There are two protagonists in How to Heal a Broken Wing, which has 75 scenes. In those scenes the two protagonists are pictured together in 38 scenes. The little boy, Will, is iterated 55 times, while the bird is featured 51 times. This information is consistent with the convention of rendering main characters in at least every scene of a picturebook for young readers. That convention is supported by my analysis (Shoemaker, 2010) of 25 award-winning naturalistic narrative picturebooks for two-to-five year olds. Evidence that this convention persists is supported by my own continuing extensive review of narrative picturebooks for young readers. While the main character is almost always in the visual scene in the service of narrative continuity, this character is not always the most salient feature, because the character frequently functions as a Given. A recent book by Oliver Jeffers (2012), Stuck, demonstrates this phenomenon. The main character, Floyd, appears on the cover, the jacket flap, and the title page and then is rendered 36 times on the next 30 pages of the book. The main character?s repeated iterations construct Floyd as a key Given and construct cohesive ties, and chains across the entire text. In this book, the strong repetition of the main character is in contrast to the variety of colours, which do not provide general cohesion, as is so often the 138  case with colour. The repetition of the visual presence of Floyd in every scene along with the repeating rhyme holds the text together and provides the stable contrast for the New surprises the text offers each time Floyd does the unexpected with what the reader anticipates is the great solution to his problem To support the continuity and the following of narrative material in visual sequencing, illustrators often give the characters and settings easily identifiable and salient characteristics. Also, as previously discussed, it is critical for illustrators to render the characters and settings with precise continuity and accuracy of detail, work that relies on a high degree of repetition. Continuity, as Halliday and Hasan (1976) wrote, ?adds a further element that must be present in order for the discourse to come to life? (p. 299). As well, they point out, ?the continuity provided by cohesion . . . enables the reader or listener to supply all the missing pieces, all the components of the picture which are not present in the text but are necessary to its interpretation? (1976, p. 299). Cohesion is expressed by continuity particularly by the continuity of reference (Halliday & Hasan, 1976), which in the case of a picturebook is expressed by the repetition of visually rendered elements: the characters, settings, and in the placement of elements within the composition. Those renderings are expressed by the use of colour, line, dot, and texture, and sometimes made noticeable by formal and informal framing, position in the composition, the use of salience, and by consistent style in line work, point of view, and scale. My discussions have dealt with some of these resources in depth such as the use of colour and framing, and colour and salience. In the discussions of each book, examples are cited of the cohesive work of colour and repetition collaborating with line, point of view, and scale. 5.5.5 Repetition and the Information System, New and Given  In Chapter Four, I described Halliday?s explanation of the three metafunctions, the ideational or representational, the interpersonal and the textual, and made the point that the current study focuses on the textual component. That component is what Halliday (1976) calls ?the text-forming component in the linguistic system. This comprises the resources that language has for creating text . . . and for cohering within itself and with the context of situation? (p. 27). Halliday and Hasan (1976) argue that the textual component operates both within and 139  outside the grammar system. It draws on the information structure and on the cohesive resources as described in Chapter Four. The information structure orders the text, and the cohesive resources construct the semantic relations that make a text hang together. As Halliday and Hasan (1976) wrote: the information structure which is the ordering of the text, independently of its construction in terms of sentences, clauses and the like, into units of information  on the basis of the distinction into GIVEN and NEW: what the speaker is treating as information that is recoverable to the hearer (given) and what he is treating as non-recoverable (new). (p. 27) My interest in reiterating Halliday?s description of New and Given is to connect the role of repetition to the realization of Given in picturebook images. As Halliday and Hasan (1976) note there are overlaps in the work of cohesion and the information structure. That overlapping work is particularly evident in visual narrative texts. It is clearly evident in the work of visual repetition, which is essential to the rendering of visual instances of Given. In those instances of visual repetition rendering visual Given, the repetition is often simultaneously constructing a variety of cohesive ties, chains, and visual configurations. In the discussions of colour and repetition, there are several examples of the deployment of resources that simultaneously construct cohesion and New and Given. 5.5.6 Repetition and Continuity A highly visual sequential narrative that has a spare verbal text with few words requires a high level of visual continuity. A wordless sequential narrative requires the highest degree of visual continuity and consequently a high frequency of repetition of many kinds of visual resources but with considerable basic item-by-item repetitions. For example, in Shirley Hughes? (1979) 32-page wordless book Up and Up there are 132 scenes, and the main character, a little girl, appears in 116 of them. The addition of other modalities to a visual sequential narrative, such as written or spoken words, sounds, music, and/or movement can alter the required amount of item-by-item repetition. The reduction in item-by-item repetition depends upon the length and complexity 140  of the narrative and the number of scenes. So, for example, in Rosie?s Walk by Pat Hutchins (1968), a book of 32 pages and only 32 words, Rosie the hen is pictured on every double page spread. The book has only 17 scenes and Rosie appears in 16 of them. There is nearly a one-to-one ratio between scenes and iterations of Rosie as there is of the little girl in Up and Up.  However, Rosie?s Walk has a far simpler narrative than Up and Up, so it has fewer scenes.  In the next chapter, I will discuss my close analysis of the cohesive work of colour and repetition in two picturebooks using the frameworks and resources described in this chapter. 141  Chapter 6: A Close Analysis of the Cohesive Work of Colour and Repetition in Two Narrative Picturebooks. 6.1 Introduction This chapter begins with a description of each of the books, a summary of their narrative content, and notes on key features of their textual systems. Next, I report the colour findings followed by my findings on repetition. As I stated previously, of the prime visual cohesive resources of framing, position, style, and colour, colour is a powerful resource in making the cohesive role of repetition visible.  I employ an approach in my analysis that reflects my perspective as an illustrator, which is to begin with a review of how the textual systems might work in conveying the narrative content of the text at hand. After each narrative summary, I provide a description of the limited colour palette and then note the relevant colour configurations and how they construct cohesive relations across the visual text. Each of the books contains some, but not all, of the possible colour configurations. The discussions of colour configurations address those that substantially contribute to the book?s cohesion. Some discussion of colour?s role in constructing literariness is embedded in different parts of each book?s analysis. The variations from book to book highlight the picturebook format?s rich and varied colour deployment potential and how this diverse resource potential promotes unique literary expressions. For example, in Where the Wild Things Are, a good deal of the literariness of the text is constructed by patterns of colour association, whereas in How to Heal a Broken Wing that quality comes from the patterns of repetition.  6.2 Topographic, Typographic, and Narrative Environments 6.2.1 Where the Wild Things Are  Where the Wild Things Are (Sendak, 1963) tells the story of a young boy?s emotional journey. In a fit of mischief, the main protagonist, Max, yells at his mother who sends him to his room without any supper. While contemplating his sorry fate Max appears to go off into a 142  fantasy adventure. This kind of narrative is considered a circular journey in which the hero leaves home, faces a challenge, and returns home with new insight and knowledge. The story has the traditional Western narrative structure of a beginning, middle, and end. Among the narrative elements are the characters, which includes Max and a collection of ?wild things.? Key settings for the story are Max?s house, the ocean, and the place where the wild things are.  All the key characters (the protagonist and the wild things) are wild things and share their wild attributes. Max, the first wild thing is visually depicted on every double spread, a practice typical of picturebooks for young children since most of the character development and character activity is visually conveyed.  Across both the verbal and visual modalities, the wild things and their four key attributes are expressed prominently. Those attributes are the repeated statement ?EAT YOU UP? and the three physical features of teeth, claws, and yellow eyes. The dominant literary theme of wildness and wild things is reflected in the character of Max who might be considered to be the wildest of the creatures as well as the creator of the wild things with their terrible teeth, claws, and yellow eyes. Wildness, its physical attributes and appetites, is a resource for the construction of cohesion in this picturebook. Thus, many of the narrative themes, content, and activities relate to wildness and provide the meaning relations that many of the cohesive colour configurations in this book express. Topographic or material aspects of the book begin with its physical format. Where the Wild Things Are is a 40-page book, which is eight pages more than the typical picturebook length of 32 pages. An additional eight pages give the narrative four more double spreads and four more page-turns to support the drama of the narrative. Across the 40 pages, there are 18 narrative scenes within the story, another two on the cover, and one on the title page spread. A quick glance at its cover, endpapers, front matter, and the narrative pages immediately conveys general cohesive design and illustration. This is primarily due to the use of a limited colour palette, which is overlaid and damped down by crosshatched black line work. Cross-hatching renders dimensionality to the drawings, and adds texture to the colours, giving them 143  nuanced variations. Most of the shapes and colours that appear throughout the text are foreshadowed on the endpapers. In addition the endpapers introduce the colours in the book?s limited palette except the colour white used on depictions of Max and the teeth and claws of the wild things. A notable feature of Sendak?s verbal text is that individual sentences stretch across double spreads and onto subsequent spreads. This dynamic uses the tension of an unfinished sentence or clause to motivate the reader to turn the page and then another. Stretching a sentence across a sequence of pages constitutes a cohesive verbal dynamic and one that is supported by the colour cohesive configurations that also connect the pages in the story sequence. Since sentences stretch across multiple double spreads, I believe, there is a subtle expectation of continuity from one spread to the next. In this book, it is achieved by the limited colour palette, the flow of the cream coloured paper that is the background colour, and by the many iterations of Max.  The verbal text illustrates a key point about picture books, that words particularize images and vice versa (Nodelman, 1988). The words alone in this text evoke a far scarier version of the tale than the composite verbal and visual text portrays. The visual rendering of the wild things with toddler-proportioned bodies dials down their potential to scare because even with pointy claws and teeth, big-eyed baby-like creatures draw on our human hardwiring to respond positively to the young. The wild things are rendered in soft, damped down colours that make them blend in with their settings while Max is the bright light and often the most salient element in the composition. Typographic cohesion and tone for the book begin on the cover and maintain consistent patterns across the text. Throughout the book, the written text is set out in one, two, and three lines of printed text per page. Although a few double spreads have six lines of type and a few have three lines per page, for the most part the written text is spare. Setting type against the cream white colour of the background paper gives it prominence while consistently placing it on a plain background heightens the contrast of the verbal text pages to the three wordless full-bleed double-page spreads that portray the wild rumpus. Bold type provides a good balance to the black ink work in the illustrations. The colour, boldness, and position of the 144  type construct its strong presence on each page and across the pages. These qualities make important cohesive contributions to the overall layout and composition of the book. Later in this chapter, there will be additional comments of on the placement of the printed type and how it contributes to the visual textual cohesion. 6.2.2 How to Heal a Broken Wing How to Heal A Broken Wing (Graham, 2008) is a naturalistic dual-focus narrative in which two protagonists meet, experience a challenge, solve the problem, and happily go off separately. It is about love and care with twists and hints of multiple meaning. The story begins on the endpapers with the image of a grey bird flying through a grey sky. The bird flies into a city of glass skyscrapers on the half title and title pages. Confusing the real sky for the sky and clouds reflected in the glass fa?ades it hits building. Falling to the ground, the bird enters the environment of city folk rushing by while it  flounders on the pavement. Simultaneously, a young boy named Will emerges from the underground, enters the street level cluttered with pedestrians, and finds the injured pigeon. Will and his mother take the bird home where he and his parents nurse the bird back to health. When the bird is well they return the bird to the spot where Will found it. Will releases the bird, which soars above them, past skyscrapers, through a cloudy blue sky, and into the high grey sky of the endpapers.  This edition is a forty page hardback book with printed endpapers. In view of the dual focus narrative, the additional eight pages serve the text well by providing space for the multiple narrative settings. Within the 40 pages, Graham shows and tells the story over a sequence of 75 scenes using the comic book convention of paneling and the device of continuous narrative illustration. The book measures 8.5? wide by 12? high. Its large format, height, and portrait orientation thematically foreground the height of the rendered skyscrapers and skyline.  The full colour illustrations are drawn in pen, watercolour, and chalk. The book was typeset in Stempel Schneidler Light font. Graham adds a note at the end of the book, ?with thanks to Rosie for her fabulous title lettering? (2008) acknowledging the hand lettering of the main title on the cover, half title, and full title pages. The verbal text is spare, a total of eight 145  sentences (81words) spread across the 36 pages of narrative. Most of the sentences reach across several double spreads. There are four wordless double spreads. The verbal text is cohesive and coherent on its own. However, the important gap, the near total lack of descriptive language in the text, allows the characters and settings to develop visually. The spare language requires well-visualized material situational settings and social contexts of situation to both deepen the narrative and to transform the spare verbal content into a seamlessly rendered picturebook text.   Graham creates an integrated text using a mixture of full page, full colour, four-sided bleeds, single pages, full bleed compositions, double-page spreads of panels, and single pages of panels. Although Graham uses panels, they are the only graphic novel convention employed. Stylistically, Graham uses pen line work with watery watercolour and a touch of chalk-work to highlight small areas. A consistent drawing style is used throughout the book, which creates continuity of characters and settings. While cartoonish, the drawing portrays elements and settings sufficiently realistic for young children to recognize easily. General cohesion is constructed by a logical and tightly rendered sequential visualization of the narrative events. Since it is a dual focus narrative, it begins with 12 iterations of the bird on the first six double spreads (including the endpapers). The second protagonist, Will, is introduced on the seventh spread, where he is iterated seven times in a series of panels, and then once more in the scene in which he first sees the bird. A number of limited colour palettes construct general cohesion and some very specific embedded narrative environments. The sequencing of action and iteration of the characters is the most powerful producer of general cohesion. It is constructed by a series of easy-to-track views and the strong continuity of Graham?s drawing style. Across all the pages of the book, the spare text of the 81 words is positioned as a running caption-like strip with, on average, four to six words per spread. Only the opening and closing spreads have significantly more, with 14 and 11 words respectively. The weight of the typeface balances perfectly with the hand-drawn panel lines, keeping the type from leaping off the page; instead, it sits in place with sufficient white space surrounding it to promote legibility. 146  6.3 Colour Cohesion Colour deployments can create immediate general cohesion and make other cohesive resources such as repetition visible. These dynamics constitute the rationale for beginning this discussion of findings with colour?s work. 6.3.1 Colour Cohesion in Where the Wild Things Are Two resources, the text?s limited colour palette and the crosshatched rendering style, give immediate general visual cohesion to the Where the Wild Things Are. Beginning on the cover, endpapers, and full-title spread, these resources, along with several significant colour associations and identity configurations, construct cohesive relationships throughout the text. The limited colour palette consists of yellows and yellow ochres, pinky reds and pinks, sea- and sky-blues, leafy greens, white, and shades of brown and black in the ink lines and printed type. Cross-hatching damps down the limited colour palette, gives its iterations greater dimension, continuity, and nuance, while providing a second general cohesive factor. In addition, cross-hatching line work on everything but the sky and ocean, gives a cohesive texture to all the visuals. Colour Association and Identity  Where the Wild Things Are has several key colour associations, some of which also construct colour identities. Each of the wild things can be identified by particular colours and attributes which associate them with other wild things and other elements in the text. Some of their colours put them into colour associations while simultaneously identifying them with the colour of the boat and of two spiky tree tops first pictured on the cover.  Variations of these two colours, the pink and pinky red, are also used on the end papers  on spiky organic shapes that look like crosses between leaves and flowers. Here, as throughout the book, cross-hatching damps down the colour. Next, the pinky red colour is used on the noses of two wild things, both of whom appear on the full title spread with Max who has a very pink face. One of the wild things also has pinky red stripes on his upper body. In the first illustration of the book, there are pale pink flowers on the fabric of Max?s tent and his face is a pinky red as it is in the second illustration of Max with a pink face on the third spread, which depicts him 147  being sent to his room. The only other pink item in the room is the bedspread. Later in the book several other elements are associated with pink, including the bowl of supper on the table when Mac returns home, a few pink plants on the ground, and the pink sky in the scene where Max realizes that he wants to go home. A collection of pink colours constructs several related sets of colour associations that create visually cohesive colour configurations. These are a bit like the verbal cohesive chains described by Halliday and Hasan (Halliday & Hasan, 1976, Hasan, 1984). Within the pink Figure 7 A collection of wild things arranged by colour association 148  associations, there are subsets: elements Max creates and those his mother creates or works with. These are all text-bound associations and text-bound ties. Semantically they are related to Max within this text only. As such, what at first appears to be a simple visual sequence of pink associations is, in fact, a far more complex set of cohesive relations that includes interactions between sets of cohesive relations. Beginning with individual colour beats the configurations of pink associations expand from page to page until, on the double spread picturing Max at the point he decides to return home, the sky is filled with pink. In this scene the sky is a flourish of pink, as if a full orchestra were playing music in colour. In another scene, Max steps into his pink boat and sails back to his warm pink bedroom where his supper is still hot. Although, the pink renderings are never the most salient on the page, they construct a sequence of visual colour associations, which possess semantic relationships.  An initial examination of the pink colour associations on these pages reveals a sequence of colour beats connecting the elements from Max?s day-to-day world with elements in his imaginary world. Max?s creations and his adventures with the wild things are within his imagination. Evidence for that conclusion is provided by Max?s drawing of a wild thing tacked to the wall in the book?s second illustration Here my interest is in how two sets of colour associations construct semantic cohesion that connects aspects of Max?s life particularly his inner imaginative emotional life with his mother and day-to-day home life. The first pink item is the boat on the cover that will transport Max to and from the land of the wild things. Also pink are the spiky treetops in this new space as well as the noses and other areas on some of the wild things, Max?s bedspread, and the bowl with Max?s hot supper. In the collection of pink items that includes the boat, parts of the wild things, the sky, and the leaves, the nature of their similarity is that they are all things Max constructs. He is related to them through his agency. Another pink set, the pink bed spread and pink bowl of food are related by being part of his home material context?or material situational setting (Hasan, 1985)?and by close association with the narrative material circumstances of supper, bed, and his mother, which in turn are related to the agency of his mother. Even without cross-referencing these visual cohesive associations to the text?s verbal cohesive chains, the visual 149  cohesion created by the colour pink makes visible, albeit subtly and poetically, the cohesive relationships in the multimodal text. The second major colour association, a set of salient white items that construct cohesive relationships and make visible another strand of poetic rhythm. The initial iterations of white are the horns, white claws, and the white snout of the wild thing on the cover. There is also a white spiky treetop, although the use of cross-hatching damps the colour to more of a grey. On the cover, the white elements are all spiky. The next image with white is the full title spread with two wild things and Max. Those two have white claws, one has white spiky horns, and they both have spiky teeth poking out of closed mouths. To the right of them is Max in his white wolf suit with its white spiky claws. Moreover, there are the whites of his eyes. So, between the cover and the full title there is a flow of colour associations generated by the white spikiness in the claws and horns of the wild things and Max. While the white colour association of Max?s wolf suit is immediately visible in its textual function, it is also functioning interpersonally as a highly salient element and ideationally showing specifically what Max looks like and how he is dressed. The wolf suit with its claws is cohesive with all the other iterations of claws as a kind of animal that has claws. Thus, the white set of colour associations also participates in a cohesive association of clawness, The set of cohesive white colour associations is interactive with the set of associations of claws and spiky things. Interaction between the two cohesive chains or associations adds to the cohesive harmony of a text and to its literariness (Hasan, 1984). Colour Associations and Literary Texts  Sendak?s Where the Wild Things Are is brilliant in good measure due to the proliferation of visually realized rhythms and themes in the text. This brilliance is more than the carefully selected and limited colour palette, more than the pink associations and the white associations; it is a product of a kind of echoing between sets of cohesive configurations such as that of the white colour associations with the spiky associations. The book?s literariness is related to the way the specific pink associations and the specific white associations visually render key literary rhythms and themes. From the standpoint of constructing a balanced composition in each illustration, another illustrator might have used bits of white to create 150  points of interest, paths of salience, and other possible dynamics. However, Sendak?s choice of salient white for all the teeth of all the wild things (including Max) and for the pointy claws, constructs a consistent cohesive salient association of wildness with his toothy ?EAT YOU UP? (Sendak, 1963) declaration. The salience of the white attracts attention to the pointy teeth and claws so that a viewer follows the sequence of them. If Sendak had given all the monsters different kinds of teeth in different colours, shapes, and spacing, these choices would construct an entirely different meaning. Not only would these different choices materially change the meaning and relationship of the wild things to each other, they would change the connection between the wild things and Max. This connection is a subtle one based on Max?s agency in imagining them. All the wild things except for Max are rendered with the same line  of teeth as Max drew  on the figure tacked on the wall in the second scene of the. Max?s drawing visually reveals him as the creator of his adventures with the wild things. This subtle visual indication of Max?s authorship illustrates the picturebook format?s potential for narrative complexity and reveals the literary conventions of foreshadowing and relevant detailing for young readers to learn and to naturalize. Observing the varied collection of pink associations is also an example of how the analysis of a visual cohesive resource in a text may lead the way to understanding cross modal cohesion and provides evidence that such a thing exists. Hasan (1984) postulates  that there can be echoes of  functional relations in looking at cohesive chains in verbal texts.  In this multimodal text there appear to be similar echoes between cohesive sequences in each modality.  One of the semantic relationships between the wild things and Max is that they are all wild creatures with spiky claws. Another relationship is that the wild things with their claws, teeth, and horns are Max?s creations. Visually the claws, teeth, and horns share the same colour and the attribute of spikiness. These relationships echo across the pages, since these claws, teeth, and horns are visually cohesive as white spiky things that form patterns of salient renderings echoing repeated verbal descriptions of ?terrible claws.? Sequentially through the text, they are also cohesive verbally with phrases such as all the ?terrible teeth? of the other wild things and with more ?terrible claws.? Visually claws and teeth are cohesive with the other white elements in the text such as Max?s wolf suit, the whites of his eyes, a 151  dog, Max?s pillows, depictions of the moon, white dots in the sky, white clouds, a bit of foam on the sea, a puff of smoke from a wild thing, a small goat-like wild thing, and on the very last page, a glass of milk. From a visual perspective the white things are rendered across the pages as many small salient beats. Visually their salience attracts attention while highlighting important thematic elements, and simultaneously constructing patterns of white beats that render liveliness to the visual narrative activity on the pages and provides some of the general colour cohesion. There is a rhythm to the beats as the volume of the salient white areas varies. Thus, some of the white, especially renderings of Max in his wolf suit, have a solid, dominant presence while most of the other white elements, namely the many teeth and claws, are like flutters of high quick beats that could be experienced as echoes of spikiness, yielding yet another relational echo between sequences of cohesive resources. A third set of colour associations is composed of iterations of smudgy yellow ochre. As with the other colour sets, this set also begins on the cover in the colour of the sails and some foliage and on the end papers, where two highly salient yellow ochre spiky leaf/flowers dominate the pattern of shapes. Then, on the full title spread, two large wild things look over at Max. By size, shape, and colour, these two wild things are the most salient features of the spread. Not only are their eyes yellow, but their bodies are rendered in yellow ochre. On the next spread, the first story spread, Max is standing on two ochre books. The next illustration has no yellow but the following one depicting Max?s room, shows a pale yellow carpet, which over the next three illustrations develops into a yellow patch of grassy ground. Max in his boat with its yellow sails is featured on the next three spreads.  The wild things form a cohesive band through their colour associations, particularly with their yellow bits, which are also cohesive with Max?s crown, boat, and other elements in his bedroom. While each wild thing has a particular colour identity, it is their colour associations, together with their generally consistent size and proportions that construct their cohesion as a group and their cohesive relations to other features in the text. In the double spread of Max?s arrival, three of the four wild things are predominantly yellow ochre and all four have yellow eyes. Of the ten different wild things in the text (not including Max), all have yellow eyes and six have ochre colouring. The wild things are strongly associated with yellow ochre, pinky red, and white for their teeth, claws, and horns. The yellow ochre allows 152  the white of Max?s wolf suit and all the teeth and claws to have salience in each spread. Although the wild things are significantly larger than Max, their damped down yellow ochre, greys, and neutral brown colours do not distract from the patterned salience of teeth, claws, eyes, and Max, ?king of all wild things? (Sendak, 1963) wearing his golden ochre crown. Max sails off in his boat with yellow sails  into his warm, pinky red and ochre room. The palette of colours used on the wild things draws from the same limited palette used to portray Max?s room. As such, the colours of the wild things reflect or echo Max?s home context of his bedroom. Another key cohesive colour association comes from the black of the typography and of the black ink crosshatching, that is, the black line work that defines, shapes, and fine-tunes the textures of the wild things, Max, and his narrative settings. As mentioned previously, crosshatching can damp down a colour and render a shape with more dimensionality by giving it shadows and defining its outline. Colour and Reading Paths In Where the Wild Things Are, there are two major colour-reading paths?one white and one black. The white one is constructed by repeated salient renderings of Max?s wolf suit, and by the whites of teeth, claws, and horns. The second path, especially for the reader of the verbal text, is constructed by the typography, which is a flow of black type across the pages of the book. The white reading path, since it contains Max, also constructs the main character with its many visual iterations of him. The black reading path uses the printed type to direct the eyes from word to word. Also, the position of the text on those pages that have a panoramic view, allows a proficient reader to read the text almost simultaneously while viewing the image, making the verbal text seem like a narrator?s voice over. Colour Flows  In some illustrated books colour cohesively flows across pages, sometimes across every page. In those books, the colour flow configurations construct cohesion as a directional process. As Halliday and Hasan (1976) explain, ?A text unfolds in real time, and directionality is built into it; hence, of the two elements embodying the cohesive relation, one 153  always follows the other? (p. 19). However, in Where the Wild Things Are except for the considerable colour flow of the creamy white of the paper, it is the colour within each of the specific narrative environments that gives these settings continuity and some cohesive connections to other textual features.  The flow of the creamy white page colour is a significant cohesive resource, with the exception of the three wordless, four-sided full-bleed illustrations of the ?wild rumpus? (Sendak, 1963). Whether or not the creamy white of the page is labeled a colour, a neutral, a basic background, or a canvas for the entire rendering of the text, the fact is that it constructs a substantial flow of colour that frames all but three illustrations and the endpapers. It gives the verbal text a plain unencumbered surface, which makes the text pop out and clearly differentiates the verbal textual environment from the visual environment. It also constructs a contrast to the image, framing it and giving it attention.   The verbal textual environment is a strong contrast to the image environments. This is an interesting dynamic in the face of the definition of the picturebook as an integration of text and image. I would like to suggest a couple of reasons why, in spite of the clear differentiation between verbal text and image on the printed pages of the book, the words and images are well integrated. First, the white flow of the paper creates a seamless flow of text across all pages but the wordless interlude. This flow forms a kind of narration. In practice, most often the reading of this book is as a read aloud while young children view it. As a viewer follows the flow of creamy background, the framing of the images is highlighted, drawing the viewer into Max?s world and deep into his imagination as the images begin to bleed off the pages. The text for many young children is an auditory one, as they listen to words read from the bottom track of the pages as Max enters the space of the wild things. The transition from illustrated pages that include printed text to full bleed wordless double spreads encourages the reader/viewer to allow herself to be drawn into the visual narrative environment because there is no frame or text between the reader the visual environment. A sequence of three wordless double spreads provides time for the reader to enter ?the wild rumpus? (Sendak, 1963). When the page turns to reveal a double spread with printed text, the text and its creamy white background bring the reader out of the rumpus and back into the role of audience member. While the printed text is not integrated into the image, it is 154  nonetheless important in its role of positioning the reader in relation to the text. In some picturebook texts typography becomes an integral part of the image. In this kind of integrated role, the reader may engage with the text more interactively..  In Where the Wild Thing Are, the verbal and visual texts are differentiated on the page by their background colours, but there are cohesive ties that connect them. In keeping with the colour focus of this chapter, one of the key cohesive ties is the black colour of the typography, which is cohesive with the black lines and black cross-hatching. The black weight of the verbal text ties to the black lines that define the shapes and textures in the image. This is because, while the verbal text and images are clearly differentiated, they are still located within the composition of each double page spread, or adjacent page. Even if a viewer cannot read the words, he or she can see lines of patterned black marks positioned to relate to the image visually. While the integration of verbal and visual texts begins at the semantic strata in a classic picturebook format, meaning making relies on the reader construing the integrated meaning from what is expressed on the pages. To construe an integrated text from the two modalities depends on the skill of the author/illustrator to construct a flow of cohesive relationships across the modes by utilizing the most appropriate mode at hand as in orchestrating a piece of music. An integrated expression is one that expresses meaning in the mode best suited to its content. Thus, descriptive information would be best expressed visually with verbal information such as sounds and speech, even inner thoughts, expressed by the auditory and/or typographic mode. Colour and the Construction of Cohesive Narrative Environments  As has been previously discussed, the verbal and visual environments are clearly differentiated in this work. Throughout the book, the verbal has a consistent typographic environment. However, the visual environment of images varies significantly because an imaginary environment of ?where the wild things are? is embedded within the overall environment. Additionally multiple material situational settings and semiotic contexts of situation challenge the construction of overall cohesion and coherence. These challenges are overcome by the general cohesive work of a limited palette, the cross hatching, and by the intentional framing and unframing of embedded environments. It is a challenge to construct 155  colour cohesion from setting to setting. In these situations colour cohesion could be a contextually contained cohesive process; however, in this book, the activities of the main character across settings link them logically and metaphorically together. Narrative characters can provide cohesive ties across settings, particularly in picturebooks for young children, because as previously stated, the main character usually appears on every page or on at least every double spread. This is certainly the case in Where the Wild Things Are because Max is present on every double spread. In addition, Max?s white wolf suit is cohesive with his white pillowcase and the glass of milk in his bedroom environment. Then, in the material situational setting of the wild things, the white colour is cohesive with their white teeth, claws, and horns. Max?s white wolf suit carries the potential for making cohesive ties and for constructing an element of continuity from page to pages since he is the main element the reader follows across the narrative Material Situational Settings There are two major, visually illustrated narrative environments in Where the Wild Things Are and they are the main material situational settings in the narrative. First is Max?s room and the second is the place where the wild things are. Even though they are very different settings, they contain several cohesive colour ties that relate one to the other. These cohesive ties are critical to expressing the semantic relationship between the two places. One setting is a place others have provide for Max and the other a place Max creates. Max?s construction is supported by the dynamic of repetition, which will be discussed in the next part of this chapter. The initial fabrication of his other place is rendered visually tree by tree on each illustrated page across a sequence of four double spreads. These renderings express the transformation of Max?s room into forest. It is left entirely to the viewer to construe from the evidence in the image that the transformation is Max?s because the text only states, ?a forest grew and grew? (Sendak, 1963, np). Visual clues to Max?s imaginative agency come from the cohesive colour ties because they construct semantic links between the two settings. Cohesive semantic links connect flows of colour associations and identities across the narrative?s intra-textual environments. For example, there is the connection between the pinky red bedspread and the pinky red foliage. Another example is the set of cohesive colour 156  ties connecting the white of Max?s wolf suit with the white of his pillows, the white of the teeth and white claws of the wild things, and with the white moon.  The tie with the moon raises a question about Max?s compositional relationship to the moon in each of the images. Are the ties clues of Max?s agency in his night journey? If this is Max?s imaginary journey, his construction and not someone else?s, then it makes sense that he constructs an environment that draws on the resources at hand, which include the colours in his home.  Just as Max might select a pinky red crayon to draw the foliage in a picture of ?where the wild things are,? in his imagination he draws on the colour resources within his personal context. This notion of Max drawing on the personal colour palette at hand is consistent with Sendak?s illustrations of the wild things, because he renders them on the pages in a manner that is cohesive with Max?s drawing (as it appears in the second illustration of the text). Max is like Harold in Harold and the Purple Crayon (Johnson, 1955 ) because, like Harold, Max is imagining his adventure. The difference between Harold and Max is that Harold is rendered holding a crayon and drawing on the book pages, which meta-fictively calls attention to his agency, whereas Max is rendered within the environment he is imagining. The reader/viewer construes that it is Max?s imagined place by drawing on the clues in his drawing  such as the horns and the mouth of teeth characteristic of  almost every wild thing. The reader/ viewer is given evidence to differentiate between the narrator?s depiction of Max?s everyday environment and the one he Max imagines, which is the embedded environment of the wild things that is tied to the everyday world in the manner of dreams that use sights, sounds, and feelings from daily life. As for the cohesion between Max?s everyday environment and the imagined space where the wild things are I am positing that besides the general cohesion constructed by the limited colour palette and the particular colour associations, that there is a semantic relationship between the two narrative environments. I base this assertion on evidence of Max?s agency in constructing a personal imagined material situational setting out of the resources of his personal daily material situational setting. The semantic ties between colour associated items 157  found in those two narrative settings serve to connect Max?s imaginative play with his drawing, bringing to mind the kind of imaginative play described by Kress in Before Writing (1997). This is where a child?s drawing on paper, is cut-out and incorporated by the child into her own playful  material situational setting leading to deep imaginative play.  6.3.2 Colour Cohesion in How to Heal a Broken Wing Colour Associations and Colour Identities In this book, colour associations and colour identities construct characters, their activities, and their settings. It begins with the cover, which features a key colour association, a spotlight of pale yellow lightness around a small boy holding a pigeon. This spotlight is an important colour association that constructs salience several times in the narrative On the cover the spotlight?s appearance foreshadows its presence later in the text. However, on the cover, it primarily draws attention to the child and the bird, in what appears to be a marketing ploy to show the book?s content. Oddly, the cover looks ?colourized? like the colour versions of old black and white movies. My suspicion is that a marketing person suggested that the colours on the original illustration needed to be brightened to give the cover a bit of salience. Although the spotlight on the child and pigeon serve the text?s meaning construction, the ?colourization? on the cover is at odds with the limited colour palettes used in the interior. Colour Associations and Narrative Environments The subtle use of colour association constructs relationships between characters and material situational settings. For example, there are colour associations between the pigeon?s association with grey, grey clouds, grey buildings, the grey of its own feathers, grey sidewalk, and the grey pedestrians. However, once Will notices the pigeon, the colours of the pedestrians and their background warm to a pale rosy brown and the clothing colours transform into warm tans. Will associates with rosiness. In contrast to the pigeon?s greyness, Will with his bright red-brown hair, is dressed in a bright red jacket, bright blue striped pants, and bright tan shoes. When Will arrives on the scene, the pigeon?s world warms up and becomes rosy as Will 158  carries it home to his family. Will and his family are also associated with blueness, the blue of the sky, of their clothing, and of the interior of Will?s home. Throughout the book, Will is clearly identified by his clothing and red-brown hair. Thus, picturing Will wearing the same pants, is a key resource for building the continuity of his character.. Continuity supports both the text?s cohesion and the easy-to-follow series of character iterations.  The book contains 75 scenes across the 40 pages of text with 75 opportunities for salience, which Graham constructs in nuanced variation, using colour in scenes with Will, and scale and position in scenes with the pigeon. By using scale and position instead of colour to construct the bird?s salience, Graham retains the softness and lightness of the pigeon, while keeping its prominence. These two kinds of salience, contrast Will?s grounded presence on earth and give the pigeon lightness for its life in the sky. Regular iterations of those two kinds of salience reinforce both the continuity and the linearity of the text. There are many scenes and many instances of salience highlighting the sequential actions of the two main characters. Because logical physical sequencing is established and maintained by reducing the gaps between actions, a close relationship between continuity and linearity is vividly illustrated. This story begins and ends with the grey pigeon flying through a high grey background of clouds. There is little contrast between the pigeon and the background until the pigeon flies into a white background with high glass buildings. This setting of blue sky and clouds  next appears as a reflection on a wall of glass. In five separate scenes, the sky and cloud scape is similarly pictured. The embedded blue sky and cloud material situational settings have high salience in each scene they appear in, which serves to highlight the pigeon?s plight. In two other scenes, one in the middle of the book and the very last scene, there is a blue sky with white clouds bleeding off the page and presented as a full material situational setting. This setting book ends the pigeon?s circular narrative. Will finds the pigeon in another narrative setting, which is rendered at foot level with a view of pedestrians from their feet to their mid torsos. Its grey and dull, grey-brown colour palette contrasts with Will?s bright red jacket and blue pants. This dull grey and brown material situational setting warms up to rose, warm grey, and warm brown once Will has the pigeon in hand, ready to take it home to heal. 159  A warm rose colour flows around Will as he travels home and fills the home environment (except for a couple of evening scenes that are bathed in pale blue). The rosy colour washes around Will and his family when the bird?s wing is healed and in the scene where they return him to the city square where he landed weeks before. In that next-to-final-scene, as Will releases the pigeon, the buildings around the square are washed with a pale rose colour. In the scenes just described, the rose shade constructs a colour association rather than an environment. It washes into a number of narrative environments but its textual work is as an association, albeit a background association.  Colour renders most of the salience in this book the majority of which relates to Will. The iterations of Will?s salience pave a major reading path that also constructs Will?s character by attracting the attention that draws viewers to follow him through the narrative.  6.4 Repetition as a Cohesive Resource in the Book Repetition is essential to the creation of patterns of cohesive resource deployment, which is the dynamic that produces the text?s overall rhythm and cohesion. The patterned instances of cohesive resource deployments will now be explored in the two books. 6.4.1 Repetition in Where the Wild Things Are Repetition and the Construction of Narrative Environments A sleeping beast with human feet sits among the trees beside a small body of water where a boat is anchored. This first material situational setting introduced in the book, Where the Wild Things Are (Sendak, 1963), is rendered on the cover of the book. Whether or not a reader/viewer has any previous knowledge of such a creature, what can be construed is that the setting must be the place where the wild things are, since that is the title printed directly above the image. The title constitutes the verbal theme, which is co-articulated by the picture of where the wild things are. This first rendering of a sleeping wild thing introduces a 160  number of narrative elements that will be iterated repeatedly throughout the book.7 An initial textual question is: What is the purpose of showing that scene on the cover? In this context of exploring repetition, the next question is: How does the rendering of this scene support the construction of a cohesive text? In addition to introducing the theme and key narrative elements, this scene foreshadows some of the patterning of visual elements and colours that will be iterated throughout the rest of the text. The title and image work together to present the theme in much the same way that the title In the Night Kitchen (Sendak,1963), together with the image of Mickey, presents the book?s primary theme of Mickey in the night. In Where the Wild Things Are, the cover title and image present the narrative theme of ?where the wild things are.? Describing the narrative theme as ?where the wild things? is a literal description of the book?s narrative theme but it is nonetheless highly plausible. It is not relevant to this discussion to explore the complex meanings of this book?s theme, but it is important to acknowledge that, in this case, it is highly nuanced, as some of the discussion of repetition will reveal.  Instances of visual foreshadowing on the cover initiate a series of visual repetitions. Similarly, the use of a visual leitmotif begins on the cover and will be visually iterated numerous times. By definition a leitmotif is a recurring item that relates to a theme and as such entails repetition. In this case, I am calling the spiky shaped leaf of Sendak?s flora (the pink spiky treetops and the small pink ground level versions) the leitmotif for this book. Visually this leitmotif is particular to this book and to the place where the wild things are. In a review of Sendak?s other illustrated works, I could find no other trees or plants rendered in this fashion. In Where the Wild Things Are, Sendak uses a collection of the leitmotifs for the endpapers. The spikiness is visually related to or repeatedly associated with spiky horns, teeth, and claws. Using the cover, endpapers, and title page spreads to foreshadow and                                                  7 Although there is a long tradition of using an illustration from inside the book for the cover of an illustrated book, one of the innovations in the evolution of the picturebook has been the use of the cover, endpapers, and title pages to introduce key thematic elements and often to foreshadow narrative elements. In view of these innovations, I believe that it is legitimate to consider the narrative intention for the cover?s content. 161  introduce narrative settings, characters, and leitmotifs in this way, has become common practice in picturebooks. In most cases, the practice is an economical and conventional use of space when there are only 16 to 20 page openings available in the standard picturebook. Three Key Material Situational Settings  Where the Wild Things Are has three key narrative environments what I have also called material situational settings: Max?s room, Max?s transformed forest room, and the place where the wild things are?a kind of island. In fact there is a fourth environment  but it is a transitional space, the ocean that ?tumbles by? and on which Max sails ?in and out of weeks? (Sendak, 1963) from the environment in his room to where the wild things live and back again. For this discussion, the focus is on the three environments in which Max is actively engaged with others. All but Max?s room is an imaginative material situational setting invented by Max. However, they all contain similar contexts of situation but with role reversals. In Max?s imagined narrative environments he takes on the role of caregiver, thus recreating the semiotic context of situation of his home in a different material situational setting. The repetition of key features in the layout of the scenes constructs the continuity of these material situational settings necessary to the critical transformation of Max?s room to ?the world all around? (Sendak,1963), which is the narrative environment of Max?s imagined voyage and visit to the place where the wild things live. As well, the repetition of language patterns constructs the slightly different semiotic contexts of situation. Although the discussion of language patterns is not the focus of this study, it is noted here because its variations collaborate with visual variations in the narrative environments.  The repetition of key features plays a role in continuity and the transformation of Max?s room. Sendak renders Max playing and transforming with what is at hand. This is the natural work of young children (Kress, 1997). Max is a young creator, who makes his world and his creatures in his own image from what he knows and can shape with what is at hand, which in this case, is what is mentally available for him. Repetition helps construct and reconstruct these contexts and it calls attention to Max?s agency in their transformations. Sendak uses the 162  verbal text to call attention to the transformation of the material situation setting and to the New events and elements in the narrative.  When Max is sent to his sparsely furnished room after he behaves mischievously and yells rudely at his mother, the verbal text sets up the semiotic context of situation. She calls him a ?WILD THING? and he yells back, ?I?LL EAT YOU UP!? (Sendak, 1963). In addition to the verbal text conveying the context of situation, it simultaneously constructs Max?s character through instances of verbal repetition, particularly word-for-word or item-by-item repetition. In addition, Max is named or referenced with a pronoun in every sentence. Max is a source of continuity in both the verbal and visual texts because he is frequently and repeatedly iterated. Transforming a Material Situational Setting  Visually, the repetition of the image of Max is important to the transformation of his room to the ?world all around? (Sendak, 1963). Max is rendered in precisely the same location in the setting over a sequence of four illustrations. His presence in the same spot in each of the illustrations is in contrast to the changes taking place within his room. There are slight variations or animations of his body. In the first scene, his eyes are open and he scowls. In the next two scenes, his eyes are closed. In the fourth, he turns toward his new transformed world, a world like the one pictured on the cover of the book. The subtle changes in Max as he transforms his world create a sequence of stop-action-like animation. Also, this series of scenes of Max in his room as scene by scene the bed and doors transform into tree trunks and trees and plants grow up out of his carpet mimics the convention of animation in which all but one or two things remain motionless. That is not only an economy in animation drawing and filming, but also a source for the contrast of what remains the same and what changes, and is essential to foregrounding the changes. In this sequence of four scenes, the things that do not change are the relationship between Max?s position and the moon, the position of the viewer in relation to Max, Max?s wolf suit, Max?s size in relation to the whole scene, and Max?s position in relation to the vertical lines in each scene. What changes are the vertical lines that begin as the lines of a door and the vertical lines of the four corner posts of his bed that become tree trunks. 163  The pattern of tree trunks first pictured on the book cover continues throughout the book. The transformation of Max?s room to the ?world all around? (Sendak, 1963) from the doorframes, window frames, and bedposts turning into tree trunks continues the patterning of vertical lines and the visual rhythm of vertical beats that run through the text. The patterning of these vertical lines also constructs informal frames and is parallel to the formal frames around some of the images. Tree trunks  rendered in vertical patterns throughout the text construct a physical sense of stability which is further strengthened by the pattern of horizontal frame lines and the solid, horizontal lines of text. This construction draws on well-documented principles of visual stability (Arnheim, 1954, 1969; Bang, 1993). Although Max is off on a great adventure to the place where the wild things are, he creates a familiar social context of situation in his new material situational setting.  Figure 8 ?And a forest grew.? A sequence of illustrations 164  The patterning of vertical and horizontal lines, lines of text, and tree trunks constructs a stable environment and background for narrative activities that, in their renderings, add new layers of patterned salience and additional rhythms to the text. The patterning constructs particular material situational settings. The two visual material settings, Max?s bedroom and where the wild things are, have visual cohesive ties that indicate Max?s likely agency in transforming one to the other. Also, the patterning of solid horizontal and vertical lines is a contrast to the patterning of pointy claws and teeth of the solid short-legged wild things. The patterning of teeth and claws, rendered in what I call pizzicato (small) beats, is added to the solid hefty beat of wild things as they mimic Max?s gestures. There is also the contrast between the repeated iterations of white teeth and claws and the repeated non-threatening body gestures and positions of the wild things. This pattern of contrast is established from the cover where the wild thing slumbers. Next, on the title spread, two wild things cower away from Max, who aims his body aggressively toward theirs. In the initial iterations of Max, he is moving aggressively toward two wild things, in the next scene he is hammering, and then in the next scene Max is chasing a dog with a fork in hand. Three sequentially portrayed forms of aggressive behaviour construct Max as a wild thing from the title page spread and on the first two illustrations. This aggressive depiction is in contrast to the two sequential portrayals of the wild things as quiet and cowed. What these illustrations show is that in four sequential illustrations the behavioral social context of Figure 9 Two sketches of pizzicato beats and tree trunks 165  situation of the wild things and Max is well established by the repetition of visually rendered body gestures. Furthermore, as the narrative progresses, the repeated renderings of the body gestures will further express the nature of the characters and their relationships to each other.  Patterns of body gestures construct a character?s demeanor just as patterns in language use do. In the case at hand, the illustration of Max depicts him as the wild thing in charge of the other wild things. Besides providing logical cohesive sequencing, the continuity of contexts is essential in order to foreground any sort of activity within them. Gibson (1986) discovered that the way humans perceive the visual environment is to scan for variations within a stable setting. In a sequence of pictures, the basic setting must remain stable in order to highlight variation. That stability is attained by the meticulous rendered repetition of elements. Similarly, changes in a narrative element such as a character must be foregrounded by stability in the depiction of the character, so that slight changes in the character?s visual or verbal depiction will be noticed. Narrative activities and elements The repeated depiction of Max is very much like a word-for-word process of the most basic kind of repetition, while the construction of contexts involves other kinds of repetition. For example, construction of the bedroom and the island environments involves the rendering of vertical bedposts and trunks, both made of wood and therefore closely related. However, it is the vertical repetition that constructs a stable environment with continuity as well as a strong rhythm of informal framing across the text, a good contrast to the short, stout roundness of the wild things. The repetition of visual textual resources makes visible the kinds of visual compositional relations that operate similarly to verbal sense relations. For example, the repetition of vertical posts and trunks is the repetition of the principle of verticalness and verticalness that is stabilized by being set on a stable horizontal. Visual repetition, that is, an exact repetition of a visual item is, as previously noted, the most basic kind of repetition. However, in the case of repeating verticals, that repetition is of a similar compositional principle, and as such, 166  it constructs a similarity chain of cohesive relations. As well, the repetition of vertical posts and trunks constructs a stable environment, which in this case may offer a sense of comfort. Repetition and Salience In Where the Wild Things Are, repetition constructs characters and narrative environments using patterns of specific renderings of salience and framing. These repeating patterns of salience and framing simultaneously contribute to the cohesion and coherence of the text. In the book, patterns of salience construct a cohesive text in several ways. First, Sendak constructs rhythms of salient beats across the text. The main character, Max, is almost always a strong beat of colour salience. In addition, as mentioned previously, Sendak rendered small pizzicato beats of salience with the horns, teeth, and claws of the wild things. The pizzicato beats begin on the cover, continue on the full title spread, and return with Max?s encounter with a wild thing in the sea, and continue on each of the double spreads where wild things appear. Max?s claws are a continuation of his wolf suit, so they do not present themselves as separate small salient beats the way the claws of the wild things do. However, Max?s claws remain associatively cohesive with other claws in the story. Several other sets of repeated iterations construct rhythmic beats. Yellow ochre shapes construct a set of solid rhythmic beats. Moreover, patterns of foot and arm gestures construct additional rhythmic movements across the text. The foot and arm gestures do not have particular colour connections, but the high repetition of those particular gestures, together with the pizzicato beats of the claws and teeth, construct rhythmic movements across the visual sequences. Almost all the wild things have the same yellow eyes and so the repetition of them across each image and across the sequence of images creates a set of salient beats. The eyes are particularly powerful because they have directional vector gazes that shift back and forth producing the illusion of movement. Repetition and the Moon Sendak?s portrayal of the moon and its phases have inspired speculation as to whether Max?s adventures take place in dream time or real time,  bringing into questioning how much real time passes in the story. In any case, the moon appears nine times in varying degrees of 167  salience and in varying spatial relationships to Max. The story begins with a crescent moon of little salience and finishes on the last page with a full moon that has high salience. When the moon first appears it has a faint presence, just a hint of the outside but its last appearance is highly salient in close spatial relationship to the table with Max?s waiting, ?still hot? (Sendak, 1963) supper and with a degree of matching salience to Max?s suit as he stands to the far right of the image. In this image, Max with his wolf suit partially pushed back to reveal his dark hair, is a New Max, a Max returned. In Sendak?s work, everything is intentional, so it is fair to conclude that the repeated renderings of the moon are there to construct meaning albeit playful and purposely perplexing. At an informational level, it may be that depictions of the moon are intended primarily to situate the activities at night. Whether or not Max is gone for five minutes or five years, or whether he makes trips in between does not alter the meaning of his adventure. It may be that Sendak is playing with the moon and with readers to give them some fun to puzzle through. The moon becomes a cohesive resource and an element in a sequence of moon iterations that constructs what can be seen as an interaction between Max?s three material situational settings and manifests as a cohesive touchstone. In each of those environments, the moon functions ideationally as well as textually. The moon is significant because, besides Max, it is the only narrative element that appears in all three narrative environments. I may be stretching Hasan?s (1984) definition of interaction in claiming interaction between settings, but Hasan saw interaction as echoes of grammatical functions occurring between similarity chains and identify chains. My claim is that the moon as a textually cohesive feature that is a quiet echo from one setting to another. In the very last scene the highly salient moon, an element from Max?s journey, is still close to him in his New settled state. Also, the moon, saliently framed by the window, relates more vividly to the room, which seems less removed from the world than it was in the first part of the story. The moon and the full bleeds on the page do not frame or imprison Max as he had been when he was first sent to his room without his supper, nor do they distance the viewer from the scene. Framing the scene of Max returning to his room would distance him from the reader. However, possibly the more important reason for the full bleed of the illustration is to create a contrast to his framed, restrained placement in his room when he is sent there as a 168  punishment. When he returns home to a warm room with a hot dinner waiting for him he is not confined by a frame. Max?s emotional change is constructed by Sendak?s rendering of the room in warmer colour tones, full bleeds, and by reducing the space between the viewer and Max. This illustration has the continuity of repeat renderings of all elements previously portrayed in the room, but with more brightness and salience in colour and size. Thus, the dynamic of repetition constructs the continuity required to foreground the variations in brightness and salience, which express key meanings in the final illustration. The Information System and Salience The last scene of Max returning to find his hot supper demonstrates how repetition made visible by deployments of salience can support the information system of New and Given. The functional work of New and Given requires a certain amount of repetition to develop the Given and the contexts of situation. In visuals, salience is frequently deployed to attract attention to the New. However, salience also contributes to other compositional/textual work, so it must be considered in light of the context, particularly in sequential visual narrative texts.  In this text, the repetition of salience constructs rhythms that provide cohesion across sequences. Max is in every scene?he is a Given and often salient, even the most salient element in the image. Although Max as a character, a narrative element, is a Given, his behaviour may be the New or one of several New things in the text.  In a sequential visual narrative, the construction of New may require additional iterations of salient items provided by the verbal mode, by position on the page, and/or by scale or prominence. This is because a visual, due to its complex composition, may contain multiple opportunities or even discreet visual episodic areas that contain the New and thus more instances of salience. 169  The position of New and Given in a visual sequential narrative is sometimes different from what Kress and van Leeuwen (2006) found in single non-sequential images. Earlier in this work, I suggested that this difference was because certain physical/logical ideational factors come into play in a visual narrative. However, I would like to add that another reason is the potential use of salience to construct rhythm across a narrative text. Additionally, within an image there can be multiple New items depending upon the complexity of the composition and narrative and their intended meanings. The rationale related to rhythm construction is in keeping with the notion of tonic prominence that Halliday (1994), in terms of language, defines as ?the main pitch movement, the main fall or rise, or change of direction . . . that calls attention to itself? (1994, p. 296). Where the Wild Things Are (Sendak, 1963) contains numerous examples of salience constructing rhythm while at the same time calling attention to a New within an episode or several episodes of a composition. The notion of episodes within the composition of a painting is applied here in reference to O?Toole?s (1994) work in the Language of Displayed Art. Halliday makes the point that, ?tonic prominence?, can be indicated by a form of graphic prominence, [such as] by bold type for print (1994, p 296). In Where the Wild Things Are, Sendak (1963) uses bold type for prominence in the third page opening when Max?s mother calls him ?WILD THING? and he replies, ?I?LL EAT YOU UP!? In fact, the bold type of the Figure 10 Episodes in a composition 170  title, Where the Wild Things Are is the most salient element on the cover. Although everything on the cover of a book is New among all the New, some elements will be more salient than others. There are patterns of salience and New across Where the Wild Things Are that construct a rhythm of salient beats. As just mentioned, the cover contains all New information and it is the bold typeface of the title that is the most salient feature of the composition. On the cover, there are several small beats of salience in the horns and claws that foreshadow coming iterations. Opening the book to the endpapers there are organic plant shapes that are cohesive to the pinky palm trees on the cover. The salient bits are two yellow leaf shapes. Those two bits of salience begin a chain of yellow ochre colour associations that are cohesive with the yellow ochre on the two wild things rendered on the title spread. As should be expected of a title page in terms of design focus, though it is not always true, the title rendered in a bold black typeface is the most salient feature of the title spread. Formal and Informal Framing Patterns of framing, both formal and informal, have an important role in constructing the transformation from one material situational setting to another. The formal framing confines and focuses on Max?s particular behaviours. As the narrative progresses, the frames expand to include larger scenes until they finally bleed away on all four sides. The repetition of the progressive change in the formal framing promotes engagement with reader/viewers. There is also informal framing that constructs episodes within individual or double spread page compositions. An example of a kind of episodic informal framing is illustrated on the cover. Episodes can be created within a single composition by informal or formal framing. On the cover of Where the Wild Things Are, there is one episode that is constructed informally by two pink trees that frame the sleeping wild thing. Another episode similarly constructed by tree trunks contains most of the image of Max?s boat. Other examples of informal framing are rendered as the forest begins to grow in Max?s room. In these examples, tree trunks frame episodes within the composition. Later during the sequence of three wordless double spreads, Sendak uses tree trunks to informally frame Max and the wild things into separate episodes. 171  Patterns of informal framing create important cohesive ties and rhythms. The patterns form the choreography of character interactions in each scene as well as a choreography of the viewer?s interactions from scene to scene. There are many informal framings created by doorframes, bedposts, and tree trunks that form cohesive ties across the key settings. Also, there are informal framings created by the positions of Max in relation to the wild things. For example, when Max arrives at the place where the wild things are his position in his boat on the water separates him from the wild things on land. In the next scene his position on the far left of the two-page spread, is informally framed by two tree trunks and the presence of a bowing wild thing who separates Max from the collection of wild things. At the start of the wild rumpus Max is in the centre of the two-page spread and he is partially and informally framed by two tree trunks, and more distinctly informally framed by the two large wild things on either side of him.  After the wordless three scenes of rumpus, Max realizes it is time to return home. At this time, he is framed by the entrance to a small tent. This tent is cohesive with the tent Max builds in the first scene of the story. However, in the first scene he is outside the tent and to its right. In the scene with the wild things he sits inside the tent, framed by it, which is something from his home along with the little stool he is sitting on. Sendak (1963) uses informal framing to create cohesive relationships, as well as relationships between characters. He is able to choreograph all the patterns of cohesive resource deployments into this narrative picturebook making it into an integrated whole work of astounding complexity.  6.4.2 Repetition in How to Heal a Broken Wing This book contains multiple material situational settings previously cited in the analysis of colour in terms of the use of limited colour palettes and colour associations. In the discussion of colour, I also noted how this book exemplifies Kress and van Leuween?s (2006) notion of the integrating code of rhythm. The focus in the current discussion is on the process of repetition as a means of revealing how it constructs and contributes to the integrating rhythm of the text. 172 Narrative Contexts  Narrative structure sets the pace and inspiration for constructing this book?s integrating rhythm. It begins in a linear fashion with the bird high in the clouds just before it hits a window and falls to the ground. Next, the second protagonist, young Will, emerges from a subway and finds the injured bird on the ground. The story continues at ground level until the boy takes the bird down into the subway to travel to his home where he and his family care for the bird. Eventually, Will, the young boy, and his family travel back into town on the subway to return the bird to the plaza where Will found it. In the final scene, Will releases the healthy bird back into its high sky home. I mention the interesting pattern of locations from high in the sky to underground and back to the sky as a subtle example of the role of repetition and patterning in creating the book?s textual rhythm.  Since there are two protagonists, one human and one bird, it is reasonable to expect two different rhythms or two different thematic expressions related to their natures and native environments. There are eight material situational settings and events and within them, there are five semiotic contexts of situation. Their relationships can be considered in terms of the three key elements of a context of situation: the field, tenor, and mode, some of which are more stable and some of which are changeable. The field is what is happening; the meaning exchange or activity. The tenor is the interpersonal circumstance; that is, who is involved. The mode is the channel of communication, which here is a physical exchange as there is no verbal rendering of spoken words or thought. However, the reader can imagine a small child?s thoughts, feelings, and possible mutterings to the injured bird and to his mother. What changes most in this text are the narrative circumstances, which are often the material contexts of situation, which I discuss next. How to Heal a Broken Wing (Graham, 2008) provides an example of a multimodal narrative text containing multiple material situational settings and multiple social contexts of situation that illustrate the distinction between the two kinds of contexts. What significantly helps to bind all the contexts together into a coherent narrative are the cohesive resources of colour and repetition. First, there are eight specific material situational settings: the bird in the sky, the bird hitting the glass high-rise, the bird on the ground among pedestrians? feet, Will and 173  his mother coming into the city, Will rescuing the bird in the plaza, Will and his family caring for the bird in their home, Will and his family releasing the bird in the city, and finally the bird returning to the sky. In addition, there are different contexts of situation: the bird alone flying, the bird alone injured on the ground, the bird rescued, the bird cared for, the bird released, and then the bird alone flying as in the beginning of the story. Those contexts are separate because of the different nature of the activity, the field of the context of situation. Some material contexts of situation have the same tenor; others have different tenors. For example, the bird lying on the ground with pedestrians walking by is one context of situation. When Will comes on the scene and enters the bird?s material setting, the context of situation changes for the bird and for Will. When Will rescues the bird, while they are still in the pedestrian filled square, they remain together in the same material context; however, Will is now in a helping relationship with the bird. Prior to their encounter in that material context Will was engaged in a different social activity?that of going into town with his mother.  The challenge in dealing with multiple material settings and multiple contexts of situation in visual storytelling is in constructing seamless transitions from one environment to another, that is, to render a flow that has rhythm and cohesive relations that contribute to a whole. Graham achieves the flow through patterns and repetitions of colour, size, salience, framing, and the positioning of New and Given. Also in portraying the tenor, that is, the social activity of characters in each field, Graham achieves a good balance of focus or salient attention that shifts from the bird to Will, to the bird and Will, and to the family unit of caregivers. Narrative Activities and Elements The shifts in focus on narrative activities and elements within each narrative environment just described, creates the flow of visual storytelling. While the shifting focus establishes a rhythm, it simultaneously conveys key information about narrative activity as well as providing a model of caregiving that expresses a key theme of the story. 174 Frequent Iterations: Repetition as a Resource for Salience The bird is iterated 12 times in the book before Will finds it, whereas the little boy is iterated only 6 times in that sequence of illustrations. One of the reasons for the more frequent iterations of the bird is that five iterations are used to detail the bird hitting the glass building and falling and then five more iterations as it lies on the ground. The first set of iterations constructs the bird?s descent and the next set constructs a sense of the bird?s fallen state and its inability to move. Each is accomplished by different repetitions that construct contrast. To create the fall, the glass tower must be repeated with the bird in a different position for each portion of the descent. The bird?s injured state is constructed by rendering the bird in the same position while the surroundings change. As well, the second set of iterations constructs a stark contrast to the bird?s flying state. A high incidence of iterations of the bird constructs a continuity of movement and a dramatic rhythm to the bird?s fall. High salience is characteristic of the frequent iterations of the protagonist in narrative picturebooks for young children. While that salience is often constructed by saturated colour, there are other ways to make salience visible as in How to Heal a Broken Wing. Throughout most of the book, Will is the most salient visual presence because of his bright red jacket. In many of the framed panels, he and his mother are generally more salient than passing people in the city scene because they are rendered with more colour and higher contrast. So the question might be why use a bright colour on Will? The primary design reason is that Will is small and, as the protagonist, he needs more salience than his mother and the other people in the scene. In addition to the use of colour to build Will?s salience, repetition both develops Will?s character and his salience by the quantity and frequency of iterations. Frequent iterations construct prominence along with a high degree of continuity. The bird?s salience is developed by frequent iterations and by location. The bird is a pigeon, pale, grey, and lacking its own natural colour salience. In most depictions, Graham (2008) brightens the space around the bird when it appears. However, when the bird first appears, it is in subtle contrast to the grey sky. Then with the first page turn to the half title page, the pigeon is in greater contrast, clearly legible against a white sky. The next iteration does not 175  feature the light aura. Instead, the bird is framed by its location on the page, that is, close to the middle and in between window bars. This framing underscores the fact that the bird has hit the window because it has mistaken the reflection in the glass facade for actual clouds. The next page turn reveals the bird beginning to fall. Here it is tiny, but primarily in a white space, and with the next page turn the falling bird is iterated in three tall panels of white. At the foot of the fourth framed, full-page panel, the pigeon comes to rest in the center of the bottom of the frame and with white around it to highlight and distinguish it from the grey ground. In this iteration of the bird, three things give it salience: the light ground around it, its center position, and the fact that it lands at the foot of a diagonal trajectory vector created by the three iterations of the falling bird on the opposite page. In subsequent iterations, the pigeon is surrounded by either a white or a bright yellow light area, particularly at key moments such as when Will picks up the bird, when he releases it, and when the bird soars above the place where he released it.  There are several reasons for the variations in the forms of salience that render repetition in this text. First, there are physical or ideational reasons, such as the fact that a pigeon has light colouring necessitating the use of other forms of salience. In that case, highlighting the space around the pigeon serves an end goal as well as creating an instance-by-instance foregrounding. The overall goal is that at the end of the narrative the subtle highlighting of the bird?s flight to freedom requires preliminary iterations to establish the convention, although it is true that similar colour highlighting on the cover also establishes the use of that dynamic. Other variations in the form of salience may be related to the construction of textual rhythms and the resource of repetition. The intention to help young readers follow main characters by frequent salient iterations of them requires the repetition of their location in the composition, their size and scale, and their identifying features. One of the most vivid variations in the salient focus on narrative activities and elements and/or characters is in the scenes when Will and his family are caring for the bird. That episode in the story is told across 11 pages containing 48 scenes, which amounts to two-thirds of all the scenes in the picturebook. These 48 scenes express the text across them: ?A loose feather can?t be put back . . . but a broken wing can sometimes heal. With rest . . . and time . . . and a little hope . . . a bird may fly again? (Graham, 2008 unpaginated sequence).  176  The home healing scenes are bookended by two pages with five scenes each, all featuring Will in his salient red jacket. Each of the bookend double spreads illustrate Will?s journey from downtown to his home and then the return trip to the downtown plaza where Will and the bird first met. Will?s red jacket is the salient focus that helps the reader/viewer follow Will?s journey. In Will?s home, there is subtle shift in salient focus from Will as an individual to Will as part of a family unit, which highlights their collaborative role in healing the bird. The increase in the number of small scenes highlights the notion of rest and time. This increase in scenes involves much-repeated iterations of Will, the bird, and Will?s parents. Because most of the pages have few to no words, the illustrated scenes render almost all of the narrative activity so more repetition of action to action is necessary to construct the visual sequence of action. Repetition, Salience, New and Given This book provides good examples of solutions to the challenges of rendering New and Given in sequential visual narratives that integrate the grammatical need of giving New some salience with the need to keep a logical order to the flow and rendering of narrative activities and elements. The further challenge in rendering New and Given in a visual narrative for young children is to integrate that work while sufficiently iterating the characters and activities so that readers can follow them. In this book, as is typical of picturebooks with spare verbal texts, visual repetition of main narrative characters and events is a key cohesive dynamic. Furthermore, this book has a dual narrative focus resulting in both the bird and the boy being iterated over 50 times each across the 40 pages of text. It is also a book with 75 scenes, so it affords many opportunities to iterate characters and events and simultaneously construct patterns of colour deployments, salience, framing, and position on the pages in the service of rendering New and Given. The endpapers introduce the subtle salient New of a grey bird flying in grey clouds, followed on the next page by the New realized by the clear salience of a blue sky and clouds reflected in the glass of a skyscraper. In both these first two iterations of New, the New is positioned on the right but in the upper right quadrant of each page. The second iteration of the bird is highly visible against a white background, but it is clearly Given and on the left and across 177  the border of the upper and lower quadrants. The third iteration of the bird is on the double title page. What appears to be falling bird is placed on the left, in the Given position. What can be construed as New is the large scale of the building and its steel beams, which accentuate the bird?s descent. However, it is not completely clear that the bird is falling. The New and salient object in this double spread image is the building and its reflection of the blue sky with clouds.  Turning to the next double spread, the New is the entire urban setting. Upon close looking, the bird is just a tiny object that is noticeable because of its placement near one of the two towers with blue sky reflections, and because of its contrast in size and shape to the buildings. The bird is Given and a stark but tiny contrast to the cityscape, but the verbal text guides the viewer: ?High above the city, no one heard the soft thud of feathers against glass? (Graham, 2008). The verbal text announces the theme of the image as ?high above the city? (Graham, 2008) and then verbally realizes ?the soft thud of feathers? (Graham, 2008) as New. Therefore, while the bird is Given, it is the ?thud of feathers against glass? (Graham, 2008) that is a New event. Then, with the next page turn, three vertical panels illustrate the bird?s body as it descends. The panels document the fall and render a sense of the timing and visual shifts of the bird?s body as it descends. This spread is an example of how the words and images collaborate in structuring the information. Words can call attention to the visual theme and the New event. Without the verbal collaboration, the illustrator might have positioned the bird further to the right into what is often considered the New position. In wordless visual sequential narratives, it is fairly common to find the New on the right side of the image. This is an interesting phenomenon that seems to reflect the tendency for wordless visual sequences to become linear and to render bits of New information in a physically logical direction. The New is then often realized with salience in addition to its frequent location on the right side. Much of the spare text in How to Heal a Broken Wing (Graham, 2008 ) is placed on pages that have illustrations with few salient features. In these spreads, the words call attention to the New. In many of the wordless pages, salient colour is used for New. Graham uses a 178  variety of the systems that integrate text, such as salience, colour, framing, and position to realize New information and to pace the story. Repetition in Patterns of Framing Framing in How to Heal a Broken Wing (Graham, 2008) ranges from extreme close-ups with full bleeds and partial unframed bleeds to a proliferation of framed panels. Framing is often the means for constructing the interpersonal since viewing perspective and distance can produce and increase interpersonal interaction with an image. Framing alters both the expression of the interpersonal and the construction of levels of interactive participation. In this work, framing also contributes to the construction of the text?s rhythm. This rhythm is constructed by both the resources of repetition and framing and by the patterns of interpersonal interaction that those resources construct. Patterns of framing influence and interpersonally engage the viewer. First, framing and paneling do several things. They create additional scenes for the iterations of narrative activities and elements. Frames provide the means of distancing the viewer from the activities and elements by outlining and dividing the space between the text on the page and the reader/viewers. As Painter, Martin, and Unsworth (2011) noted, it can also make an illustration seem more like a picture. As well, as previously noted, framing contextualizes an illustration (Painter, Martin, & Unsworth, 2011). The absence of framing tends to bring the viewer into the context at hand. Nevertheless, that also depends upon the size of the image and its narrative and compositional context. How to Heal a Broken Wing (Graham, 2008 ) contains examples of different kinds of framing and non-framing that have clear influences over readers?/viewers? meaning making because the number and size of the frames and/or panels affects the reading rhythm and the narrative pace. For example, often, the pace of the narrative is slowed down by the use of large panels and full-bleed double spreads and also by the use of many small frames or panels. However, in the case of multiple small frames, the content of each illustration or panel determines how much time will be spent on it and whether its size and content make it a strong beat or a complex one.   179  This chapter reports on the close analysis of the cohesive work of colour and repetition in two picture books. It aims to demonstrate the power of colour to construct cohesion and to show the collaborative resources of repetition that add to this power, particularly in the hands of great picturebook artists such as Maurice Sendak and Bob Graham. In the next chapter, I discuss the significance of these findings and consider further investigations into the ways of meaning-making in the picturebook. 180  Chapter 7: Reflections on Looking for Ways of Meaning 7.1 Introduction I borrow the phrase ?ways of meaning? (Hasan, 1996) for my chapter title not only to recognize the influential role of Halliday (1975, 1978,1994) and Hassan (1984, 1985, 1996) on my research and because, in essence, this is the focus of my research?a search for how ways of meaning are constructed in the sequential narrative picturebook and, particularly, how the cohesive visual resources of colour and repetition contribute to information focus and flow. What began as a singular quest to explain multisemiosis in picturebooks turned into a study that foregrounds the picturebook format as a resource for multimodal literacy development in young learners. Undertaking this research necessitated the formulation of a methodology, which when applied to picturebook analysis resulted in two multimodal learning and assessment resources. The first resource is a framework of resources and the intra-textual environments they construct. The second resource is a defined set of cohesive colour configurations, associations, and identities that support assessing and editing visual cohesion in sequential multimodal forms. A further outcome of my study is a set of descriptions of the ways visual repetition constructs continuity and cohesion.  The use of SFL and the application of a systematic approach to meaning-making in picturebooks were instrumental to several of this study?s broader implications for educators such as ways of assessing visual cohesion, information flow, and information focus of multimodal work. Explanations of applications of SFL in these areas will be woven into the discussions of the findings and conclusions. I will also suggest the benefits of the use of the picturebook for multisemiotic analysis for learning about language in context. The literary picturebook is brief, concise, and an ideal format for the delicate analysis of multisemiosis. Moreover, the picturebook?s accessibility and sequential narrative structure make it suitable for educators who wish to respond to the growing imperative to involve current and future students in multimodal learning and literary response activities.  181  This research study is in the long-term service of understanding and articulating the visual and verbal meaning-making resources in picturebooks and other forms of visual sequential narrative. It adds to a corpus of systematically collected information on how resources construct cohesive multimodal texts. Adopting a systematic approach to the analysis of meaning-making in picturebooks by using my frameworks of the resource systems of the picturebook and the intra-textual environment and its textual environments confirmed a number of my assumptions about colour as a general and particular resource for cohesion. In addition, this study expanded on notions about the role of repetition in building coherence. Given the significance of these frameworks as outcomes of this study, I begin this chapter with a discussion of them and their potential for further study especially in the field of education.  7.2 The Frameworks of the Resource Systems of the Picturebook and the Intra-textual Environment and its Textual Environments  In a complex narrative picturebook, the great challenge in analyzing how its meaning is constructed stems from the fact that it is an orchestration of multimodal expressions. A recurring question usually concerns what aspect to investigate first. As discussed in Chapter Four, SFL offers the resource of the metafunctions to deal with the sequence of investigation. Metafunctions describe the three functions in a semiotic message: the representational information, the interpersonal, and the textual. Although the representational concerns what is expressed and the interpersonal explains who is involved in the communication, throughout this study the first line of investigation has been the textual because it addresses how the meaning is constructed.  Applying a theory of language to image analysis and multimodality is often criticized due to the linearity of language and the spatiality of image. However, in this case the use is applied to a sequential visual narrative form with a high degree of linearity, which makes it an appropriate fit. Additionally, the issues of how cohesion, continuity, and coherence are constructed visually in sequential narratives share much common ground with their construction in narrative language.  182  Addressing only the textual functions of the modes in the picturebook narrows the investigation. Nevertheless, it remains a complex endeavour because the picturebook is not only a multimodal text but also a physical artifact with a materiality that influences the construction of its text. The materiality of the picturebook is yet another factor that can be addressed by the adaptation of several of the theoretical resources used in Systemic Functional Linguistics (SFL). SFL and a number of significant applications of its key notions to multimodal analysis inspired my framework of resource systems because they model the systematic organization of resources (O?Toole, 1994; Kress & van Leeuwen, 2006; Matthiessen, 2007). Specifically, it was Halliday?s (1994) description of systems of resources as networks of potential that led me to conceptualize a set of picturebook resource systems.  7.2.1 Resource Systems The application of Halliday?s notions of strata, context, metafunctions, information flow, and focus to picturebook analysis, were especially influential in formulating the parameters of these resource systems. Halliday?s view of resource systems or networks of potential offered a means of isolating important features of form that build meaning. In the course of investigating cohesion, I discovered colour configurations. Also, inspired by Halliday, I conceived the resource systems and their environments as a way to systematically view features at work. Kress and van Leeuwen (2006) and O?Toole (1994) reinforced the viability and productivity of looking at meaning in the picturebook as constructed by a network of deployed resources from a set of systems. Their work along with SFL (Halliday, 1994) made it clear that the analysis of a complex network requires a theoretical framework that affords the systematic definition of each resource, its potential, and a robust analysis of each resource. Table 3 lists the six resource systems and Table 4 enumerates the six intra-textual environments created by the systems. While the two frameworks are closely related, several systems share connections with more than one environment. 183  Table 3 Resource Systems in a Picturebook Narrative the narrative, characters, settings, activities or events, narrative conventions Verbal the words and grammar of the language used Topographic the physical properties of size, height, width, paper, binding, paper textures, quantity of pages Typographic the font style, size, weight, colour, texture Visual colour, tone, line, shape, dot, texture Cohesive  this is a mixture of the integrating and cohesive resources of colour configurations, associations, identities, palettes, repetition sense relations, rhythm, salience, contrast, position, information  value, information focus, Theme and Rheme, New, and Given    Table 4 The Intra-textual Environment of the Text and Its Textual Environments Narrative  The material and semiotic contexts of situation of the story. Material situational contexts are the narrative settings of the story. Verbal The overall flow of language realized by the printed typographic text, by the phonology of the text when read aloud or when embedded in a text as in a recording or video. Visual  The appearance of the text?its colour, tone, line, style, texture, shape, and size. Can be divided  into the Typographic environment and the Topographic environment Topographic  The overall material properties of the book, its binding, paper, paper texture, weight, scent, size, and shape. Typographic  The overall typographic design of the book. Cohesive The collaborative cohesive work of the verbal and visual modes. Includes colour configurations, cohesive verbal ties and chains, and cross-modal cohesive chains and ties.  184  It would be possible to describe a single overall material environment of the text including all its physical attributes and all its visual features of images and printed text. However, for the purpose of analysis and from a practical text-producing perspective, I chose to describe the material environment from three perspectives, the topographic, the typographic, and the visual environment of images and graphics. For my analysis of cohesion, I combined the visual and verbal cohesive resource into what I call the multimodal cohesive resource environment as a way of looking at cohesive work in all its modalities. My decision to take this approach acknowledges the frequent occurrence in picturebooks of cross modal relations. Notably, narrative environments can be viewed as conceptual entities in the sense that the narrative is a semantic environment expressed by all resource systems. Narrative environments are made visible or realized by resource deployments from each of the other systems. However, not all narrative environments will be fully and visibly represented on the printed page since some narrative environments may be solely evoked by language, both aural and printed. Nonetheless, the narrative environments are critical to the picturebook text as they are the expressions of the narrative content, its elements, its environments, and its fabula. They express content (narrative structure, elements, and events) derived from the narrative system and they are realized by the deployment of resources from the other systems. From an SFL perspective, the six systems and the environments they build resemble a stratified model of communication in that the narrative system (a semantic stratum) is expressed in the narrative environment which consists of narrative settings (semiotic contexts). Those settings include semiotic contexts of situations and material situational settings all of which are realized by the expressive systems of the verbal and visual.  Another reason for SFL?s influence on the creation of this framework of systems and environments is that it is a functional theoretical model of a semiotic system. Its functional explanations of how a semiotic system works provide practical ways of supporting multimodal literacy learning and teaching. One of those ways is by defining systems of resources. Another practical benefit of a functional approach is that in addition to acknowledging sets of resources, each set can be described by how it functions in the context of a stratified model of communication.  185  Working with a stratified model of communication enables greater focus on the role of context in meaning making. Indeed, a benefit of using with SFL comes from the fact that it acknowledges that the relationships between contexts, meaning, and expression are essential to the assessment of multimodal teaching and learning. As such, this set of resources can be useful in describing picturebook creation to educators and students. Principally, this ability is due to the provision of a systematic means of assessing key text forming features. These sets also provide a model for considering the contexts of their creation and their viewing and the contexts within the expressed content since, in the case of a picturebook; the narrative might take place in a number of different contexts.  Working with this set of resource systems educators can choose to work with one system at a time depending upon the context of the classroom and the particular book at hand. For example, in a class with new language learners whether children or adults, a teacher using wordless books can draw on the visual system of resources to look at how its features construct a narrative. In fact, in this circumstance a teacher can use the inventory of visual resources to assess and select wordless books for classroom use by looking for general colour cohesion, and for the repetition of main characters and narrative elements. In a wordless book, the viewer looks for clues to meaning from particular visual elements, semiotic contexts, and material settings in the book. A good example of the latter is Shaun Tan?s The Arrival (2007), a wordless story of immigration in which the narrative context is fully visually realized as the material situational setting of a new world in which the viewer enters as a true immigrant. The viewer?s attention is guided by the visual work of salience, position, and framing to follow a path of key elements and events. Wordlessness  enacts the immigrant?s experience of not understanding the language of a new land and being forced to rely on visual signs to make meaning in a strange new world.  For illustrators, whether young classroom students or emerging professional practitioners, describing the construction of a picturebook in terms of making choices from six sets of resource gives them a set of conceptual tool kits for visual communication. The systems also help in the orderly editing of rough drafts of sequential visual narratives such as picturebooks and comics and other multimodal classroom work that involves the visual mode. Here the use of the systems to edit student work is a sorel