Go to Semiotic Review of Books Home Page
Go to SRB Highlights
Go to SRB Archives


SRB Archives

This review appeared in Volume 1 (3) of The Semiotic Review of Books.

Speaking of Pictures

by Philip Alperson

Pictorial Concepts: Inquirys Into the Semiotic Heritage and Its Relevance to the Interpretation of the Visual World by Göran Sonesson, Maimö, Sweden: Lund University Press, 368 pages, 1989, ISBN 91-7966-062-2

Göran Sonesson's Pictorial Concepts is an ambitious book. In it, Sonesson attempts a critical survey of European and American semiotic theories and present developments in psychology, philosophy and linguistics in order to assess the prospects of a semiotics of the pictorial which would "discover the laws and regularities attendant on each and every case of depiction." Sonesson does not hide his basic conviction that semiotics is "an indispensable tool of intellectual understanding of our time." (7) His attempt to be thoroughly critical in his consideration of the theoretical underpinnings of pictorial semiotics is a reflection of his laudable intention to insure that the discipline of semiotics, if it be such, advance on something more secure than the enthusiasm of some of its aficionados.

What sort of a discipline would pictorial semiotics be? For a start, Sonesson argues, it would have to be a genuinely ecumenical study. It would be scientific if the sense that it would take as its main task the discovery of general laws presumed to obtain in the realm of pictorial depiction. It would draw upon the insights of art history, philosophy, cognitive and social psychology, linguistics, anthropology, literary theory and more. The picture painted here leads rapidly to the suspicion that the sort of visual semiotics Sonesson has in mind seems hardly a discipline at all, that it seems more a tissue of borrowings than a branch of learning with a unified, specific and characteristic set of rules and goals to guide its operations.

It comes as no surprise, then, that the first of the three parts of Sonesson's book begins with a critical discussion of the aims and methods of semiotics in general and of pictorial semiotics in particular. On Sonesson's view, the main pillars of semiotics are the disciplines of linguistics, psychology and philosophy, each of which earns its foundational status by virtue of a distinctive contribution to the study of meaning. A pictorial semiotics in particular would build on the basic insight of linguistics, especially the idea of "textual analysis", without making the mistake of assuming that all (or even most) examples of visual depiction could be understood on the model of a "linguistics" of the visual image. Sonesson values the linguistic model of meaning less for its account of the mechanics of linguistic meaning than for its heuristic value: general semiotic features such as the sign function, articulation and connotation may be discerned in a putative semiotic system even though there be nothing which answers exactly to such specifically linguistic features as the phoneme, word order, predicates, syntagmas, etc. Psychology contributes much to the understanding of pictorial meaning, not only in terms of the psychology of perception, but also from the perspective of social psychology. Philosophy provides pictorial semiotics with important methodological and conceptual tools as well as with careful analyses of the meanings of meaning, especially in the phenomenological tradition of Husserl and his followers.

The discipline of semiotics, then, is best viewed as a particular intellectual tradition concerned with the understanding of meaning. The tradition may be understood in a preliminary way according to Locke's formulation as the study of "the nature of signs the mind makes use of for the understanding of things, or conveying its knowledge to others," if it be added that not all meanings are signs strictly so-called. The tradition includes both a philosophically oriented, primarily German concern with the classification of signs into types and a more empirically oriented French strain. The tradition is basically nomothetic, concerned with general principles according to which meaning is constituted by human beings, but it embraces a "constellation" of methods. As against those who would withhold the label of "discipline" to such an enterprise, Sonesson remarks that "most other disciplines would not come off much better, but since they are well-established, they no longer have to justify themselves." (29)

Having dispensed with general methodological issues in this manner, Sonesson proceeds to identify and examine various concepts from semiotic theory and from related fields which seem promising for the understanding of pictorial meaning. It should be said at the outset that the ambitiousness of Sonesson's project stems not only from the panoramic view he attempts of the field of pictorial semiotics but also from the very detailed analyses of a myriad of concepts, theories, studies, neologisms and internecine quarrels which at times threaten to overwhelm the reader (if not the argument) of Sonesson's book. Sonesson tries to keep order amongst the various terms, theories, allusions and roads not taken by indicating (in virtually every paragraph) another section (or sections) where the topic at hand is discussed and by providing frequent summaries. But it must be said that when Sonesson himself characterizes sections of his book as "tortuous" (25), even "long and tortuous" (342), he speaks the truth.

Space precludes a detailed examination of Sonesson's treatment of such complex and wide-ranging concepts as indexicality, contiguity, perception, category, hierarchy, structure, configuration, operativity, figurativity, pertinence, relevance, iconisity, and more, or of the wealth of examples Sonesson adduces to support his analyses. Sonesson's general approach to these concepts, relations and operations is to analyze them in the twin contexts of a Peircian understanding of the nature and function of signs and the Husserlian concept of a "lifeworld". The strategic importance of these two points of reference serve to anchor the operations of a pictorial semiotics in a field of constancies. From Peirce comes (among other things) the triadic conception of signification which assigns a crucial role to the interpretant and process of interpretation according to which a sign may be said to have a significance, as well as the Peircian notion of abduction, the movement of the mind according to which hypotheses are formulated on the basis of sensory experience. From the notion of the life world comes the idea that within a particular time and culture, the world presents itself as an "open horizon" and perspectival, yet at the same time possessed of general types of things and regular patterns of behaviour. This phenomenological strain is consistent with Peirce's theory of signs insofar as Peirce, as David Savan has noted (Savan 25-7), insists on the importance of context to explain how it is that signs not only convey but extend knowledge. The Husserlian borrowing also opens up the rich vocabulary of phenomenological philosophy and psychology to describe the synthetic cognitive operations according to which meaning is constructed.

In this context, pictures may exhibit different kinds of indexicallity, for example, in the general sense of a relation between an expression (Peirce's "representament") to a content (or "object"). Sonesson goes to great length to detail the variety of pictorial indexicalities. We may reconstruct the meaning of the Minoan fresso, "Ladies in Blue", for instance, or of certain cubist portraits, both of which seem initially to exhibit rather puzzling positions and shapes, on the basis of a "contiguity of potential expression planes . . together with abductions from principles taken for granted in the particular Life world." (41-3) A picture of an arrow involves an indicative indexicallity in which the sense of direction indicated by the arrow is rooted In an anticipation ("protection") of a following moment rooted in the sort of bodily experience described by Merleau-Ponty. (47) A picture of a clock on a signboard outside a watchmaker's shop exhibits an indexicality which is both abductive, requiring an inference based upon prior knowledge about a relation between a clock and the kind of services to be expected of the watchmaker, and performative in the Austinian sense, creating a state of affairs in the act of utterance (in this case designating the site to which it is contiguous as the watchmaker's shop). (53) On Sonesson's view, certain types of indexicallity are characteristic (though not -criteriological) of pictures, but not all pictorial indexicalities can be understood in terms of the sign function, strictly so-called. The most that we can say by way of general principles is that "order of some sort is essential to all meaning" and that the understanding of pictorial meaning is built up out of relations of mediation, relevance, signification, structure and abduction. (110)

Similarly, an examination of well-known pictorial analyses by Barthes and Floch leads Sonesson to conclude that though we may speak of the connotational "language" and plastic "language" of pictures, these modes of meaning are linguistic only in an attenuated sense: their sign-like meanings are guided (but not determined) by principles of relevance rooted in the life world. The iconisity of pictures receives a comparable treatment. Sonesson argues against the view that the iconisity of pictorial signs is essentially arbitrary or conventional (220-50) he does not go so far as to claim that similarity is a necessary condition for pictorial iconisity, but, he argues, there must be some property "held in common by all pictures, and by no other objects. For not only is there a coherent Life-world notion of pictures, but there is no other way of explaining that pictures have meaning." (249) What is required, in his opinion, is that there be the impression of similarity whose possibility he seeks to explain by means of a Husserlian gloss of several well-known contemporary theories of perception. This aspect of pictorial consciousness is explained largely in terms of "a presentation", a process of pictorial understanding involving both perception and imagination.

Just how far do these analyses take us in understanding what a pictorial semiotics would look like? The picture is not clear. Sonesson is in command of a very wide range of literature in philosophy, linguistics, phenomenological psychology and the psychology of perception and his analyses suggest many promising areas of further study in each of these fields. His insistence on understanding pictorial meaning in the context of the life world's complexities is to be commended. Sonesson ends his book with a succinct sketch of "the prototypical picture" as "a surface, which has been covered with textures foreign to it, in such a way, that it conveys the idea of a scene, which could possibly take place in the ordinary perceptual Life world and which, at the same time, creates the impression that there is a similarity between the surface and the scene." (343) This formulation has the virtue of intuitiveness - it sounds very much like what the person in the street would have said about the meaning of-pictures had he or she been possessed of a richer than normal philosophical vocabulary.

On the other hand, at the end of over 300 pages of close commentary on scores of theories in linguistics, philosophy and psychology, Sonesson acknowledges that he has established neither any new doctrine of pictorial signs nor a final synthesis. (343) Furthermore, Sonesson's approach threatens to undermine, or at the very least, complicate the idea of a discipline of pictorial semiotics. It is not so much the ecumenical or interdisciplinary aspect per se of his approach which is the threat here or even Sonesson's willingness to embrace a "constellation" of methods. The appeal of semiotics stems in large part precisely from its promise to establish a framework for all disciplines which seek to understand the meaning of artifactual signs. One does well to remember that too narrow a conception of what constitutes a "discipline" will lead to an impoverished attempt to understand the world in all its richness and complexity. To abandon the idea of a pictorial semiotics because of its "interdisciplinary" nature is to give up the game before it has been allowed to start.

The real problem which threatens to explode the notion of a discipline of pictorial semiotics comes rather from the daunting task that Sonesson has set for the pictorial semioticist. Early in his book, Sonesson argues for the viability of the "principle of alterfunctionality", the idea that "the criteria which pick out the relevant features and determine the limits of their variability, are to be found in something which is experienced as on another level of reality than the phenomenon itself." (18) The significance of this principle becomes clear toward the end of the book when Sonesson acknowledges that the full elaboration of the sort of pictorial semiotics he has in mind requires a much larger framework - "the more general conditions of possibility of all Life world hierarchies found in the different sociocultural Lifeworlds". Sonesson goes on to say that, "even supposing that to be a task which is possible to accomplish, it lies beyond the scope of the present essay." (282) But if the prospects of a rich pictorial semiotics de require this level of contextualization, it may well be the case that pictorial semiotics is a discipline, but an impossible one.


Barthes, Roland (1957) Mythologies. Paris: Seuil.

Barthes, Roland (1965) Le degre zero de l'esriture/Elements de semiologie. Paris: Gonthier.

Floch, Jean-Marie (1978b) "Roland Barthe: Semiotique de l'image," Bulletin du Groupe de recherches semiolinguistiques, no.415.

Husserl, Edmund (1931)Ideas: General Introduction to Pure Phenomenology. New York: Macmillan.

Husserl, Edmund (1964) The Idea of Phenomenology. The Hague: Nijhoff.

Merleau-Ponty, Maurice (1962) The Phenomenology of Perception. New York: Humanities Press.

Peirce,C.S. (1931-58) Collected Papers of C.S. Peirce. Cambridge: Harvard University Press.

---. (1977) Semiotics and Significs: The Correspondence Between C.S. Peirce and Victoria Lady Welby, C.S. Hardwick, ed. Bloomington: Indiana University Press

---. (1962-) Writings of Charles S. Peirce, Max H. Fisch, gen. ed. Bloomington: Indiana University Press.

Savan, David (1988) An Introduotion to C.S. Peirce's Full System of Semeiotic, Toronto: Toronto Semiotic Circle.

Wolheim, Richard (1980) Art and its Objects, second revised edition. Cambridge: Cambridge University Press.

Philip Alperson is Associate Professor of Philosophy and Chair of the Division of Humanities at the University of Louisville. He is editor of What is Music?: An Introduction to the Philosophy of Music (Haven, 1987) and The Philosophy of the Visual Arts (Oxford, forthcoming) as well as the author of articles and reviews in the field of philosophical aesthetics.

Go to Semiotic Review of Books Home Page
Go to SRB Highlights
Go to SRB Archives