Go to Semiotic Review of Books Home Page
Go to SRB Highlights
Go to SRB Archives
This article appeared in Volume 1 (2) of The Semiotic Review of Books.
Sémiologie du langage visuel, by Fernande SaintMartin. Sillery (Quebec): Presses de l'Universite du Quebec,1987.307 . ISBN 2-7605-04334.This book is due to be published in English in September 1990 by Indiana University Press under the title: Semiotics of Visual Language (Advances In Semiotic Series) ISBN G25S 35057-3 The tasks set the reviewer by Fernande SaintMartin's "Semiologie du langage visuel" is not an easy one. No doubt it must be included, together with four or five other books, in the small set of really essential contributions made so far to pictorial, or more broadly, visual, semiotics. On the other hand, it presents itself as "the" book: it would like its reader to abandon his/her ancient creed (or, if you prefer, paradigm) and become a convert to another, while leaving behind, in the process, all that depraved knowledge which had been formerly so painstakingly acquired. That may be a little too much to ask. Yet Saint-Martin's magnum opus is so replete with innovative propositions, striking observations and enlightening comments that the reviewer feels himself obliged, at this point, to suspend judgement on the exact value which should be attributed to the different pieces which make up her "system". What follows, then, is merely a preliminary fragment of a continuing review. Pictorial semiotics could perhaps be described as that part of the science of signification which is particularly concerned to understand the nature and specificity of such meanings (or vehicles of meaning) as are colloquially identified by the term "picture". Thus, the purview of such a speciality must involve, at the very least, a demonstration of the semiotic character of pictures, a study of the peculiarities which differentiate pictorial meanings from other kinds of signification (particularly from other visual meanings, and/or other meanings based on iconisity, or intrinsic motivation), and an assessment of the ways (from some or other point of view) in which different species of pictorial meaning may differ without ceasing to inhere in the category of pisture. Furthermore, visual semiotics, if indeed it is a reasonable subdomain of semiotics, wouid comprise pictorial semiotics, in addition to the semiotics of soulpture and architecture, perhaps the semiotics of ciothing, of writing, and so on, and certainly the semiotios of gesture.
If semiotics as a whole may be said to take its origin in antiquity, or at least in the times of Locke and Leibniz, the part concerned with pictures must be considered a very recent spesiality indeed. If we except a few scattered remarks which may be found in the classic works of Peirce, Saussure, Dégerando, and a few others, Barthes' famous 1964 articie analyzing a photograph which extols the delights of Panzani spaghetti may be taken as its starting point. That articie has often, not unreasonably, been thought to constitute a false start, for, on one hand, it empioys a model taken from linguBstics, which is not very adequate for the purpose, and which, in addition, is erroneously and even confusingly implemented and, on the other hand, it is in fast devoted to an investigation of the significations conveyed, not so much by the picture itself, as by the depisted objects; that is, it really amounts to a kind of "rhetoric of the referent", when we would have expected a pictorial rhetoric. More or less modified, this model was later taken over by numerous researchers in different countries preoccupied with the study of advertisement pictures; while in France it was transferred with little modification to the study of fine arts by such writers as Damisch, Marin, Schefer, and Lyotard, who were then rapidly disenchanted with such tools as were provided by the model.
A more serious beginning was made in the early seventies, which seems to have been developed in parallel fashion by the Paris school, represented, notably, by Floch (1984) and Th&uurlemann (1982) and by the Belgian Group MU (1980), both of which introduced a distinction between two layers of the picture, the iconic, by means of which it produces an illusory rendering of a real-world perceptual scene, and the plastic, whereby more abstract meanings are conveyed using the properties possessed by the pictorial surface as such. In other respects, these two approaches are very different, as had already become clear from the fact that the Paris school applied a series of close readings to an array of particular pictures, where Group MU produced a complex taxonomy, illustrated by exemplary pictures, which crossclassified figures of visual rhetoric according to the way in which they generate divergences from the plastic and/or iconic norms, that is to say, from that which is normal, or expected, in the plastic and iconic layer of a picture, respectively.
It is the conviction of the present reviewer (though not, it would seem, of Saint-Martin) that both these approaches have contributed enormously to our present understanding of the nature of pictorial meaning, as have, in addition, numerous other investigations of a more theoretical character, such as those initiated by Eco (1970), Krampen (1983), and Espe (1983). At the same time, there are excellent reasons for believing none of these approaches to be entirely satisfactory, not only because they all tend to concentrate on limited aspects of pictorial meaning to the neglect of others, but also because their treatment of these aspects is, in some respects, inadequate. A crucial question is, for instance, how we are to discover what kinds of meaning are really conveyed by the plastic, as against the iconic, language, and how the plastic layer of a picture relates to the iconic one contained in the same picture.
This question is posed, in a particularly urgent fashion, by the existence of pictures, the iconic layer of which appears to be nonexistent, that is to say, by the presence of abstract or non-figurative paintings. In his attempt to attribute a meaning to Kandinsky's "Compostion IV", Floch must in fact suppose, as we have shown elsewhere (Sonnesson 1987: 1989), that plastic language is entirely redundant in relation to iconic language, and in fast highly impoverished, because of its being more abstract. If so, then the interest of analyzing plastic language separately is not apparent. It seems reasonable to posit other sources of plastic meaning than those reflected on the iconic level, but it is not at all clear how they could be determined. In this respect. Saint Martin's approach turns out to be highly productive, for she is almost exclusively concerned with what, in the above-mentioned conceptions, would be called plastic language, and she certainly takes its sources to be autonomous.
According to Saint-Martin, students of visual language so far have failed to discover an elementary unit equivalent to the phoneme, because they have expected to find too great a similarity between the units of the diverging systems. In the case of visual language, this unit is the "coloreme", which may perhaps roughly be described as a minimally perceptible surface point, or as that amount of information which is taken in by the viewer in one glance as it is organized around a single point of fixation. Having introduced this intriguing notion, the author proceeds, in the second chapter, to explore the nature of six visual variables, which turn out to be dimensions on each one of which every surface point must evince a value: these are termed colouritonality, texture, dimension/quantity, implantation into the plane, orientation vectoriality, and frontiers/contours generating shapes.
After these unavoidable preliminaries, Saint Martin goes on, in the third, very short chapter, to present us with the rules of visual syntax, by means of which the coloremes are "chunked" together. There seem to be three kinds of such rules: the interaction of colours; topological relations, such as proximity, separation, encasing, envelopment, and orders of secession, and finally Gestaltian relations, such as figure ground, contiguity, similarity, closure, good forms, vectoriality and, curiously, notion having a high degree of familiarity. Thus far we have been concerned with visuality generally, but, according to Saint-Martin, "coloremes" and visual variables are differently constrained when they are permitted to intermingle inside the basic plane, which is presented as a precondition on the creation of each and every picture, in the virtual cube which is the condition of possibility of sculpture, and in the environmental cube which determines the potentialities of architecture. After having elucidated the constraints which characterize the picture plane, the author turns to a long review of distance effects and pictorial perspectives, which are said to number about two dozen, all of them supposedly conventional. Then the virtual cube is discussed (but nothing is said about the environmental cube), and the book closes with some fairly general remarks on how to apply this multifarious analytical tool kit to the analysis of particular pictures.
It is to be regretted that no concrete picture analysis has been included in the book which, in its present form, cannot avoid leaving its reader wondering about the exact meaning of the spectacie he has been witnessing. Elsewhere, it is true, Saint-Martin (1986) has published an elegant and very thorough analysis of "Mascarade" by Alfred Pellan, and another work by the same artist is analyzed by Marie Carani (1986) along very similar lines. But these articles, in turn, seem largely incomprehensible to a reader unacquainted with the book under review (as was the case of the present reviewer when first encountering the articles in question).But let us now turn to the strengths of this approach. One of these is no doubt the author's determination to find a separate, autonomous, origin for plastic language, which, in addition, permits the analysis of fairly "abstract" pictures, such as those created by Pellan. In particular, it is interesting that, in so doing, she makes a systematic use of the relations defined by mathematical topology which were found by Piaget to make up the primary spatial conception of humans as manifested in the reasoning and drawings of small children. On the other hand, it is unfortunate that Saint-Martin should devote such a limited space, hardly more than three pages, to the explanation of these topological notions, which are, in a way, the principal innovation of her book. Her originality should not be exaggerated, however. Earlier, Pierre Boudon (1981) introduced mathematical topology into architectural semiotics. Also, although he does not insist on their topological character, Floch (1984) actually elaborated a much more complex taxonomy of the different varieties answering to the topological concept of inclusion, in particular in his analysis of Boubat's "Nu" and in his study of the "News" publicity. A further deficiency of Saint-Martin's use of topology is her failure to account for the relevance of these relations to perception generally, and pictorial perception in particular, if we except the reference to Piaget's work. Such a justification may be forthcoming: as we have pointed out elsewhere, the modal vectorial properties, which may be conceived, along the lines of the cognitive psychologist Howard Gardner (1970; 1973), as a kind of bodily generalization paralleling the cognitive ones studied by Piaget, may in the end turn out to be reducible to topological properties, once a dynamic element is added to each one of them (cf. Sonesson 1989 (9:90-109)
Saint-Martin's merit is that of having applied topology explicitly and systematically to the analysis of pictures. Yet we must note a further shortcoming of her approach, which consists in her neglecting to clarify the relationship between the topological notions and the Gestaltian ones, which constitute her second class of syntactical relations. Some of these notions are clearly related, as is the case of proximity and contiguity. Furthermore, it would be important to elucidate the relationship of these latter notions to indexicality and to the syntagmatic axis, both of which involve contiguity.
The second important contribution of the book is the idea of there being certain constraints imposed on the visual variables when they contribute to form a picture, and other such constraints restricting the possibilities of sculpture. This idea is important in many respects. First of all, Saint-Martin accounts in this way, at least to some degree, for the differences between at least two kinds of visual meaning, painting and sculpture. In relation to the requirements stated at the beginning of this review, Saint-Martin is thus seen to contribute to the analysis of the difference between pictorial meaning and other visual meanings. On the other hand, she has nothing to tell us about different pictorial kinds, and very little is said about the semiotic nature of pictures. Furthermore she entirely neglects iconisity, simply accepting Eco's rather elementary critique (cf.Sonesson 1989).
In the second place, the very fast of outlining a semiotics of sculpture is in itself an important expioBt. There have been very few studies of sculpture in semiotics and although both Dora Vallier and Rosalind Krauss have suggested a few characteristic features of sculpture, their definitions are really too aprioristic in character to be taken seriously.
Thirdly and perhaps most importantly, at least in the case of the pictorial basic plane, Saint-Martin's descriptions of these constraints turn out to form a very important prerequisite for the interpretation of the visual variables which it circumscribes. In fast, the idea of there being a basic plane stems from Kandinsky, but in a very lucid discussion of his conception, Saint-Martin shows it to be marred with contradictions. Yet it never becomes clear why she should reject the asymmetric variety of the basic plane which she has so elegantly extracted from Kandinsky's writings and for which she even quotes some perceptual evidence.
Her own view of the basic plane, according to which it contains four formative angles, two diagonals, which give rise to two pairs of partly overlapping triangles, a cruciform, and a lozenge, is presented without any real justification, and although it may seem intuitively satisfactory, it appears that many other divisions could be equally satisfactory. Her model also gives rise to too many contradictory potentialities. Thus, there is supposed to be a kind of energetic radiation which emanates from each of the corners and which decreases when approaching the centre, but, on the other hand, there is also an augmentation of energy in the centre, which is due to the crossing of the diagonals and the cruciform structure. An interesting suggestion is, however, that the occurrence of certain shapes, as for instance open forms, may have the effect of potentializing energy near the limits, while other shapes produce the opposite effect. But, here Saint-Martin is merely describing a field of inquiry, where work has hardly begun.
A further interesting aspect of Saint-Martin's work is her attempt to classify a number of different perspectives or distance effects. Unfortunately, these are not characterized in relation to each other in any more systematic way so that the model fails to afford us with any intersubjectively applicable criteria for identifying and separating the distance effects. Indeed, only a few already well-known perspectives, such as the linear perspective, the Cubist perspective, the isometric perspective, and so on, could be reliably identified.
We have been concerned so far with those respects in which Saint-Martin's book proposes an array of innovative conceptions and original tools which should be incorporated into any future model of pictorial semiotics. If, at the same time, we have not hesitated to note some problems connected with these tools and conceptions, it is because we believe that the only fruitful way of absorbing such instruments consists in trying to modify and elaborate them further. In what follows, however, we will have a constructive look at a few more obvious shortcomings of Saint Martin's model.
Two opposite, but complementary, requirements could be imposed on a semiotic approach. First, when analyzing an object which has been studied also by other disciplines, it should take full account of what has been learnt so far. In the second place, in so doing, it should retain its particular proper point of view, and not simply merge with one or another of the other disciplines studying the object in question. On a very general level, Saint-Martin's book, contrary to the many other putative works of pictorial semiotics, manages to steer free of both perils. Still, on more particular points, both prerequisites appear to pose some problems.
If there is one thing which is particularly characteristic of semiotics, then it is the determination to take the point of view of the sign user himself, that is, to describe the expression merely to the extent that it conveys a content and vice-versa. Prieto, repeatedly quoted by Saint Martin, has often insisted on this point. But if Saint-Martin had taken this requirement seriously, we could have been spared the long and tortuous discussion (23 pages) of what kind of colour analysis is relevant to semiotics. More importantly, she should not have produced a list of visual variables in the definition of which perceptual, physical and physiological criteria are used indiscriminately. It is true that Jakobson has often been accused of a similar confusion in his famous definition of phonological features, but (to put it in the terms of a classical doctrine of art~ even the faults of giants are faults and should not be imitated.
As for the other requirement, Saint-Martin turns out to be remarkably unfamiliar with at least some aspects of resent psychology which would seem to be of importance to her theme. She fails to mention the recent important inquiries into the psychology of picture perception by Gibson (1979~ Hochberg (1978), Hagen (1980) and Kennedy (1974,1984), the latter, like Saint Martin, a contributor to the monographs of the Toronto Semiotic Circle. Also she never acknowledges recent advances in cognitive psychology, some of which tend to put the work of Piaget, on which she relies heavily, into critical perspective. The consequences of this omission are not always serious: thus, for instance, in suggesting that the colours must be conceived as "chromatic poles", she seems to rediscover for herself the important concept of the prototype, introduced by Rosch (eg.1978) which could be described as the use, for the determination of category membership, of approximations to the best instances, instead of sufficient and necessary criteria. It is unfortunate, in any case, that lacking a clear notion of the prototype, she fails to make any real use of it in the sequel. Since Gestalt psychology is an important source of information to Saint-Martin, it is particularly regrettable that she should ignore Rosch's demonstration that "good forms" are kinds of prototypes.
Much more damaging is her neglect of resent psychology of perception. At one point this neglect is even somewhat embarrassing: Bishop Berkeley's old conception, according to which depth is not seen but somehow cognitively reconstructed using clues derived from touch, is quoted with approval from the art historians Riegl, Wolfflin, and Worringer who still propounded this theory around the turn of the century. However, if the idea of depth being a merely tactile sensation has by now been unacceptable in psychology for almost a century, even the more general notion that is has to be constructed instead of being directly perceived at present seems incompatible with the facts. When late James tieDson rediscovered the old insight of Husserlean phenomenology according to which it is the thing itself, not some two-dimensional perceptual field, which is directly seen. Yet the opposite, erroneous conception turns out to be fundamental to much of Saint-Martin's reasoning, in particular in her discussion of perspective and in the analysis which she performs of the potential cube supposedly defining sculpture.
Serious doubts must also be voiced about Saint Martin's claim to having discovered, in the form of the "coloreme", the elementary unit of visual meaning. The issue of determining to what extent there is in pictures and sculptures something even remotely comparable to the phoneme and/or the morpheme, is a complex one, as is the further question of discovering how pictorial features differ from linguistic ones. Saint-Martin's discussion of the visual variable is useful and may perhaps fruitfully be taken as a point of departure for a reflection on the peculiar character of visual features. The "coloreme", on the other hand, merely causes further complications. After its introduction in the first chapter, it is relegated to a shadowy existence all through the book, only to be brought up again in the last chapter, where it turns out to be something very different from what the first pages tended to suggest.
The "coloreme", it will be remembered, is an aggregate of visual variables perceived in the visual world through a centration or fixation of the eye, giving rise to a macular, and then a peripheral field surrounding the foveal area of clear perception. In its primary sense, the "coloreme" is thus a segment, not of the object perceived, but of the very process of perception, and in that respect, as well as in its being structured as an array of adumbrational variants organized around a thematic centre, it is clearly comparable to what is known in Husserlean phenomenology as the perceptual noema and to what the psychologist of perception Julian Hochberg terms, more colloquially, a"glance".
There can be no one-to-one mapping of the units of the perceptual process onto the units of the object, however. Joining together the kindred traditions of Gestalt psychology and phenomenology, Gurwitsch (1957) made a very thorough analysis of perception showing that the object is entirely contained, differently adumbrated, in each one of the corresponding noemata, his result still seems compatible with what we know from perceptual psychology. But if the object is merely an "internoematic system", as Gurwitsch puts it, it must be made up of an infinite number of different adumbrations which more or less overlap each other, permitting the constitution of Gestalt-coherence. Any selection of elements from the internoematic system will be arbitrary and is thus unable to lay the basis for an analysis of the object, that is, in this case, the picture.
In the case of picture perception, however, investigations, since the time of Boswell, have tended to demonstrate the existence, if not of an order of reading, then at least of certain points of fixation where the glances tend to cluster. In this sense, the noemata would seem to offer a kind of indirect access to the thematic centres of the picture, to the extent that those points attracting recurrent glances could be taken to define a small set of privileged noemata extracted from the internoematic system. However, since noemata have no clear limits, but tend to shade into each other and encompass one another, they could not, even in this case, delimit the real building blocks of the picture. Furthermore, the topological relations of the noemata are not at all those of the parts of the object contained in them. In any case, when returning to the "coloreme" in the last chapter of her book, Saint-Martin denies every relevance to these results without even attending to the difference between the hierarchies defined by the order of glances and by their recurrence.
Instead, the "coloreme" now turns out to be a unit defined ad hoc by the analytical scheme resulting from Saint-Martin's model. In order to analyze a picture an arbitrarily divided grid is imposed on it. Such a grid, according to Saint Martin, may well possess 15, 20, 25, or any other number of squares, although she arbitrarily decides to employ the latter number, resulting from five horizontal and five vertical divisions. Moreover, each square, designated by a letter of the alphabet may be further divided by the numbers 1-5, corresponding to each of the sides of the square, in addition to its central point. And here comes the surprise: these 125 points are identical to the "coloremes"! Thus the points of fixation defining the "coloremes" are simply imposed ad hoc by the model!
Curiously, Saint-Martin then proceeds to suggest that each of the squares is comparable to the basic plane as a whole in containing a differential distribution of energies corresponding to the cruciform structure. But this is a strange suggestion, if we remember that the number of squares of the grid, and thus the localization of the limits between the squares, was arbitrarily determined. Saint-Martin herself here seems to forget the conventional character of the framework imposed by her model.
None of the preceding remarks serves to deny the operative value of this analytical grid when applied to the analysis of concrete pictures. It should be clear from our description that Saint-Martin's model has quite a different coverage from the models suggested by Floch, Thürlemann, and Groupe MU: it actually scrutinizes every single point of the pictorial surface. Elsewhere, we have observed that, whereas Fioch tends to take his departure from some very large, higher-order structures, and then goes on to reduce this organization into a series of underlying bundles of features, Thürlemann, at least in his Klee book, rather isolates single elements, out of which he tries to build up complex wholes (cf. Sonesson 1989;1990). In spite of her repeated references to Gestalt psychology, Saint-Martin goes even further in the sense of atomism.
Saint-Martin's procedure is, of course, a good check on many an overly adventurous hypothesis; at the same time, however, one wonders if there is not a risk of getting lost in empirical facts without beginning from some specific hypothesis, such as Floch's binary divisionism. Such a specific hypothesis is, in fact, built into her analysis of the pictorial base plane. On the whole, the systematic character of Saint-Martin's model is an essential contribution to pictorial semiotics.
It may seem strange that, having declared Saint Martin's book one of the very few essential contributions to pictorial semiotics, we have proceeded to criticize most of the ideas it contains. This procedure must be interpreted in connection with the present state of the art: when pictorial semiotics finally begins to emerge as a serious discipline, what we need are not another Summa semiotisae, which closes off the discussion. We need books that makes us think and which give us opportunity to revise and recast our models. Ordinary science is not for tomorrow.
This book will be of interest, not only to specialists in pictorial semiotics, but to all those who take an interest, professionally or otherwise, in the meaning structures of pictures. Although Saint-Martin, unlike most other workers in the field, would seem to consider only fine arts to the exclusion of all other pictures as her domain, there is really no reason why the elaborate model she has built should not be fruitfully applicable to other kinds of pictures. The problems raised here could certainly not impede this extremely meticulous model from yielding interesting results to those researchers having the patience to go through all its phases and moments.
In an earlier review of this book, Luc Régis (1987) claimed that the issue of pictorial interpretation could not be resolved on the level of mere perception, but would have to be posed instead on the level of culture. He is certainly quite right. But in order to reach that level, we need to have access to more knowledge, not less, about pictorial perception than that which is afforded us by Saint Martin's approach.
Barthes, R.(1964) "Rhétorique de l'image." Communisations 4:40-51.
Boudon, P.(1981) Introduction è une sémiotique des lieux. Montréal: Presses de l'Université de Montréal.
Carani, M.(1986) "Le corpus Pellan: une relecture semiotique." Protée 14 3:15-26.
Eco, U.(1970) "Sémiologie des messages visuels." Communications 15:11-51.
Espe, H.(1983) "Realism and some semiotic functions of photographs." In T. Borbe (ed.) Semiotics Unfolding Proceedings of the second congress of the International Association for Semiotics Studies Vienna, July 1979. Berlin: Mouton de Gruyter. (1431;1442)
Floch, J.-M. (1984) Petites mythologies de l'oeil et de l'esprit Paris: Hadés.
Gardner, H. (1970) "From mode to symbol." British Journal of Aesthetics. 10, 3:359-375.
---. (1973) The Arts and Human Development. New York: Wiley and Sons.
Gibson, J.J. (1979) The Ecological Approach to Visual Perception. Boston: Houghton Mifflin.
Goodman, N.(1968) Languages of Arts. London: Oxford University Press.
Gurwitsch, A.(1957) Théorie du champs de la conscience. Bruges: Desclée de Brouwer.
Hagen, M.(ed.) (1980) The Perception of Pictures. Vol. 1: Alberti's Window. Vol. 11: Dürer's Devices. New York: Academic Press.
Hochberg, J. and Brooks, V. (1978) "The perception of Motion Pictures." In Carterette, E. and Friedman, M. (eds.) Handbook of Perception. Vol.10. New York: Academic Press (259-304).
Kennedy, J.M. (1974) A Psychology of Picture Perception. San Francisco: Jossey-Bas Inc.
---. (1984) Vision and Metaphors: Emperical Investigations. Publications of the Toronto Semiotic Circle. Toronto: Victoria University.
Krampen, M. (ed.) (1983) Visuelle Kommunikation und/oder verbae Kommunikation. Berlin: Olms Verlag, Hildesheim und Hochschule der Künste.
Lindekens, R. (1976) Eléments de sémiotique visuelle. Paris: Klincksieck.
Mu, Groupe (1980) Plan d une rhétorique de l image. KodEcas/Code. 3:249-268.
Regis, L. (1987) Review of: Fernande Saint-Martin, Sémiologie du langage visuel. Bulletin du Group de Recherches Sémiolinguestiques. 10, 44:61-62.
Rosch, E. (1978)"Principles of Categorization." In Rosch, E. and Lloyd, B. (eds.) Cognition and Categorization. Hillsdale (NJ): Lawrence Erlbaum (27-48).
Saint-Martin, F. (1985) Introduction to a Semiology of Visual Semiotics. Publications of the Toronto Semiotic Circle. Toronto: Victoria University.
---. (1986) Sémiologie et syntax visuelle. Une analyse de Mascarade de Pellan. Protée. 14, 3:27-40.
Sonesson, G. (1987) "Notes sur le macchia de Kandinsky: Le probleme du langage plastique." Actes Semiotiques: Le Bulletin X 44 (23-28).
Thürlemann, F. (1982) Paul Klee. Analyse sémiotique de trois peintures. Lausanne: L'Age d'Homme.Göran Sonesson holds doctorates in Linguistics and Semiotics respectively from the University of Lund (Sweden) and the Ecole des Hautes Etudes en Sciences Sosiales (France). From 1974 to 1983 he conducted research in general semiotics and the semiotics of gestures in Paris, and in Mayan linguistics in Mexico. In 1983 he was appointed director of the Semiotics Project at the University of Land. He represents Sweden on the Executive Committees of the International Associations for Semiotics Studies and of the Nordic Association for Semiotic Studies, and recently took an active part in the founding of the International Association for Visual Semiotics (Tours, France, 1989). He is the author of Pictorial Concepts. Inquiries into the Semiotic Heritage and Its Relevance for the Analysis of the Visual World. (Lund: ARIS/Lund University Press, 1989). His most recent contribution to visual semiotios include: "Le mythe de la triple articulation. Modèles linguistiques et perceptifs dans la sémiotique des images." Fourth congress of the International Association for Semiotic Studies, Barcelona/Perpignan, Mars-April 1989, to be published in the proceedings. "Tjugofem ars soppa pa Panzanis pasta. Postbarthesianska betraktelser över bildsemiotiken." Norwegian Association for Semiotic Studies, November 1989 to be published in Livstegn, 1. 1990, and "Semiotik der Bilder. Forschungsbericht am Anfang der neunziger Jahre," forthcoming in Zeitschrift für Semiotik (1990).