Semiosis and the Picture-Book:

On Method and the Cross-medial Relation
of Lexical and Visual Narrative Texts

Peter Pericles Trifonas

Ontario Institute for Studies in Education / University of Toronto

Semiotic Dimensions of Textual Form and the picture-book

The term "text" has evoked various meanings according to particular disciplinary perspectives (Trifonas, 1993). In cognitive psychology, it has been represented as the sum of authorial propositions stimulated by the linguistic variability of forms; in semiotics, as the set of lexical or visual signs that act as "cues" to guide the reader's inherent predilection for mental decoding operations. Structuralist theorists after Ferdinand de Saussure (1916) determined the text to be "an object endowed with precise properties, that must be analytically isolated" and by which the "work can be entirely defined on the grounds of such properties" (Levi-Strauss cited in Eco, 1979, p. 3). Some proponents of poststructural theorizing have conceived of "textuality" as the substantive equivalent of the author's productivity, a polyphony of (a)synchronic voices realized via intertextual processes of communication for the social exchange of thought surmising the tensions of knowledge, power, and desire (Kristeva, 1969; Barthes, 1970). Others (Eco, 1976; 1979; Peirce, 1931; Derrida, 1974) have cultivated a non-metaphysical conception of text/uality where meaning-making on the part of the reader is considered to be a generative movement embodying a semantic glide or elision of infinite, yet playful, regressions that negate objectivity and render the content of the written word undecidible in relation to a seemingly uncontrollable labyrinth tracing a network of possible interpretation. The picture-book genre offers an interesting case for illuminating the formal dimensions of textual structure and for exploring the meaning-expressive potential of the lexical and visual forms of signification characteristic of such cross-medial narrative texts (Kiefer, 1988).

AS/SA nº 11-12, p.67

Even though the picture-book has displayed the propensity to be a highly unconventional or experimental literary genre employing both lexical and visual systems of signification (Kiefer, 1988; Lewis, 1990), the dominant paradigm for educational research of the picture-book reflects three main "types" of analytical strains: 1) pedagogic, where the printed word supersedes the pictorial aspects of the text as the focus of examination in the meaning-making process; 2) aesthetic, where the rationale for research is drawn from art criticism and/or art history as an appreciation of form leaning toward the pictorial aspects of the text at the expense of its lexical co-text; and 3) literary, where the picture-book is subsumed in the vast oeuvre of children's literature "as a marginal genre, or a larval stage of literature proper" (Lewis, 1990, p. 140). David Lewis (1990) has identified the metafictive, postmodernist, or "non-mainstream" features of the picture-book that belie the recalcitrant minimalism of any staid or stagnant notions about the inventiveness of its authors and illustrators and the extent of boundary-breaking occuring within the genre at large. The picture-book is essentially an open form, a fluid textual entity, incorporating lexical as well as visual signs variably codified in an unceasing interaction of word, image, and reader (see Lewis, 1990; Kiefer, 1988). Lewis (1990) notes that: "An adequate theory of the picture-book must directly address the bifurcated nature of the form (word and pictures) and must account for the whole range of types and kinds including the metafictive" (p. 141). Moreover, because the picture-book genre, by definition, is dependent upon the interaction of two separate yet integrated systems of signification (lexical and visual), it is a unique combination of literary and pictorial elements exhibiting "high semantic or semiotic capacity" (Landes, 1987, p. 320). And the complementarity of textual media is what animates the natural predisposition for the creation of personal cognitive, affective, and aesthetic meaning in the reader-viewer. It is this obvious semantic-semiotic endowment of the picture-book — the formal properties of its construction and structure making it ideal for the purpose of teaching young children by establishing "contexts for literary and real world understandings" (Kiefer, 1988, p. 260) leading to learning experiences — that merits the focus of educational inquiry (Schwarcz, 1982). In semiotically reconstituting the picture-book (Lewis, 1990), it is necessary to step back from the well-worn research paradigms discussed above. And by so doing, to take the liberating scope of another vantage point, one more likely to fuse together the sometime disparate pedagogical, literary, or aesthetic interventions exemplifying the study of the genre thus far, by taking into account this codic interdependence, the cross-mediality both within the narrative unfolding of the lexical and visual textual milliuex of the picture-book and also between the picture book and the reader-viewer (see Kiefer, 1988). How does the textual form of the picture-book work, both lexically and visually, as a semiotic system of signs and codes to create the interpretative possibilities for meaning-making?

Cross-mediality and the picture-book

The textual formulation of the picture-book, as in any literary or visual artistic endeavour, functions to enhance the hermeneutic potentiality of the general genre across an ever-expanding field of semiosis (Kiefer, 1988; Lewis, 1990; Landes, 1987; see also Eco, 1979; Greimas, 1983; Arnheim, 1974; Lotman, 1990). In order to understand the semiotic intraspecificity of how meaning-making is produced through the cross-medial dynamics of picture-book construction, it is essential to identify the formal elements of its lexical and visual textual components, hence, establishing a structural foundation for an exegesis of their interactionwithin the presence of a reading-viewing consciousness capable of actualizing the text's latent expressive potential (Trifonas, in press; see also Trifonas, 1992; 1993). A semiotic explanation of the relationship between the reader-viewer and the text is concomitant to isolating those aspects of textual form that serve as a mediative vehicle for the facilitating of lexical and visual "markers" upon which to furnish cognitive, affective, or aesthetic hypotheses from the conscious and subconscious responses required or initiated during the reading-viewing process as a meaning-making activity.

AS/SA nº 11-12, p.68

Toward this purpose, I will outline a method of textual analysis designed especially for the cross-medial make-up of the picture-book, one incorporating somewhat "traditional" semiotic techniques (e.g., Eco, 1979; Greimas, 1983; Barthes, 1964, Saint-Martin, 1990) that have been utilized for the examination of lexical and visual texts in order to address the previously posed question. The various "boxes" composing Table 1 denote the procedural landscape of a semiotic matrix for clarifying the complexity of such textual investigation. The cells of the matrix itself are used to identify the "levels" of interaction between the picture-book and the reader as well as to isolate the narrative structurality of lexical and visual texts within the cross-medial codic domains of the picture-book itself (see Greimas, 1987; Barthes, 1970). With specific reference to the "structural semantics" of approaches for probing the constructedness of lexical and visual texts as identified in semiotic methods of narratological inquiry, the emphasis of the investigation is twofold: 1) upon the examination of the syntactic composition of the picture-book as a synthetic narrative of lexically and visually encoded text(s); and 2) upon the mental operations (cognitive and affective, conscious and subconscious) required by, or initiated in, the reader-viewer by the text as a codic set of lexical and visual signs leading to the creation of cognitive and aesthetic response(s).

A recent study of the picture-book, as "event," focused upon the semiotic dimensions of the transaction between the reader, or "performer," of a text and the listener, or "spectator," of the performance exhibited during a class reading (Golden & Gerber, 1990). The researchers were primarily concerned with studying the effects of paralinguistic cues (performative and instructional) initiated by the performer upon the subjects' interpretations of the text as an intersubjective "social event." Semiotic methodology was not used to investigate the evocative dimensions of the picture-book genre nor to identify or to explain how the interaction of lexical and visual signs and codes in the textual form of the picture book work to yield effects motivating the invention of meaning-making for the reader-viewer. This study therefore did not address the expressed need for a semiotic analysis of the genre at large that would account for the meaning-generating potential of an overall text comprised of lexical and visual systems of signification that characterizes the cross-medial nature of the picture-book form as such (see Lewis, 1990; Kiefer, 1988). To this end, semiotics offers a highly developed epistemological, theoretical, and methodological framework for de-constructing the structurality of lexical and visual signs embodied in picture-books as communicative sign systems or codes that work to facilitate the conveyance of meaning. And it affords an opportunity to interrogate the bifurcate structure of the picture-book text as a medium for the exchanging or dissemination of knowledge. This is an essential area of research if we hope to understand the role of such multi-media texts in the learning-teaching process. Semiotics inevitably allows the researcher: 1) to take into account levels above and below the sign (Greimas, 1983); 2) to examine the means of signification as well as to project the content of signification (see Hjelmslev, 1943); 3) to ground the analysis in the text itself and to examine how the structures of signification are engendered "globally" in codic terms to form sophisticated cross-medial systems of signification (Eco, 1979); and 4) to examine the roles of both the sender (e.g., a text) and of the receiver (e.g., a reader/viewer) in a pragmatic act of communication through linguistic and non-linguistic modes of representation (Eco, 1976; 1979).

AS/SA nº 11-12, p.69

Definitions and Features of a Semiotic Methodology: Some Assumptions

The epistemological, theoretical, and methodological principles of semiotics (see Eco, 1979; Greimas, 1983) incorporated within the matrix provide the basic tools and metalanguage for the semiotic study of cross-medial textuality articulated in a narrative form. These presuppositions are useful only to the extent that they allow for the particular phenomena under study to be circumscribed in terms comprehensible to the human intellect (Eco, 1979). It is in this sense that a "methodological structuralism" as the operational program for scrutinizing lexical and visual texts is crucial to the model I have developed after the method of Umberto Eco (1979) because without the necessity of this reflexive metalanguage, there would be no way to achieve the purpose(s) of semiological inquiry relevant to the examination of the picture-book form as outlined above (see also Eco, 1976; Nöth, 1990). A semiotic method of textual analysis is accordingly considered to encompass metatextual means or devices (e.g., a metalanguage, a "model," figures or other visual schemata, etc.) that conceptualize in hypothetical, rather than empirical, terms the intensions and extensions made by the reader-viewer in the act of meaning-making relative to the lexical and visual structures of signification occupying the picture-book form. Intensional responses are defined as the consciously motivated acts of meaning-making demanded of, or brought forward in, the reader-viewer to realize the signifying contingency of the "total text." Extensional acts are not consciously motivated, but are determined extratextually in direct relation to the signifying structures that in-form the articulation of the text. Their affectivity is surmised by contextual factors that influence lexical and visual sign perception in the pragmatic act of communication (e.g., culture, education, competence, etc.). A model reader-viewer (Eco, 1979; see also Iser, 1978; Trifonas, 1993), one who can fully apprehend the intensional and extensional "requirements" of the picture-book form, is postulated here as an integral feature of the semiotic method for the analytical aims of this essay.

The model of textual analysis for lexical narrative and the role of the reader as proposed by Eco (1979) (adapted from Petöfi, 1973) forms the edifice upon which the concrete framework for the lexical component of the method is laid out. Eco's (1979) method, however, is modified and expanded (see Table 1) to include the features and function of visual text manifest linearly in the picture-book as a narrative progression. Like Eco's (1979) model, the position of the individual "boxes" that concretize the visual representation of the method in Table 1 does not preclude to any suggestions of an hierarchy of levels encompassing the ends of reading-viewing itself or to a sequential ordering of the reader-viewer's intensional and extensional responses to the textual form. The levels of possible abstraction at which meaning-making occurs are addressed metalinguistically. Eco (1979) explains this misleading premise of semiotic practice:

The notion of textual level is a very embarrassing one. Such as it appears, in its linear manifestation, a text has no levels at all... "level" and "generation" are two metaphors: the author is not "speaking," he has "spoken." What we are faced with is a textual surface, or the expression plane of the text. It is not proved that the way we adopt to actualize this expression as content mirrors (upside down) that adopted by the author to produce the final result. Therefore, the notion of textual level is merely theoretical; it belongs to semiotic metalanguage. (p. 13)

Table 1 is likewise intended not as a guide to hierarchical planes of lexical and visual text or to a "step series" of acts or responses that the reader-viewer may execute in relation to the signifying structures of the picture-book form, but to reveal and to reinforce the interdependence among the metatextual "boxes" in detailing a semiotic method for textual explorations (see Eco, 1979). The only way in which the method depicted in Table 1 provides a concrete case for textual interpretation is the fact that all intensional and extensional performed by the reader-viewer are apprehended in dependence with the linear lexical-imagistic manifestation of the picture-book such as it appears linguistically and visually in existing lexematic and colorematic surface forms (Eco, 1976; 1979; Saint-Martin, 1990). Furthermore, in Table 1 a horizontal line separates the "actualized content" from the given set of codes or subcodes the reader-viewer applies to these expressions of textual form as discursive structures or plastic and perceptual variables to transform them into meaningful content.

AS/SA nº 11-12, p.70

The Reader: Lexical Codes and Subcodes

Basic Lexical Dictionary. The reader utilizes the graphic and lexical signs provided by the text to construct kernals of semantic sense from semes, or minimal content units, embodied in the expression(s). This constitutes a primary tentative effort toward an amalgamation (in a general sense) from which meaning is generated as the productive effect(s) of consciousness or memory over time (Greimas, 1983; Eco, 1983). For example, the sentence /Effie is a gregarious ant/ contains composite syntactic and semantic indicators within the elements of the expression that function to elicit cognitive and affective responses from the reader. The noun /Effie/ is a deictic referent to a human name, that of a girl or perhaps a woman, in itself, promoting mental associations representative of the conceptual properties of the word manifest anthropomorphically as the referential equivalent of a human state of being (e.g., a woman as experienced, in reality, having human proportions, characteristics, demeanours, behaviours, etc.). The development of meaning-making is dependent upon the projected existence of a basic lexical dictionary in the conscious mind of the reader that can be drawn upon to reference associations stimulated by distinctive graphical clues in the form of word arrangements. The syntactical properties (e.g., singular, feminine, noun, etc.) of the lexemes, or words, do not completely actualize the meaning potential of a total expression until connections between other terms in the total expression are established by degrees of difference through co-referencing. Thus, the isolation and actualization of the virtual semantic properties latent within lexemes is contingent upon the syntactic structuring of expression that facilitates the reader's ability to realize semantic disclosures.

Rules of Co-reference. The various shifters in the text work to orientate the reader on the grounds of the first semantic responses to the words (Greimas, 1988). /Effie/ as a sememic unit of some sense is undercut with reference to non-human associations, since the noun qualifier /ant/ is anaphoric in re-expressing and re-establishing a previously made semantic relationship that the reader must now adjust to contextualize the meaning of the sentence. Initial reader expectations are also displaced with the awareness that /Effie/ refers to a non-human entity and the sememic level of meaning becomes redefined textually through other syntactic relations between lexemes (e.g., /Effie/ "human" ; /Effie/= "ant" ). Co-references are textually based and disambiguate meaning from surface to "deep levels" within the discursive structure of the sentence (see Greimas, 1979). If this is not possible, the reader relies upon further (co-)textual clues for clarification.

Contextual and Circumstantial Selection. Beyond the co-textual manifestations of meaning in linear text (e.g., word forms), contextualized selections provide indicators to the referentiality of a given term in comparison with other terms originating from the semiosphere of the same signifying system, such as in a language (Eco, 1979; Lotman, 1990). In this case, reference is established in an encyclopaedic knowledge framework where one lexeme can denotatively and connotatively generate a series of associations with which the reader may or may not be familiar as determined contextually through experience. For example, a lexeme like /hen/ can refer to "bird" or "poultry" in different cultural contexts and could point to radically diverse associations for individual readers depending upon the experience(s) of the lexical sign substituted for the concept as real world object. The distinction is produced by the reader when possible interpretations of a word within an expression are selected and rejected according to textually suggested correlations of lexical signs with external referents that are used to build mental interpretants, or imagistic representations of concepts.

AS/SA nº 11-12, p.71

Circumstantial selection is based upon "bookish," or intertextual, competence, and the ability of the reader to reconcile the presence of elements external or foreign to the semiotic code to which a particular text adheres. This might include aspects of vocabulary and specialized expressions or jargon. In narrative texts, circumstantial selections become contextualized, or are linguistically defined, in order to avoid confusion (Eco, 1976; 1979; see Iser, 1978).

Rhetorical and Stylistic Overcoding. Rhetorical and/or stylistic cues alert the reader as to whether language is being used literally or according to aesthetic convention. For example, the phrase /Once upon a time/ is an overcoded expression in that the reader is in possession of, and inserts into, the semiotic exchange the part of the code required to complete the purpose of communication (Eco, 1979). The reader is aware of and alerted to the fact that a story beginning with this overcoded expression is fictional and written according to a certain style commensurate within the genre expected. The interpretation of the textual indicator is not naïve but purposeful in setting up and meeting structural or thematic expectations. To this end, rhetorical and stylistic overcoding can be used as a literary device in aesthetic texts.

Inferences by Common Frames. Frames are data-structures that are used in lexical texts to represent stereotypical situations experienced in reality (Winston, 1977). There are specific elements within frames (e.g., courses of events, individuals, objects, actions, relations, and facts) outlining certain modes of cognitive action (perception and language comprehension) that are necessary to understand the situation purely as an experience of familiarity (Eco, 1976; 1979). Beyond the visual aspects of a text, to be discussed later, a narrative contains references to visual objects, the features of which are isolated and identified in order to engender an overall common frame. For example, the lexical description of a farm might contain references to objects with overtly visual dimensions (e.g., hen, pond, haystack, mill, etc.) in order to establish a particular conceptual context for the scene depicted that is recognizable to the reader as a common frame. The listing of these objects is in itself an overcoding of the information required to understand the specific situation and the subsequent building of further data structures to complement the common frame of reference.

Inferences by Intertextual Frames. No text is read independently of the reader's experience of other texts (Kristeva, 1969). This is another case of overcoding where the extratextual experiences of the reader act as an encyclopaedic source for information that can be used to disambiguate the complexity of a text. What Eco (1979) describes as literary topoi, or narrative schemes of understanding rooted in intertextual frames of reference, may aid the reader to the extent that a text is immediately invested with metaphorical properties that are the discursive outcomes of intertextual reference (see also Iser, 1978; Kristeva, 1969) (e.g., allusions to stock elements of literary experience such as "the villain," "the Cinderella tale," "the happy ending," etc.) (Frye, 1957).

Ideological Overcoding. Ideological structures are outlined discursively within a lexical text through the progression of narrative sequences of action. The extent to which the reader can apprehend textual ideological structures is determined by an individulaized capacity for actualizing ideological subcodes, or gestalt (Iser, 1978). If a text is open, it allows for interpretation against a different code and is personalized in being uniquely invested with subjectivized meaning (Eco,1976; 1979; 1984). In the case of a closed text, however, a given ideological background can help to uncover or to inhibit the operation of the text on the level of fabula. For example, ideological bias can work to switch codes and lead the reader to interpret the code manifest in the text aberrantly, or other than that "intended" by the writer. That is not to say that the reader can know precisely what aspects of the author's ideological subcode is incorporated in the text; nevertheless, tentative ideological subcodes can be attributed to the writer when authorial judgements are isolated, usually in the form of overtly philosophical statements of justification, approbation, or criticism (Eco, 1979; Barthes, 1964) (e.g., in some instances, texts ask for ideological sympathy from the reader). Of the genre of fable, one may ask: What are the affects upon the reader of a story of an ant who — once exiled from a society of ants because of a naturally inherited physical trait — returns to heroically save the day because of that particular trait and now commands respect and love from those who once despised her? If the reader perceives the ant as being vindicated in the outcome of the action, then there is a sympathy between the embodied textual points of view and implicit authorial ideology and that of the reader. If not, then the ideological code of the reader succeeds in promoting an aberrant decoding of the text because of subjectivity.

AS/SA nº 11-12, p.72

The Reader as Viewer: Visual Codes and Subcodes

Basic Visual Dictionary. With reference to the recognition of the structural properties of visual representations, if the form depicted in a visual text is distinctively analogous, or representational, the viewer is able to juxtapose the figuramatic properties present of the form against the basic properties of natural forms as experienced in reality according to external visual paradigms. Forms are iconic to the extent that the actual properties possessed by corresponding real world referents, are reflected in and not possessed by the represented figures (see Morris, 1946; Krampen, 1973). The conventionality of the imitative code of the visual text is brought to bear upon the expressive plane. But the content plane, the meaning sense, may also be affected if the analogous image comes to arbitrarily represent something outside of itself. So, the visual text becomes symbolic, or contains digital imagery, the expression and content of which, are determined according to internal visual paradigms of a particular work or intervisual paradigms drawn from the viewer's other encyclopaedic sources (e.g., the fox as a symbol of "cunning" , the color red representing "danger", etc.) (Trifonas, in press).

Rules of Visual Co-reference. After the initial figuramatic analysis resulting in the detection of visual syntactic properties, the viewer disambiguates spatial, or toposensitive, relations among the forms in a pictorial plane. The first tentative attempts at visual co-reference are confirmed by a more detailed scanning of the forms as co-textual items within the pictorial plane and subsequent judgements are noted mentally. In this way, the visual text doubly articulates meaning on both the expressive and content planes (Eco, 1976; 1979; Sonesson, 1989; Trifonas, 1992).

(Visual) Contextual Selections and Circumstantial Selections. These are coded and displayed through the figure of form. To be considered iconic, the figures represented in a text must exhibit properties that are distinguishing characteristics of particular types of form as determined by external visual paradigms . For example, an animal depicted visually can be distinguished by its physical characteristics. If the illustrator wishes to distinguish between two or more types or species of the same animal, then the properties endowed the animal in the illustration will be precise enough for the viewer to cognitively facilitate the distinction. It is likewise the "critical responsibility" of the viewer to eliminate the possibilities of alternative selection while drawing from an encyclopaedic source of knowledge.

Conversely, if the figures contained in the text are foreign to the viewer's experience, then the viewer must resort to some external point of reference for guideance and clarification. Quite often, the text contextualizes explanations of items foreign to the viewer, codifying them, in order to expediate the meaning-making process.

Visual Stylistic Overcoding. The cumulative elements that comprise the visual text are stylistic features coded within the work itself (e.g., the depiction of figures, choice of setting, perspective, color choice, variation in textures, etc.) and cannot be extricated from the particular context of expression. These stylistic features act as overcoded cues in the visual text when the viewer is alerted as to whether it is being used to meet structural or thematic expectations according to the purpose of communication. For example, an abstract treatment of form is itself a stylistic feature of visual text that sets up a series of associations, expectations, and judgements for the viewer with respect to the means of accepting, decoding, and interpreting the images presented.

Inferences by Common Visual Frames. Utilizing the definition posited earlier (see "Inferences by Common Textual Frames"), it is necessary to stress that common visual frames are not necessarily inchoate texts (see Eco, 1979; 1984; Saint-Martin; 1990). The features that create the overall common frame are identified and isolated visually to produce overcoding. For example, a farm scene could depict some of the major elements that are traditionally associated with rural agricultural life: particular animal types (e.g., hen, fox, goat, etc.); naturalistic settings (e.g., trees, crops to be harvested, grassland, etc.); farm architecture (e.g., barn, hen-house, farm-house, windmill, etc.); and agricultural artifacts (e.g., tractor, cart, etc.). This can be described as a common visual frame because of the stereotypical nature of the scene contents.

AS/SA nº 11-12, p.73

Inferences by Intervisual Frames. It has already been stated that no text is read independently of the reader's experience of other texts. Where the external visual experiences of the viewer are elicited to act upon the interpretation of a visual text, visual topoi, or visual schemes of understanding, may aid the process of meaning-making to the extent that the ocular plane is immediately invested or overcoded with properties that are the products of intervisual frames of reference. The viewer must supply the necessary intervisual knowledge to make meaning from the visual text (e.g., stylization of forms according to convention, symbolic shapes, other culturally relevant information, etc.) (Trifonas, in press).

Visual Ideological Overcoding. In a visual text, the interaction between the forms depicted, both open and closed (see Arnheim, 1974), produces visual contexts consisting of formally structured pictorial elements that function on the thematic level to develop a distinct visual code objectifiable, to some extent, through recourse to language. The ideological interpretation of a visual text is dependent to a great degree upon the viewer's powers of visual perception because internal variables (e.g., the ability to perceive color, depth, topological disjunctions, etc.) may influence the interpretive outcome regardless of the openness or closedness of the text itself.

Actualized Content: Lexical Intensions and Extensions

Discursive Structures

The responses to the lexical level of a text must be actualized by the reader to allow for the progression to further amalgamations from differences between word-signs. Meaning is created through semantic disclosures made by the reader relative to discursive structures that isolate the immanent semantic properties of the lexemes virtually present in the reader's store of culturally grounded information (Eco, 1979; Greimas, 1988). Therefore, the word-signs in a lexical text actualize no meaning without the reader (Trifonas, 1993). The topic, or theme, of the lexical text functions as a guiding force to insure communication and to delimit the extent of possible semantic properties within the lexemes to be actualized by making them textually relative. The isotopies, or actual textual verifications of the topic, present in an expression also direct the meaning making process by providing a single level of sense from which the reader guides semantic amalgamations (Greimas, 1983).

(Bracketed) Extensions

Once the discursive structures of the text are actualized, the reader is certain of the characters, the actions, and the events that comprise the plot, since, the intensional semantic disclosures performed by the reader are realized through the interplay of lexical structures in relation to and within a total narrative sequence. Suspension of disbelief is then facilitated by the first overt recognition of a possible world with an inherent underlying logic corresponding to that of the characters, the actions and the events in the plot (Hodge, 1990; Eco, 1979).

Narrative Structures

Whereas the plot is the fundamental action of the text, the basic elements from which the story is generated is the fabula, the make up of the characters, the inherent logic of the action(s) and the time-line sequence of events (Greimas, 1987; Eco, 1979). Realization of the fabula involves a continuous series of abductions, or inferences, experienced linearly by the reader in the process of disambiguating a narrative text (see Peirce, 1931; Eco, 1984). Ultimately, the reading process leads from micropropositions emanating from those expectations initiated through semantic disclosures on the level of discursive structures to more definitive macropropositions such as themes, motifs, narrative functions, and the determination of various levels of abstraction regarding the fabula upon which the story itself generates meaning for any given action in the text (Eco, 1979; Greimas, 1983; 1988).

Forecasts and Inferential Walks

Since the fabula is always experienced as a linear and sequential set of abductions, a disjointing effect is necessarily experienced by the reader, thereby marring the vicarious imaginative experience of the text to some extent. An extension of the imagination to presuppose further action results in aporia, or concerned curiosity presupposing some resistance to closure, at the major or relevant disjunctions of the fabula that are set at the level of plot. Here, the reader infers by gathering intertextual support for the hypotheses created through the discursive structures of the text. In this way, the expected and/or the unexpected is made explicit "as individuals and properties belonging to different possible worlds imagined by the reader as possible outcomes of the fabula" (Eco, 1979, p.218)

AS/SA nº 11-12, p.74

Actantial Structures

The lexical text, as narrative, works to verify reader forecasts with respect to the fabula and its development (Eco, 1978; Greimas, 1983). Narrative is segmented into programs, or stories, where the fabula and every other narrative structure can be further abstracted and reduced to the formal positions of actantial roles (e.g., subject vs. object, sender vs. receiver) according to the modal predication of lexical actants, those acting and those acted upon, that function thematically on the level of discourse to produce textual actors, or characters (Greimas, 1970). The active interaction between the lexical actants within expressions on the level of discursive structures creates thematic meaning as the fabula is unfolded through the interplay of actors in the narrative structures. Lexical actants take the roles of actors when the thematic functions of a text are reinforced as discursive and narrative structures, thus, reliably pointing to sources of meaning-making within a text.

Elementary Ideological Structures

In comparing and contrasting actantial and actorial structures manifest in the lexical text, so as to distinguish dimension of "textual truth," there is an acknowledgement of the verisimilitude of the fabula on the part of the reader. This implies a comparison of the textual world with the reader's own real-world vision and a suppression of further suspension of disbelief (Eco, 1979). Elementary ideological structure oppositions can be translated into truth assignments where the reader, utilizing already formulated schemata, makes ideologically motivated interpretive decisions about the ideology expressed in a given text.

Textual World Structures

Once "textual truth" has been accepted, the text is reduced to oppositions and there is a subsequent assignment of truth values between the textual world structures determined. The given relations between the lexemes at the actantial level are considered insofar as they are predicated in the textual world structures as true or false (Eco, 1976; 1979). Ultimately, the reader makes final decisions about the credibility of the text as a series of reported events, the sincerity in embodiment of ideological beliefs through convincing characters, and the accessibility of the textual world as a fictional experience.

Visual Intensions and Extensions

Plastic and Perceptual Variables

Just as the lexical text is constituted of the sum of individual features that work to create meaning as a whole, the visual text is comprised of readily identifiable elements meaningful within an integrated form of expression. Consequently, the relationships between the manifest properties of coloremes, or minimal color units comprising a visual text, disclosed at a point of ocular centration during the act of viewing, may also be analyzed syntactically and semantically (Saint-Martin, 1990; Arnheim, 1974). The cumulative effect of two sets of visual variables, plastic and perceptual, upon the apperception process, isolates the latent properties of the coloremes virtually present in the viewer's store of culturally determined visual encyclopaedic knowledge (Gombrich, 1960; Saint-Martin, 1990). Exploring the prevailing chromatic relations between coloremes in a particular pictorial text, creates an awareness of how the visual variables determined through the formal structure of the work interact with respect to the perceptual processes of the viewer and engender meaningful visual experiences. Color, value, and texture are plastic variables while line, shape, form, vectoriality (focal point and directional tension), and implantation (position/balance) are perceptual variables (Saint-Martin, 1990; Trifonas, in press).

Visual Anaphoric/Deictic Extensions

Anaphora, for language, is characterized as a network of a relations between two or more terms, on a syntagmatic axis, establishing linkages in discourse (Greimas, 1983). On the level of visual text, anaphora can be regarded as the unity and coherence between the elements which comprise the work that must be maintained to create "pictorial sense." The recognition of form, from schema as objects, in a visual text is deictic because it is dependent upon the recognition of changes in the intensification or regrouping of coloremes aggregately within a visual field. Distinct contours between figures (open or closed) creates analogous forms isomorphic with the reality of experience and results in a stable and organized visual field of perceptual construction; whereas, digital, or symbolic, forms rival viewer interpretation because distinct form contours may or may not be present within the figures represented. The spatialization, or placement of forms, within the fore, middle, or background of a pictorial plane is a crucial determinate of the viewer's interpretation of a visual text resulting from variables in perception(s) according to individual gestalten approximations derived from experience (see Arnheim, 1974; Saint-Martin, 1990). Ultimately, the viewer can discern visual forms in a definitive spatiality of relations and the setting of which they are a part, consequently, setting up the imaginative conditions of a possible visual world that invites the potential for the suspension of disbelief.

AS/SA nº 11-12, p.75

Visual Metaphorical Structures

It is primarily through closed forms that regions or subregions in a pictorial plane lend themselves to iconization and are interpreted in relation to the properties manifest in relative natural forms external to the world of the visual text (Mitchell, 1986; Saint-Martin, 1990). It is on the level of visual metaphorical structures that a lexicalized equivalent can also be connected to the representation of form, thus, allowing for the linguistic differentiation of the pictorial elements of the text that adhere to vraisemblance, or display a direct correspondence with the apprehensibility of real world entities. In "global" terms, the extent to which the visual text reinforces the lexical text can be described as cross-medial agreement. If there is a immediate correlation between the visual and lexical possible worlds projected, then an objective correlative, or concrete visual representation, of the possible world referred to on a total textual level is established and elaborated upon through linear visual narrative . If not, then there is a chiasmos, or separation, between alternative world visions posited, visual and lexical, that the reader must juxtapose as fabulaic alternatives. The products of this type of visual stylistic overcoding are literal and figurative visual frames that may or may not reinforce reader abductions irrespective of stylistic considerations. On this level, the visual text works to secure thematic considerations as well as the functions of visual metaphorical structures from which abstractions in the from of macropropositions of the visual fabula (e.g., themes, pictorial motifs, etc.) are abduced by the viewer.

Visual Indexes

Visual indexes are the result of generative or repressive cross-medial image indicators built into the conventions of the text as a supportive visual framework for the inferences drawn from the lexical text. Beyond replication of possible lexical world constructs, the visual indexes set up cross-medial frames of reference with respect to internal and external paradigms applicable to a particular text that suppress disjunction and support thematic concerns on the level of the "global" fabula by providing points for comparison/contrast and clarification/elaboration upon the narrative structures of the lexical text through the linear visual narrative of the pictorial text. Therefore, the visual indexes serve to limit and define the viewer/reader's extensional responses in accord with the aesthetic conventions of the text by aligning the visual contexts appropriately to insure indexicality for the interpretation of signs and codes, lexical and visual, within a specific schematic and textual framework.

Visual Actantial Structures

Through the isolation of visual actantial structures, the viewer attempts to furnish hypotheses necessary for an analytical approach to the pictorial text as part of a sequential linear visual narrative. The viewer's approach to decoding, however, is non-linear but correlational in that the interaction of forms within the pictorial setting results in an awareness of the visual actants comprising a supersyntagm, or combination of elements co-present in the visual text, as they function to elicit thematic meaning (Saint-Martin, 1990) over an extended series of visual frames that constitute the visual fabula. The active or passive interaction of forms creates visual actantial roles (e.g., subject vs. object, sender vs. receiver) within the picture plane and as the visual plot is unfolded pictorially through the interplay of visual actants with distinctive thematic functions in the action and events of the linear visual narrative, the viewer is able to discern the visual actors .

Visual Ideological Structures

In essence, "textual truth" is determined pictorially when the visual text is acknowledged as "real" and the subsequent assignment of truth values placed upon a particular form or relation(s) between forms, as visual actors depicted in a linear visual narrative, is correlated with the truth values disseminated by the same relations between relative actors in the narrative structures of its lexical compliment. The reduction of the visual text to propositions of opposition determines if there is an incongruency that must be resigned before the lexical and visual texts are aligned on the level of fabula to consolidate the total ideological vision of the text.

AS/SA nº 11-12, p.76

(Visual) Veridiction

Through (visual) veridiction, there is an attempt at corroboration of assigned truths, both lexical and visual, within a single textual world structure. The extent to which the "textual truth" assignments of the lexical text and visual text are aligned thematically on the level of fabula, determines the aesthetic success of the work as a whole and the viability of the vision embodied within the cumulative effects of the synthesis of its formal features..


This essay has presented an epistemological, theoretical, and methodological framework (see Eco, 1976; 1979; Greimas, 1983; Saint-Martin, 1990; Arnheim, 1974; Barthes, 1964) for the semiotic analysis of the picture-book by explaining the cross-medial protocols of reading-viewing concerning narrative lexical and visual texts that are relevant for the aforementioned purpose(s) of the study. It details a semiotic foundation for the re-reading of the picture-book form to identify, to define, and to explain the various "levels" of interactivity between the lexical and visual elements comprising the signs or codes that engender textual structure in relation to the cognitive, affective, and aesthetic responses required of, or initiated in, the reader-viewer in intensional and extensional acts of meaning-making. The pragmatic aspects of the communicative act between the lexical and pictorial texts of the picture book and the reader-viewer are embodied within the methodological framework that addresses the cultural dimensions of signification systems by building them into the intensional and extensional approach to textual analytics in the form of extra-textual influences upon the circumstances of utterance (e.g., "Information about the sender, time and social context of the message, suppositions about the nature of the speech act, etc." [Eco, 1979, p. 14]).

Several important conclusions may be drawn about how the formal aspects of lexical and visual systems of signification embodied as signs and codes within the textual structure of the picture-book as a literary genre work to create the interpretative potential for meaning-making. The model demonstrates how the formal dimensions of text/uality in the picture-book work to "guide" the reader-viewer through the circumstances of its lexical and visual production, or structuration, for example, from the recognition of elements and levels "below" the sign (e.g., semes or coloremes) to elements and levels "above" the sign (e.g., possible worlds or fabula), where meaning-making is dependent upon the reader-viewer's ability to actualize intensionally and extensionally motivated responses (cognitive, affective, and aesthetic) according to individualized systems of conceptual apparati based upon real world experience(s) (Eco, 1979; Greimas, 1983; Saint-Martin, 1990). In essence, the unique aspects of the picture-book's cross-mediality function to induce meaning constructions by provoking and evoking aesthetic responses on the lexical and visual expressive planes of the text while allowing for fundamental cognitive and affective communication to take place "globally" on the content plane. It is in the cross-medial relation of lexical and visual forms that the integration takes place to semiotically imbue the text with meaning potential.

Even so, the case of the picture-book is not simply the reconciliation of the expression of content within the lexical and visual texts of a work, but how the expression of content leads to the creation of personal meaning for each reader-viewer. What the method reveals is how the consciously motivated acts of meaning-making required of, and initiated in, the reader-viewer to realize the signifying potential of the text (at different levels) are reconciled with the extratextual responses achieved by the reader-viewer relative to the signifying structures in a text, but dependent upon contextual factors that influence their perception (e.g., "culture," "education," "training," etc.). The contextual influences of learned codic systems (e.g., lexical or visual) upon individual perception and other experiences that determine "competence" (as defined in relation to cognition of sign structures, e.g., words, colors, etc.) are identified, explained, and accounted for according to the presuppositions of semiotic theory and its methodological arsenal. The dimensions of text in its linear manifestation, both lexical and visual, as a narrative based upon the temporalization of a sequence of events acted out by characters is revealed through the elementary structures of signification, the primary signifying features of which (e.g., a word, a sentence, a color, a line, etc.), convey semantic potential through syntactic construction extending over an larger narrative structure (see Trifonas, in press). It is in this sense that a sequence of related visual frames can be conceptualized linearly as a narrative textual progression and warrant a method of semiotic analysis (similar to that of lexical narrative) developed especially for the purpose of deconstructing how a linear visual narrative is structured.

AS/SA nº 11-12, p.77

The encoding of these elementary structures of signification through which a work achieves meaning and life as narrative is analyzed in terms of the reader-viewer's creation of a "possible world" conceived as a construct (from individual experience) upon which disbelief is suspended. It is true that the elementary structures of signification engender the textual form of lexical and visual narrative structures in the picture-book (e.g., sentences, paragraphs, visual frames, etc.), but it is not in direct relation to them on a microstructural level that the reader-viewer makes abductions in the form of macropropositions or comparative responses (e.g., forecasts and inferential walks or visual indexes) regarding the resolution of the plot as fabula (and as an intertextual or paradigmatic entity). Some of the reader-viewer's cognitive, affective, and aesthetic responses are clearly subconscious interpretative acts that concede disclosures around which is generated a force field of semantic potential (e.g., the seme is a postulate for this type of reaction). The reduction of the narrative into "sequences" according to the interplay of actants (e.g., subject vs. object, sender vs. receiver) allows the thematic roles of the actors (characters) or visual actors governing the narrative structures of the lexical and visual text to be revealed occurs however at the macrostructural level (see Trifonas, in press; 1992). The thematic roles of the actors being acknowledged as "real" develop the ideological motivation of a given text and predicate the action of the plot accordingly through the characters on the level of narrative structures and elucidate the fabula. The culmination of the aesthetic experience of reading-viewing a text is dependent upon the accessibility and the viability of the vision in relation to the textual world structures, both lexical and visual, and the extent to which they are aligned on the level of the fabula within the "global" possible world of the text. The picture-book itself demonstrates that there is a definite self-supportive framework of cross-medial engagement between the lexical and visual components of the text on all levels that functions to develop the linear narrative manifestations of the plot in each codic milieu.

The progression from the possible world visions portrayed in a text to deeper real world understandings is a matter of suspending disbelief and accepting the conventions of the genre as applicable fabulaic alternatives for everyday life (see Kiefer, 1988). Although it is not true in a literal sense that art is more vivid than life, the imagination reconstitutes life through art and vivifies it as a heightened portrayal of the human condition from which we learn more about ourselves. The picture-book, by employing both visual and lexical modes of communication, serves through cross-mediation to supply the reader with an experience novel to the work but dependent on the world of the self.


AS/SA nº 11-12, p.78



Arnheim, R. (1974) . Art and visual perception: A psychology of the creative eye. Berkeley: University of California Press. (Original work published 1954).

Barthes, R. (1964) . Rhetoric of the image. In R. Barthes (Ed.), Image — music — text (pp. 32-51). New York: Hill & Wang

Barthes, R. (1967) . Criticism and truth (K. Kevnenan, Trans.) . London: Athlone Press.

Barthes, R. (1970) . S/Z (Richard Miller, Trans.) . New York: Hill and Wang.

Carter, C. (1972) . Syntax in language and painting. The Structuralist, 12, pp. 54-50.

Eco, U. (1976) . A theory of semiotics. Bloomington: Indiana University Press.

Eco, U. (1979) . The role of the reader. Bloomington: Indiana University Press

Eco, U. (1984) . Semiotics and the philosophy of lang uage. Bloomington: Indiana University Press.

Frye, N. (1957) . Anatomy of criticism. Princeton: Princeton University Press.

Golden, J.M., & Gerber, A. (1990) A semiotic perspective of text: The picture story book event. Journal of Reading Behaviour, 22(3), pp. 203-219.

Gombrich, E. (1960) . Art and illusion. London: Phaidon Press.

Goodman, N. (1968) . Languages of art. New York: Bobbs-Merrill.

Greimas, A. J. (1983) . Structural semantics: An attempt at method.

MacDowell, R. Schleifer & A. Velie, Trans.) . Lincoln: University of Nebraska Press. (Original work published 1966).

Greimas, A. J. (1987) . On meaning: Selected writings in semiotic theory (P. J. Perron & F. H. Collins, Trans.) . Minneapolis: University of Minneapolis Press. (Original work published 1970).

Hjelmslev, L. (1943) . Prolegomena to a theory of language. Madison: University of Wisconsin.

Hodge, R. (1990) . Literature as discourse: Textual strategies in English and history. Baltimore: John Hopkins University Press.

Iser, W. (1978) . The act of reading: A theory of aesthetic response. Baltimore: John Hopkins University Press.

Kiefer, B. (1988) . picture-books as contexts for literary, aesthetic and real world understandings. Language Arts, 65(3), pp. 260-271.

Krampen, M. (1973) . Iconic signs, supersigns and models. Versus, 4, pp. 101-108.

AS/SA nº 11-12, p.79

Kristeva, J. (1969) . Semiotiké. Paris: Seuil.

Landes, S. (1987) . picture-books as literature. In B. Harrison & G. Maguire (Eds.), Innocence and experience: Essays & conversations in children's literarure (pp. 315-322). New York: Lothrop, Lee &Shepard.

Lewis, D. (1990) . The constructedness of texts: Picture books and the metafictive. Signal, 62, pp. 131-146.

Lotman, Y. (1990) . Universe of the mind: A semiotic theory of culture. (A. Shukman, Trans.) . New York: I. B. Taurus & Co.

Mitchell, W. J. T. (1986) . Iconology: Image, text, ideology. Chicago: University Press.

Morris, C. (1946) . Signs, language and behavior. New York: Prentice-Hall.

Nöth, W. (1990) . Handbook of semiotics. Indiana: Indiana University Press.

Peirce, C. S. (1931) . Collected papers. Cambridge: Harvard University Press.

Saint-Martin, F. (1990) . Semiotics of visual language. Indiana: University of Indiana Press.

Saussure, F. de. (1916) . Course in linguistics (W. Baskin, Trans.) . New York: McGraw-Hill.

Sonesson, G. (1989) . Pictorial concepts: Inquiries into the semiotic heritage and its relevance to the interpretation of the visual world. Sweden: Lund University Press.

Schwarcz, J. H. (1982) . Ways of the illustrator: Visual communication in children's literature. Chicago: American Library Association.

Trifonas, P. (1992). Explorations in the semiotics of text: A method for the semiotic analysis of the picture book. Unpublished master's thesis. University of British Columbia, Vancouver.

Trifonas, P. (1993). Conceptions of text and textuality: Critical perspectives in literary theory from structuralism to poststructuralism. Interchange, 24(4), 381-395.

Trifonas, P. (1998). Cross-mediality and narrative textual form: A semiotic analysis of the lexical and visual signs of the picture-book. Semiotica.

Winston, P. (1977) . Artificial Intelligence. Reading, Mass.: Addison-Wesley.

E-mail to the editors
Pour écrire à la rédaction

© 2002, Applied Semiotics / Sémiotique appliquée