Lecture 1: The Quadrature of the Hermeneutic Circle Historical and Systematic Introduction to Pictorial Semiotics

The first lecture will present pictorial semiotics within the framework of general semiotic theory. It will construe semiotics as a particular point of view taken on everything which is human or, more generally, endowed with life, rather than simply the continuation of the mixed or separate doctrines due to Saussure and Peirce. The historical part will describe briefly the development of pictorial semiotics and the peculiarities of its different schools and traditions, following upon the somewhat premature founding gesture attributed to Barthes.

Contents I.1. Introducing pictorial semiosis: the way pictures mean 4 The picture as a picture and as a sign 5 The act of recognition and the categories of reality 8 The painter caught in the act 9 The picture as part of the world 12 The depictions in the picture and the picture depictions 13 The picture as a thing in the world 15 The inscribed positions of the observer and of the painter 16 Reflections of (and on) other rooms 17 The title and the theme – and the act of depiction 17 A Saussurean look at “Las Meninas”: The sign 19 A Saussurean look at “Las Meninas”: The system 20

 Peircean perspectives on “Las Meninas” 23 The picture as a cognitive-semiotic operation 25 The picture as socio-semiotic operation 25 Summary 25

1.2. Theoretical beginning : Semiotics as a distinctive discipline 26 Out of the semiotical soup shops: Semiotics and Philosophy 26 Two ways of defining sciences: Semiotics vs. the History of Religion 30 The social institution of science: Semiotics vs. Archaeology 33 Meta-analyses in our time: Semiotics and Cognitive Science 36 From Phenomenology to Ecological semiotics 38 The three sciences of communication: rhetoric, hermeneutics, semiotics 42 Summary 45

1.3. System and History: Beyond Barthes and Eco in Pictorial semiotics 45 The domain of pictorial semiotics 47 From Panzani pasta to the theory of iconicity 49 The emergence of pictorial semiotics 52 Some contributions from philosophy and psychology 56 Summary 57

References 5

At least one thing is implied by the very proposal of a course in pictorial semiotics: that a subject matter for such a discipline can be found. It supposes that there is such a thing as a pictorial sign, or, more broadly (in a sense which will be specified in the second lecture),  Fig. 1. Velázquez’ ”Las Meninas”

that there are pictorial meanings. In other contained in the uses to which pictures are words, that homely object known as a pic- put. ture either consists in a sign or some other It is not enough to show that pictures kind of meaning, or is made up of signs may be construed as signs. We must also and/or other meanings. demonstrate that this approach has not The claim contained in the formula- been taken, or not systematically so, within tion of the subject matter of our discipline disciplines antedating pictorial semiotics. is not really about what is “out there”. Mini- Indeed, I will argue that, contrary to many mally, the claim is merely that it is possible others parts of semiotics, pictorial semi- to construe pictures as signs (and/or mean- otics is a quite new endeavour. While the ings), and that, in so doing, we are in some way semioticians have discussed literature way able to illuminate some properties of or culture may appear to be just one in a pictures not evident from other construals. long series of possible approaches to their In fact, however, I am making a somewhat object, one of several version of “literary stronger claim: pictures are really used by theory”, “anthropology”, and so on, there is people (at least some of the time) in ways no precedent for the manner in which pic- which imply that they are signs. This does torial semiotics looks at pictures, certainly not mean that people would ordinarily say not in art history, and only vaguely so in pictures are signs. Indeed, they would not. philosophy. Even this affirmation needs to Rather, the sign character of pictures is be qualified: in some respects, what has

 happened in perceptual psychology follow- generally. ing upon the work of James Gibson may be more important for pictorial semiotics than I.1. Introducing pictorial some things done under this very heading. semiosis: the way It should be pointed out from the start that, in my view, pictorial semiotics pictures mean is something rather different from art his- Pictorial semiotics is the science of all tory, even construed as some more general pictures and all picture types. As a part “Bildwissenshaft”. It is not uninterested in of general semiotics, its primary aim is to pictures classified as art, but it does in no investigate how pictures differ from other way attribute any privilege to them: art is signs and other meanings, and in what simply one of several categories of pictures ways they are similar to them. At the same defined by the social use to which they a time, however, it is involved with explain- commonly put. Instead, we are involved ing the ways in which signs which are pic-  with pictures as the result of a particular tures may differ from each other. cognitive operation, which is, more spe- A picture is primary a perceptual cifically, a semiotic operation, a sub-class task. But it is not just that. Or, rather, it is a which also includes such things as ver- peculiar perceptual task. In spite of all that bal language, gestures, and much more. I has been said above, I will start out from would like to add, as a caveat, that as pic- a picture considered to be a work of art. torial semiotics is dedicated to the under- It has the advantage of being well-known. standing not only of the general category of And this particular case also has other ad- “picture”, but to all possible subdivisions vantages, from the semiotic point of view, of this category, it still retains an interest as we will see shortly. So we will begin by for pictorial art. looking at Velázquez’ “Las Meninas” (Fig. In order to understand the latter 1.). But we will do so as semioticians. point, it will be necessary to go rather deep Adopting a semiotical point of view, into the nature of semiotics as a discipline, we would ask ourselves for instance how well beyond the easy alternative of Saus- similar to other pictures it is, and, more sure versus Peirce. This will be accom- in particular to pictures within the same plished by using both Peirce and Saussure picture category, such as all paintings, all as our stepping-stones, in order to reach a oil paintings, all pictures considered to be modern conception of semiotics, inspired works of art, all pictures in museums, all in perceptual psychology and cognitive pictures referring to the act of creating art, science. Finally, we will have a brief look, etc. We would be less interested than art to be pursued later on in the course, at the historians in knowing how similar it is to most important traditions within picto- other paintings by Velázquez and to other rial semiotics, which have developed from several different critical stances taken on  In this section, my aim is simply to offer Barthes’ early proposal, into a series of some approximate ideas about what kinds of ques- tions are addressed by pictorial semiotics. There conceptions more or less corresponding to are very few references here, and hardly any men- the ideal which we have set for semiotics tioning of names. We will return to theses issues at another level of complexity in later lectures.  Fig. 1b. Detail of Velázquez’ ”Las Meninas”

paintings from the 17th century, but even occupy as spectators, a man and a woman these questions may acquire a semiotical not very similar to us latter-day observers sense, as we shall see later. The answers are reproduced. The mirror image and the to these questions may throw light on what viewing position inscribed in the painting a picture is, as opposed to being a verbal thus seem to compete for the part of ob- sign, a gesture, and so on, and it may also servers. tell us something about the peculiarities of The paradoxes of Diego Velázquez’ painting, art, and other picture categories. painting “Las Meninas” are well-known, At the same time, it may also enlighten but largely unresolved (cf. Foucault 1966; us as to the particular way in which this Boudon 1979, Searle 1980). These para- particular picture makes use of the general doxes seem to have two sources: first of all, properties which pertain to it because of the fact that six of the persons depicted, the simple fact of being a member of the including the painter himself, stare hard category of pictures, the category of paint- at the observer; and, in the second place, ings, the category of a art, and so on. the specific position of the mirror image. The painter is inside the painting, The surface of paint which corresponds and he is looking at as intently. He seems to to the depicted scene at the same time ap- be in the process of painting our very like- pears to form a transparent wall facing on ness. But in a mirror in the background, di- to another space opposite to it in which the rected at the position which we are likely to spectators of the scene are placed. Indeed,

 Fig. 2. Picasso’s work ”Velázquez’ Las Meninas”

we can imagine another possible picture, ous in the present case. In fact, already as in which the spaces have been inverted, a perceptual artefact, the picture depends and the people we now see are looking at for it functioning on a set of environmen- whatever is at the other side of the pictorial tal spaces: the one in front of the picture, surface, which would normally include us. the one outside the margins of the pictorial The presence of the large surface, which surface, the one on the sides as well as be- we see from behind, may seems to indi- hind the final wall of the depicted scene, on cate that there is such a painting. And the so on. These are complex issues, but before very optical laws according to which mir- we return to them, we will have to attend to rors function suggest that the mirror image some more simple facts about the picture. in the background renders a portion of the We are actually looking at an ex- space which happens to coincide with the ceptionally large oil painting – or, more space of the observer. exactly, at a reproduction on paper (itself reproduced on the monitor) of such a can- The picture as a picture and as a vas. Both form flat surfaces (ignoring for sign the moment the more pregnant texture of But even before these paradoxes emerge, the oil painting), on which spots of differ- we must look upon the picture as a picture, ently coloured pigment have been irregu- that is, as a kind of sign, and we must take larly disposed. But we immediately see for granted that pictures can be perceived “through” this surface onto something else: and interpreted in certain peculiar ways. a scene taking place in a room illuminated Most obviously, of course, what has been by the light emanating from an aperture said about the transparent wall of the pic- in the background; the walls of which are ture facing the observer apply to all pic- almost completely covered with paintings; tures (as has been more famously pointed and which is occupied by nine persons (one out about the stage), but the gazes and the of which is standing in the opening in the mirror image make this fact more obvi- most distant wall) dressed in typical 17th

 century upper class clothing, to which are the sign “the painter” is really made up of added a dog, and a canvas seen from be- smaller signs such as “the moustache”, hind. The two persons to the far right ap- “the face”, etc.; and that the face is made pear to be dwarfs. Of the other three per- up of even smaller signs such as “the eyes”, sons in the foreground, one is a little girl, “the nose”, and so on. Indeed, this may who is surrounded by what appears to be sometimes be a rather adequate way of de- teenage girls, one kneeling and the other scribing the facts. However, if we look at curtseying. All the persons depicted have Picasso’s revised version of “Las Meninas” very serious looks, and most of them direct (Figure 2), it will be clear that this can- their gaze on us. The person in the close not be true as a general fact. We can still background to the far left holds a brush and identify some strokes and dots as stand- a palette, and an over-sized canvas takes up ing for different facial traits, but only by most of the left margin. Among the paint- referring them to the whole. It is only after ings in the background there is a mirror, having recognised some of the constella- which is illuminated in a peculiar way, in tions of surfaces as forming, for instance, which two other persons are to be seen (cf. the content “the female dwarf”, that we Figure 1b). Another luminous area in the can attribute the meanings “head”, “eyes”, background represents a door opening in “mouth”, etc., to some of the strokes. The which another person is standing. parts are thus meaningless before they are To see something as something oth- made to form wholes, just as in the case of er than what it is immediately is to see it verbal language; but unlike what happens as a sign. As we will see (in lecture two), in verbal language, the meaning of the this is far too broad a definition, but it will whole is redistributed back to the parts: in have to do for the moment. From the point a whole identified as a human body, some of view of our perception, certain constel- parts are made to signify “the head”, oth- lations of areas formed by pigment make ers “the trunk”, and so on. This, of course, up higher-order units which we identify does not happen in language. as such things as “a man in 17th century According to the psychologist James clothing”, “a small girl of the upper classes Gibson, only human beings are capable of from the 17th century”, “a female dwarf”, seeing lines on a surface as signs for some- etc. It is the combination of an expression, thing else. Indeed, apart from irregularly in this case some areas of paint disposed and arbitrarily disposed spots on surfaces, in particular ways, and a content, such as which we identify as dirt, and totally regu- for instance, “a man in 17th century cloth- lar shapes, which we identify as ornaments, ing”, which forms a sign. But the picture there is a peculiar phenomenon which is in is a peculiar kind of sign. As soon as we some ways intermediate between these two: have recognised the painter as such, we can irregularly disposed spots, which however identify his moustache, his eyes, his hair, have the property of corresponding in more his clothing, his brush, etc. The parts of the or less regular ways to things which might content “the painter” are distributed onto be perceived in the common sense world of certain parts of the expression. our immediate experience, the Lifeworld. We might even be tempted to say that  It should perhaps be pointed out that the  Fig. 1.a. Detail of Velázquez’ ”Las Meni- nas”

In this sense, Picasso’s “Las Meninas” is also of some other hominids, at least when as much a picture as is the original one by they have received some training. Velázquez. On the other hand, there can Indeed, our perception of pictures is be no doubt that this definition will cause twofold in a double sense: it is not only that some problems for those how would like we see through the surface to the depicted to attribute the status of pictures to some scene, while retaining consciousness of works by Pollock, Klein, and Vasarely. this mediating space. We are also often Later on, we shall see what we can do to aware of the properties of the surface itself. help them out. Is has been said by one of the eminent pro- Gibson cannot of course have meant ponents of pictorial semiotics, Jean-Marie to claim that other animals are unable to Floch, that it is possible to consider any or- identify what the picture represents. Even dinary picture as if it were a non-figurative doves have been known to react to pho- painting. In Velázquez’ “Las Meninas” we tographs of human beings exactly as they can hardly avoid noting the preponderance react to the humans in person. But that is of extended, dark, relatively homogeneous, exactly the point: they do not perceive the rectangular, fields and straight, almost par- picture as a picture – as a sign. A picture allel, lines in the upper part of the surface, of a human being is to them just another as well at the extreme right and left mar- instance of the category of human beings. gins, in opposition to the small areas of dif- So what they lack is the capacity for double ferently coloured, mostly rounded shapes, perception characteristic of human beings in the central, lower part of the surface. It – and perhaps, in spite of Gibson’s claim, can immediately be observed that Picasso’s version preserves little or nothing of this idea of construing pictures as being in some way intermediate between dirt and ornaments is my “plastic” layer of the picture. In fact, Picas- own, though Gibson records these three categories sos “Las Meninas” seems to convey very of arrangements on surfaces.  different meanings from those which can place the infanta and the menina to the left be grasped in the original. This makes us in the 17th century because of the configu- suspect that, in some sense, the plastic lay- ration of their dresses. The different per- er may also contain its own series of signs, sons in Picasso’s work are deformed and with their own expression and content. simplified to different degrees: in the case of the two right-most figures, nothing more The act of recognition and the than a general sense of humanity seems to categories of reality be retained. But although a rather classical For the moment, however, it will be bet- Cubist transformation has been applied to ter to delve deeper into the nature of the the Velázquez figure, it seems to conserve signs of depiction which make up the pic- some more personal traits: the moustache ture. In “Las Meninas”, we can identify and perhaps also some of the facial traits more than just general categories such as are recognisable. “man” or “woman”, “child” or “adult”, There is of course an extensive art “dwarf” or “person of normal length”, etc. historical literature about Picasso’s paint- We can identify individuals. The painter ing, as there is about Velázquez’ work. is Velázquez himself. The little girl is the Art historians have been known to claim, Spanish infanta Margarita. The couple in for instance, that in Picasso’s version, the the mirror is Philip IV and Queen Mari- man standing in the door opening in the ana. The other persons are members of background represents Picasso’s own sec- the Spanish court which also can be given retary. In fact, of course, the schematic their name tags. Surrounding the infanta, figure created by Picasso hardly retains of course, we discover the two meninas, or any property permitting us to identify it as maids of honour, whose function explains anything more than a human being with a the title nowadays given to the painting Spanish cloak leaning on the door post. It (cf. Figure 1a). Velázquez, the king and his is of course quite possible that Picasso was court have (or, more precisely, had) an ex- thinking about his secretary while paint- istence outside of “Las Meninas”, and out- ing this figure. But there is nothing in the side of the world of pictures in general. The painting which conveys this fact to us. So have formed part of the privileged sphere although the secretary has (or had) an ex- we call reality. They are not only contents istence in the real world, he not only does but also referents. not form part of the content of the painting, The case of Picasso’s paraphrase but he is no referent either. On the other is somewhat different. We may recognise hand, if it is true that Picasso changed the the persons also in this picture, but only breed of the dog appearing in the Velázquez because we can compare it to Velázquez’ painting into a dachshund, because he was original. Considered independently of himself the owner of one, then of course Velázquez’ work, the figures appearing in it makes sense to say that not only has the Picasso’s painting can hardly be identified content been changed (on the general level as anything more than, in the most favour- on which “dachshund” is opposed to “Alsa- able cases, men and women, perhaps also tian”), but also the referent. adults and children. Perhaps we could still

 Fig. 3. Hamilton’s work ”Picassos’ Las Meninas”

The painter caught in the act their portraits. The reference is really only made up of relationships between pictures. It is often said that, contrary to language, And at some point there is a label: “this is pictures may only show individuals. We Velázquez”. Indeed, this could just as well have seen that this in untrue, at least if pic- be a completely fictional world, if there tures like those created by Picasso are taken were not so many other historical threads into account (and traffic signs, pictograms, (not having to do with their looks) pointing and so on). But perhaps we have exaggerat- to the real-world existence of Velázquez ed the difference between the paintings by and the king who employed him. So we can Velázquez and Picasso. After all, we only be fairly sure they are real-world persons. know who these individuals are because the But we cannot know much about their real- identifications were made by their contem- world looks. So this gives us one of the rea- poraries and have been conveyed to us by sons for retaining the distinction between books. There are of course also other self- the content of the picture and its referent. portraits by Velázquez, and Velázquez and others even painted other pictures showing And this is also the sense in which some of the other individuals. But contrary we can identify a genuine Velázquez sign, to (some of) their contemporaries, we have as well as a Philip IV sign, and so on, but never seen Velázquez, Philip IV, and the not a “secretary-of-Picasso” sign. The others in reality. We only know them from content of the former sign is sufficiently 10 characterised to be recognisable in other relationships between objects which are pictures. The latter one is not. On anoth- present in an equivalent perceptual situa- er level of generality, we can identify a tion, even though it might be seen as em- dachshund sign, because there are other phasising some of them. This is not true in dachshund pictures. We cannot identify a the full sense of Picasso’s variant. Nor is it “Picasso’s dachshund” sign, even though true of traffic signs, pictograms and other it seems probably the change of breed was “schematic” pictures. And it is not true of motivated by Picasso’s having a dachshund typical pictures of saints and archangels, closer at hand. On the other hand, in Ham- for even though the person and his or her ilton’s paraphrase of Picasso’s “Las Meni- attributes may be rendered descriptively, nas” (Figure 3), we can identify a “Picas- the relationship between them is not. so’s bull” sign, not because we have any In the second place, Velázquez’ “Las reason to suspect Picasso had such a bull Meninas” shows one isolated phase in the at home, but because it appears in many of sequence of the actions termed “painting a Picasso’s paintings. canvas”, or, more precisely, in the sub-se- Even if we would have been unable quence which could be described as “look- to identify Velázquez, there is of course no ing at the motive while simultaneously problem in delimiting a painter sign. We picking out one of several shades of colour see Velázquez in the process of painting. /in order to adjust the pigment on the can- In fact, it is probably more exact to say that vas/”. In fact, this particular phase may eas- we see him as he steps back to look at his ily be redescribed at other levels of abstrac- picture (and perhaps to compare it with its tion, both higher up and lower down. And, subject, the persons in front of the picture of course, the picture renders at the same we see). But we do not even need to ob- time phases of several other sequences of serve the act of painting. The brush and the actions: we can catch the left-hand maid of palette may be described as stereotypical honour in the process of kneeling down, as attributes of the function “painter”, much well as holding her hand, and even her fin- as the archangel Michael is identified by gers, in a particular position. a flaming sword, the evangelist John by a As for what has happened before this goblet, and so on. But these attributes are constellation of phases of different action not placed in an indeterminate relationship sequences, and what is going to happen be- to the body of the painter (as they might be fore, we can only know that to the extent said to be in Picasso’s variant), nor are the that we are familiar with the type of action outside of time: Velázquez has been caught going on: painting, kneeling down, curt- at a particular moment in the process of seying, etc. In Picasso’s variant, however, touching the palette with his brush, while the options for redescribing the actions at examining his motive, and deciding where different levels of abstraction, as well as and how to apply to paint to the canvas. the possibilities of anticipating and project- In this double sense, the original ing other phases of the actions backwards “Las Meninas” is a prototypical picture: in time, are much more limited, and dif- first, it covers the spatial scene densely, ferently so for different figures. Of all the taking into account more or less all those figures, the leftmostinfanta is probably the

11 only one in Picasso’s picture which offers coloured shapes on the surface as persons a semantic richness coming close to that of and things, we seem to go beyond what is Velázquez’ personages, in the sense pres- immediately given. And yet, it is almost ently addressed. impossible not to do so. We do not need All this no doubt has something to do any “key” to see what is depicted, at least with the famous paradoxes of Velázquez’ not at the highest levels of abstraction: “a “Las Meninas”: they will only occur to the man”, “a girl”, “a dog”, etc. No doubt we extent that we are able to identify part of need some experience of the world. Some what is seen in the picture as a representa- rudimentary historical knowledge is also tion of the painter looking at his motive. useful, in order to see the figures as 17th And for their existence, these paradoxes century personages. But we do not need are also dependant on our identifying any specific “key”, such as may be required the painter depicted with the one who is for some “picture riddles” (or “droodles”, as known to have painted the canvas we are they have been called), such as the square looking at. In other words, both the prop- with a small circle and a triangle, which erty of being depicted on this particular could be taken to show an olive dropping canvas in the act of painting (which is of down into a martini glass, or the navel of course a likeness, that is, an iconical sign), a girl in a bikini seen through an opening. and the properties of having produced the In fact, even in Picasso’s “Las Meninas”, painting by applying paint to this particu- there is a lot to see without having recourse lar canvas (which is a sign dependant on to any such specific instructions. To go fur- a relationship of contiguity, that is, an in- ther (to less abstract levels), however, we dexical sign), are properties ascribed to have to use the original “Las Meninas” as the peculiar persona we call Velázquez: a key. that is, they are part of our construal of A second sense in which the pic- the referent. Differently put, these are two ture immediately seems to give access to things we are immediately aware of about more than it literally contains was consid- the referent Velázquez, as he appears in ered in the last section: our anticipation, the picture while knowing all the time that for instance, of the painter’s putting the there is a lot more to know (some of which brush onto the palette, than to detach it we will never get to know). from it, and applying the brush to the can- vas – which goes together with our paral- The picture as part of the world lel ability to project into the past, at least There are thus several senses in which our vaguely, the actions preceding the scene in perception of the picture immediately goes the picture: the first touches of the brush beyond that which is directly present to us applied to the canvas, even the raising of on the canvas. Strictly speaking, of course, the canvas inside the room, etc. We will there are only colour spots on the canvas: certainly need some not so very specific already when perceiving the differently knowledge of the world in order to identify the type of action going on. So the scene  Or, rather, in our case, what we see is the reproduction of this canvas, that is an iconical sign which we perceive directly is placed within of an indexical sign – on which more is to follow the framework of a temporal sequences of later in this lecture. 12 typical happenings in the world with which opening and the window. But contiguities we are familiar. as well as relationships between parts and There are yet other senses in which wholes are very important in other ways to the picture can be seen as a part of the real the interpretation of Picasso’s picture. In the world given to our perception. Already as case of some of the figures, we even have a surface the picture forms part of a greater to put body parts in rather different styles whole: the lines extended from the objects which appears more or less independent of close to the edges of the picture are felt to each other together again in order to iden- continue on outside of the pictorial surface. tify the personages. This is perhaps more More obviously, this is true of the scene obviously true in the case of the painter fig- which we observe in the picture: the canvas ure, which in a typical Cubist manner (of erected at the left-hand limit of the scene, which Hamilton gives other examples) is for instance, is not perceived to be cut off given in separate perceptual perspectives. where the depicted scene ends. If so, we What particularly seems to be lacking in could hardly have identified it as a can- Picasso’s version, however, is the fron- vas. Nor do we doubt the presence within tal space: the gazes of the personages are the same fictional space of those pictures either directed elsewhere or are too inde- hanging on the wall in the background terminate to suggest any presence in front which are partly covered by the canvas of the picture. And the mirror can only be or the bodies of the depicted persons. We read as a mirror by using the original “Las also take for granted that the door in the Meninas” as a key. background leads on to further rooms, and that the window opens up onto the outside The depictions in the picture world. What is more, we never doubt that and the picture depictions fictional space continuesin front of the pic- Velázquez’ “Las Meninas” is related in two torial scene, beyond the limit of the depict- rather obvious ways to other pictures: it de- ed space. Most probably, there are other picts some of them, and it has itself been walls there, other doors and windows, and the source of a number of later pictures. perhaps other paintings on the walls. Other The pictures on the wall, just as the per- persons may also be present in that room, sons, can be identified – or, at least two of as suggested by the fact of most gazes be- them can. They are works by Rubens and ing turned in that direction. We know that Jordaens, both treating themes from Ovid’s what most tends to catch the attention of “Metamorphoses” – or, more exactly, as art human beings is the presence of other hu- historians are wont to tell us, they are cop- man beings, and, in the second and third ies after those works made by Martínez place, other animals or other things which de Mazo (though this is, as far is I under- move. In this particular case, there is also stand, not a piece of information which can the presence of what we tend to identify as be gained from scrutinising “Las Meni- a mirror image which confirms this inter- nas” itself but merely something which can pretation. be extrapolated from your knowledge of Not so much is left of this is Picasso’s the royal inventories). As part of the sur- version of the scene, apart from the door face of “Las Meninas”, these pictures are

13 thus pictures of pictures of pictures – but tuted by Velázquez’ original work. But it to the observer, they are indistinguishable would be curious to maintain that this is from pictures of pictures. The interesting true in the same sense in which the word thing about them, however, is that they are “Velázquez”, in its written form, contains representations of specific pictures. There two tokens of the letter type “z”. clearly are other pictures seen to be hang- There is actually two differences ing on the wall of the room depicted, but here. It seems reasonable to say (at least they cannot be identified. They are thus if you are not a Platonist), that previous to only generic picture representations: they the creation of the first token of Velázquez’ merely represent the category “picture”. “Las Meninas”, there was no type corre- The crucial factor here is not whether we sponding to “Las Meninas”. So there was can see what is represented. It is whether a creation, at a particular time and place, was is seen can be projected onto some of a first token, which constituted the type, original picture known from other sources. which then could give rise to further tokens. The villas of Roman antiquity are full of It makes no sense to say the same thing generic picture representations, painted in about the letter “z” or the corresponding trompe l’œil on the walls. The motives can sound. Now doubt the letter was invented often be clearly distinguished, but we, at at some point in time, and the sound may least, are unable to refer them back to some has metamorphosed from some differ- originals by Appelles and Zeuxis. ent sound at some moment: but, as part of “Las Meninas” is also at the origin of some language (or some name system), it is many other pictures. Most obviously, to us, part of the resources always available to the it is at the origin of many reproductions, one user of a particular language – as long as it of which is found as an illustration to this is (the same stage of) the same language. lecture, and others which appear in picture But there is of course also another books about Velázquez’ work, or are sold difference: by putting together a lot of as posters in museums and art shops. In a letters and words (just as I am doing just rather similar way, it is probably also at the now), one creates a type of a text, which origin of some other painted pictures. Per- then may be reproduced in numerous haps Martínez de Mazo made a copy of it books – or downloaded to a number of also, though I have never read about it. But web-browsers. So language does also lend surely, since Velázquez’ time, there have itself to the creation of types originating in been art students visiting the Prado and a first token. Traditionally, however, we do trying to imitate the master’s work as best not tend to think much about the first token they could. These copies no doubt differed of a written or printed text (unless it is one more from Velázquez’ original work than of the incunabula). Pictures are different. our photographic reproductions (though of Or at least, pictures which we call art are course they were more similar to it in some different. The case is less clear in other in- respects, such as texture). It seems proper stances. In part, at least, this has something to say, then, that those painted works, just to do with the kind of channels in which as the reproductions, are in some sense ex- pictures circulated, and the social purposes amples, or rather tokens, of a type consti- which art attributed to them. Some such 14 sets of channels, in combination with a world number of socially anticipated intentions, Possibly the most interesting picture de- together with some peculiar ways of pro- picted in “Las Meninas” is the one we can- ducing pictures, have, under historically not see, or more precisely, which we can changing circumstances, been identified only see from behind. What is the canvas with the sphere of art. showing to the persons at the other side of Then there are of course those pic- the pictorial surface? Perhaps it is mere- tures which have been engendered by “Las ly an empty canvas. But the gaze of the Meninas” in a more convoluted way. Picas- painter suggests that it is at least partial- so’s “Las Meninas” is not just a token of ly a representation of the frontal space in Velázquez’ “Las Meninas”. Nor is Hamil- which we are also located. Some commen- ton’s “Picasso’s Las Meninas” really a to- tators, however, have suggested that what ken of Picasso’s work. One would rather Velázquez here is in the process of paint- think that each refers to the earlier work ing is precisely “Las Meninas” (cf. Boudon while adding some commentary to it. In 1979; Searle 1980). In many ways this is fact, as we shall see in a later lecture, Ham- a strange proposition: it certainly serves ilton’s work actually refers to, and com- to augment the paradoxical character of ment upon, both Velázquez’ and Picasso’s the representation. One of the reasons for “Las Meninas”. One may wonder, then, in this suggestion (formulated by Searle) is what way these works are different from that “Las Meninas” is the only painting by those of the art student trying to imitate the Velázquez showing the royal couple – even masterwork in the Prado – and those of a if only in the mirror. If so, the couple only imagined Martínez de Mazo churning out make up a small detail of the scene repre- copies of Velázquez’ painting for all the sented in “Las Meninas”, the mirror image. members of the court. In part, no doubt, the Interestingly, however, Searle’s argument status of the former works are different be- is based on relationships between pictures. cause they differ from the original in much The picture does not only represent more obvious and systematic ways. And, a fictive scene. It also organises the space in part, they are different, because Picasso around itself. In fact, the picture is not just and, to a lesser extent no doubt, Hamilton a flat surface: it is an object with three are important figures in the world of art. dimension and six sides. But only one of It has been said that, by depicting these sides attracts our attention and/or is other pictures, Velázquez emphasises the traditionally considered to be the front side sign character of his work (Lotman 1990: of the picture thing: the surface. The other 54ff). If so, by depicting Velázquez’ work, sides of the object we call a picture do not Picasso and Hamilton continue strengthen- signify in the same way – not in the way of ing the sign character of the picture, even a picture. We can to some extent see two of though the first picture depictions, of the these others sides of the picture depicted in Martínez de Mazo copies, become less our picture, and if we were in front of “Las prominent in the latter works. Meninas” in the Prado, we would be able to see all but the back side. In the reproduc- The picture as a thing in the tion accompanying our text, on the other 15 hand, the other sides are hardly visible to picts from behind, it is easier to see that is the human eye. This observation applies to occupies a particular position in the world the lateral side of the picture, as reproduced (which is only vaguely inherited by the in a book, for in most books nowadays, the book reproduction and the web picture). It back side of the picture would be identical places the observer in front of the frontal to the front side of another picture. On the side. In a picture using linear perspective, Internet, as in computer space generally, as “Las Meninas”, the inscribed position the idea of the back of the picture does not of the observer may be more exactly de- have any interpretation at all. termined. There is only one place of ob- It could be said that the museum, the servation from which the depicted scene book, and the web are different channels is rendered in a geometrically correct way. of circulation for the picture: the ways in This position could be marked out on the which they are brought from the “sender” ground in front of the picture. In this sense, (creator, patrocinator, or other kind of orig- the picture object contains its implicit ex- inator) to the “receiver” (the one perceiving hortation: this is the place from which I and/or interpreting it). Often enough, but like to be observed. not necessarily, different channels of cir- This is not to say that the picture can culations are connected to different man- only be seen from this position. Our per- ners of construction. In this case, it seems ceptual system has the ability to recognise that the surface is becoming ever more in- objects from all but the most extremely di- dependent of the thing of which it consti- vergent points of view (such as anamorphic tutes one of the sides. In order to circulate perspective). In fact, we do not observe the through museums and galleries, Velázquez’ perceptual world in the way presupposed work may remain an oil painting. If it is to by linear perspective: with a single eye im- circulate more widely, to homes and public mobilised in a fixed position. Therefore, it places, it would stand more chances being is no more difficult to interpret the scene transformed into print. And if it is to circu- of the book reproduction or the web pic- late on the Internet, is has to be made into ture than that of the original canvas. Lin- pixels. Manners of construction and chan- ear perspective is simply an additional sign nels of circulations are also in principle in- impressed on the pictorial surface. dependent of socially anticipated functions In this sense, “Las Meninas” is just or effects. In the present case, the picture like any other picture. What is peculiar we are considering was meant to be a work about Velázquez’ picture is that most of of art (which in itself was not the same so- the persons depicted, including the painter, cial practice as it is today), but we are using seem to be looking in the direction of the it here mainly as an example of any case of point in front of the canvas coinciding with depiction. It is put to a pedagogical use. the position of the inscribed observer. The fact that the painter is looking in that direc- The inscribed positions of the tion is especially remarkable, for there is observer and of the painter every reason to think that what the painter If we think of “Las Meninas” as a solid is looking at is also what he is in the proc- object, similar to the canvas which it de- ess of depicting. 16 Not only is the position of the ob- rooms server inscribed in the picture; the posi- Among the paintings on the wall in the tion of the painter also appears to be so. background, there is a square which has al- According to Searle, at least, we look upon ways been taken to be a mirror. The reason the depicted scene as something the paint- for this is no doubt its higher degree of re- er himself has observed from the particu- flexivity and illumination, together with a lar position indicated by the perspective. greater fuzziness, which distinguishes this There is nothing paradoxical about that: square from those which represent paint- it simply means that what the painter sees ings. If so, the scene rendered is a direct is also what the observer is meant to see. imprint of real-world objects (in this case, Nor has the problem anything to do with persons) present in the world imagined to the simple fact that the painter depicts him- continue in front of the pictorial surface. self in the process of painting: in Courbet’s From this point of view, the mirror is like view of his own workshop, there is nothing the photograph, or like the tracks left by strange, because he depicts himself as any an animal on the ground – except that the other painter might have done. In this par- mirror only retains the imprints as long as ticular case, however, the relationship be- the objects continue to be present in front tween the inscribed positions of the painter of its surface. Since we are unable to see and the observer together give rise to a par- the depicted couple directly, however, the adox: the depicted painter is looking at the mirror is here the only sign adverting us position where we should be. It thus seems to its presence on this scene. Unlike what that he is depicting the very observers of is often said to define the sign, the mirror his painting. He is depicting us. thus requires the presence of its object. But The paradox is the result of one (his- it allows it to be present elsewhere – not torically very important) way of construct- where we are looking. Indeed, the royal ing pictures. In would not arise in many couple is absent from the scene which we mediaeval pictures, nor in Russian icons of are observing. The mirror is not necessar- a later age, because at least the central parts ily redundant, as in the case when some- of these pictures have been conceived as if body is scrutinising his own looks. As the observed (by the painter, directly, and per- rear mirror of the car, it here allows us to haps by God, metaphorically) from the back discover something new. side of the painting (cf. Uspenskij 1976a, b, In this case, this gives rise to another c). The painter is not depicted. Yet his posi- paradox: the scene reflected by the mirror tion is marked in the picture. As observers, is identical to the inscribed position of the nevertheless, we are incapable of taking up observer. We are the couple in the mirror. the same position as the implied painter. We are the royal couple. But of course we We are excluded from the dialogue. The do not quite believe the sign. There is a gap possibility of identifying with the percep- between the sign and the reality. tual perspective of the painter is only given Like the door opening, the mirror is a in a particular type of picture. way of opening up the depicted space onto

Reflections of (and on) other  This is of course not Eco’s view, but we will see in lecture three why he is wrong. 17 the environment; it adds a dimension, Lot- of the royal family, and the main members man (1990: 56), maintains, much like the of it are only present in the mirror. sound in the cinema coming from some- None of the titles actually seems to thing which is not visible in the scene. In account for what we seen in the picture. another picture by Velázquez, “Venus They differ in singling out as important at her mirror”, the mirror image actually different portions of the depicted scene. shows us something which is in the scene, But none of them are very convincing in the face of Venus, although it is invisible that respect. The construction of the scene from the point of view which the picture may be more important. The geometrical adopts on the scene, which only allows us mid-point is found exactly between the to observe Venus in profile. In “Las Meni- eyes of the infanta. But the middle of that nas”, however, what the mirror shows is not semi-circle which is created by the painter, part of the scene, even from another per- at one extreme, and the right-most courtier, spective. It displays another part of space. at the other, is occupied by the mirror, in The effect of the picture is to crown which the royal couple is to be seen (cf. the observers into king and queen. Since Foucault 1966). Both the two imaginary the painting was made for the royal couple, mid-points of the scene, of which one is this operation has not always been unreal- somewhat displaced in relation to the other, istic. To Philip IV and Queen Mariana, the thus correspond to depictions of members painting was really a mirror (cf. Boudon of the royal family. This suggests that the 1979). We, however, are only participating royal family really forms the central theme in the long row of visitor to the Prado, who of the picture. have been king and queen for a few min- But it could also be proposed that the utes. real theme of the picture is not in the scene depicted, but in its relation to the potential The title and the theme – and the observers. Not only does the scene include act of depiction a depiction of the painter himself in the act Some people maintain that pictures, like of painting, but its very organisation sug- perceptual reality, only gain a meaning gests that what is depicted on the canvas once we assign to them a verbal label. forming part of the scene is the observer We know this picture under the title “Las himself. Meninas”. It tells us rather little about what In a famous passage, Michel Foucault is seen in the picture. It refers to the two (1966: 31) suggests that “Las Meninas” is a young women surrounding the infanta, representation of what he calls “classical the maids of honour. In fact, this title was representation”, in which the sign appears th only invented in the 19 century. There is to be separated, not only from its model, no known contemporary title, but some but from the subject which is at liberty to early inventories describe it as showing the create it as well as to perceive it. This is painter painting. Later on it was known as a curious affirmation, for, if anything, our “The family of Philip IV”. This title also discussion of the painting has shown it to seems curious: several persons, including point very clearly to the interrelation be- Velázquez and the dwarfs, are not members tween the picture and that which is depict- 18 A Saussurean look at “Las Meninas”: The sign Many people think that semiotics consists in some kind of development, or applica- tion, of things said by the linguist Ferdi- nand de Saussure, as well as by the phi- losopher Charles Sanders Peirce, possibly to objects not considered by these thinkers. This is a strange idea, already because the domains staked out by Saussure and Peirce cover very little, if any, common ground. Nevertheless, we shall just for a short while entertain the fiction of looking at “Las Meninas” from the point of view of Saus- Fig. 1c. Detail of Velázquez’ ”Las Meni- sure and Peirce. As far as I know, there is only one ed, and to the relations of the latter to the passage (not included in the Cours), in world around it; and it has shown the posi- which Ferdinand de Saussure [1871-1913] tion of the observer to be prescribed by the mentions the picture as a possible object of very object making up the picture. semiotics (or “semiology”, as he preferred According to John Searle (1980), to call it). In fact, the picture is also present “Las Meninas” is about two things, one in another passage, by implication. To of which is invisible (the front of the de- Saussure, semiotics is only concerned with picted canvas, supposedly identical to “Las conventional signs. But he tells us about Meninas” itself), and the other one being the mime that it is sufficient for it to have outside the picture (the imaginary position a rudiment of conventionality for it to be- of the painter). Even if these elements are come a subject matter for semiotics. Such important, they could hardly be so, other- a rudiment of conventionality is certainly wise than as part of a more general strat- found also in pictures. The explicit remark egy: they single out the act of depiction by Saussure about pictures concerns their itself. Jurij Lotman (1990.55f) maintains dimensionality: he observers that they that, because it depicts other depictions, have several dimensions, contrary to lan- and even the painter in the process of creat- guage which only has one, that of time (or ing a picture, and because it lets the mirror of space on the written page). This is, as reflect the subject matter of the painting, I will suggest below, a passage in which “Las Meninas” transforms the relationship Saussure forgets his own essential lesson: of the picture to the depicted object and the that the important thing is not what is ma- nature of visual thinking into the theme terially there, but what is relevant. of the painting. We have seen that there is The basic concept of Saussure is the some truth to this. But to a large extent, sign, understood as the union of expression something similar could have been demon- and content, where the latter is distinct from strated in the case of any picture. the object of reference. In verbal language, 19 the expression is (an idea about) a sound, a small bundle of traits which are present and the content is a concept, or (an idea also in Picasso’s version, and even, given about) persons, things, and other phenom- some detective work, in Hamilton’s ver- ena of the world. Analogously, in a picture, sion. If the content form is “The Spanish the expression is (an idea about) colour infanta Margarita who was five years old in spots disposed in a particular way on a sur- 1656”, however, much more details are re- face, whereas the content is (an idea about) quired, which are only found in Velázquez’ a visually perceptible situation. What is version (and even so, of course, only there common to the different “Las Meninas” by for those having some background knowl- Velázquez, Picasso, and Hamilton is (on a edge; cf. Figure 1c). At the other extreme, very high level of abstraction) the content: the content form for “human being” re- a room with nine persons, a dog, a canvas, quires a very small amount of expression some window openings, a door opening, a features: in some cases also Picasso, and mirror image, etc. The three correspond- even Hamilton, manage to convey this con- ing expressions have much less in com- tent with very limited means. mon, apart from consisting of (the idea of) It will be noted that these different paint (no doubt made out of rather different meanings are not placed after each other: materials), or, more generally (to include rather, they are incrusted into each other. Hamilton), marks on a surface. Not even The bundle of features necessary to con- the dimensions are the same. vey the content “human being” is part of The content can be further divided the content representing “a small girl with into smaller parts, but not in the same way long hair adorned with a bow wearing a as language, where you add one word to crinoline and having a stiff corporeal at- another. Contents such as “a man with 17th titude”, which itself is part of the content century clothing”, “a young woman with a “The Spanish infanta Margarita who was crinoline”, “a small upper class girl from the five years old in 1656”. This, of course, is 17th century”, “a female dwarf”, “an Alsa- an important way in which pictures (and tian”, etc., can readily by conveyed without perceptual reality) are different from lan- all the details included in Velázquez’ ren- guage. dering. Many of them are still identifiable in Picasso’s version. Those parts of the ex- A Saussurean look at “Las pression which cannot be exchanged with- Meninas”: The system out giving rise to another content, as well According to the Saussurean view, the as those parts of the content which cannot sign, as a whole, made up of expression be substituted for others, without another and content, stands for reality, and there expression being the result, are called form is an arbitrary or conventional relation- by Saussure: other traits, which may vary ship both between expression and content, freely, are called substance. If the content and between the sign as a whole and real- form is “a small girl with long hair adorned ity (that is, the referent). In the case of lan- with a bow wearing a crinoline and hav- guage, this means, in the first case, that the ing a stiff corporeal attitude”, then the cor- sound making up the expression is usually responding expression form only requires not similar to the content it designates; and,

20 in the second place, that the way language In the case of pictures, the first re- carves up reality is not (entirely) given by quirement is only slightly problematic, (perceptual) reality itself. The picture sign but the second is much more so. It is of- is obviously different in these respects, but, ten maintained that, contrary to language, like the mime, it certainly retains a rudi- pictures are unique, but this is misleading. ment of arbitrarity. The expression of the The assignment of particular value to the content “Velázquez” is no doubt to some first instance of a picture is a norm prevail- degree similar to reality (or to other pic- ing in the sphere of art, but there, too, even tures containing the same sign); but it is before the advent of mechanical reproduc- also flat, and made up of paint. And while tion, copies of masterworks have always the sign “Velázquez” in Velázquez’ own been made. They were of course slightly painting, and that by Picasso, must be sup- different from the original, more so than posed to correspond to the same figure of the photographic reproductions of more re- the real world, that is, the same referent, cent ages; but every realisation of a standard their contents are quite different; indeed, unit is somewhat different from the others. in this case it is quite clear that they ana- The problem is only whether we consider lyse the referent in quite different fashions, this difference to be part of the form or the Velázquez exclusively in terms of proper substance; in other words, whether it is a parts, but Picasso also in terms of percep- difference which makes a difference. In the tual perspectives. sphere of art, it is. But as far as other uses Another important distinction made of pictures are concerned, there is no rea- by Saussure opposes the system to its dif- son for it to be relevant. ferent realisations. The system is made up Discussing pictures as we just have of a series of units and the rules for their is like talking about language at the level combination. In language, the units are the of texts. But it is not impossible to gener- sounds (corresponding more or less to let- alise, to some extent, the argument to parts ters), which are combined into words ac- of pictures. There is an infanta sign which cording to certain rules, and, perhaps more does not only recur from one reproduction obviously, the words, which are put togeth- of “Las Meninas” to the others, but also, er to form sentences following particular given some suitable principle of relevance, rules. This implies two things: there is a to other pictures by Velázquez and Mar- repertory of standard units, which may be tínez de Mazo, and, with some even wider repeated over and over again, at different definition, to “Las Meninas” by Picasso moments in time and space, still remaining and Hamilton. The trouble is of course that the same units; and there are restrictions there is not just one fixed principle of rel- on the way which these units may be put evance, and there are even possible units together to form higher-level units. In the which overlap with the infanta sign, such first case, we are concerned with the dif- as the “little girl sign” (another level of ab- ference between the way in which a word straction) and the crinoline sign (another appears in a dictionary, and the way it ap- part of the whole). pears in a text. In the second case, an ex- The reason pictures are different is ample would be the rules of grammar. that they do no normally form repertories.

21 This is not true of all visual signs: there are to the left and above the infanta sign rather well-established gesture languages, from than the reverse: there is not even any rules sign system used by the North American stipulating that the Velázquez sign goes Indians in inter-tribe communication (as together with the infanta sign, whereas analysed by Garrick Mallery) to the dif- the Picasso sign can be exchanged for the ferent varieties of Sign languages used by Velázquez sign (as it is in Hamilton’s pic- deaf-mutes today. More in particular, there ture). Thus, even if we consider syntagm are repertories of pictures used for particu- and paradigm apart from their deployment lar purposes, such as traffic signs, the pic- in time and space, as conjunction and sub- tograms used in public spaces, the pictures stitution, respectively, they do not apply to on packs of cards, the ideograms which in pictures. some historical contexts were at the origin Or, rather, they only apply to pic- of the alphabet, etc. Some such systems tures in special cases. Given, Velázquez’ may be analysed into smaller units which “Las Meninas”, Picasso’s version may be are then put together to form complete seen as a different syntagmatic variation, signs, as most certainly is the case with using (at some rather high level of abstrac- the figures used on clothing as washing tion) the same units; and Hamilton’s ver- instructions, but also perhaps with prehis- sion would rather appear as a case of filling toric petroglyphs. Most pictures, however, the same syntagm with different paradig- seem to exist on their own, without form- matic units. But, first of all, this only has ing part of a series; or they become part of a sense when relating these pictures to the the series only post hoc, as in the case of first version of “Las Meninas” – and most the different “Las Meninas”. of the time, there is no such primary ver- And this is why one of the most fa- sion. And, in the second place, there is no mous Saussurean distinctions is of little use limited number of variations which are al- when talking about pictures, that between lowed by the syntagmatic rules, nor any syntagm and paradigm. The syntagm is finite number of paradigmatic units capa- the order in which units are combined. The ble of occupying the different positions in paradigm is the set of units able to occupy the syntagm. We are here concerned with the same position in the syntagm. As Saus- a very much derived meaning of meaning, sure much insisted, language is linear, and which is better understood at the level of thus units can only be put together one af- rhetoric: as producing meaning by deviat- ter the other, in time in spoken language, ing from that which is expected. and in space in writing. This is the context Syntagms and paradigms also have in which Saussure argued that pictures, in other uses in pictures. There are no doubt contrast, are multidimensional: their units some particular pictorial genres in which can be combined on multiple axes. Yet I there is no absolute freedom of placing all think Saussure was wrong. Not only is kinds of things together and in any order. there no rules for pictures stipulating that Thus, for instance, Russian icons allow the Velázquez sign must appear somewhat only saintly persons to be placed in the  The latter term was really coined by Hjelmslev, but the general idea is present in the  An important domain of pictorial semiot- work of Saussure. ics, which we will discuss in a later lecture. 22 middle of the picture, and ordinary people appear either before the picture, in what at the margin (cf. Uspenskij 1976a, b, c). is depicted, or after the picture, in the re- Conceivably, other pictorial genres have lationship between sets of already exist- rules for the concomitant presence and or- ing pictures. They are either referential or der of combination of colours and shapes. intertextual. It does not make much sense In the second place, paradigms may be to talk about a picture system. Rather, present in much of what is depicted. For there are resources at hand which permit instance, when Picasso places a dachshund the creation of meaning by means of pic- where Velázquez painted an Alsatian, he tures. These are different in many respect has recourse to what may very well be from the resources at hand – units and their termed a real-world paradigm (a classifica- combinations – allowing for the creation of tion system) of dogs; and when Hamilton verbal signs. introduces a bull instead of a dog, he refers to the over-arching classification system Peircean perspectives on “Las of animals. Another system having both Meninas” syntagms and paradigms depicted, most Contrary to Saussure, Charles Sanders obviously in Velázquez’ “Las Meninas”, is Peirce [1839-1914] did not use language as clothing: there are rules (somewhat differ- his model. However, he was not much in- ent today than in the 17th century) for what terested in pictures either. The advantage of pieces of clothing may be combined, and the Peircean model is also to a large extent for what units may be substituted for one its disadvantage: it is so general it fits in another at a particular place. Indeed, the with everything. It will not tell us anything clothing system really is what Saussure er- about the specificity of the picture sign. roneously claimed about the picture: mul- In fact, I am not sure it tells as any- ti-dimensional. There are rules for which thing about the specificity of the sign, in pieces of clothing may go together both on the sense in which the Saussurean sign different body parts and at different lay- is one, or in the sense in which we have ers in relation to the bodily surface. In the above tried to characterise the picture picture, however, only the first of these di- sign. It appears to involve meaning in a mensions is visible. much broader sense. The Peircean sign Of course, Hamilton does not only seems much more akin to a situation of substitute a bull for a dog and a Picasso communication. The object, then, is that figure for Velázquez : he puts his picture which attracts the attention of somebody, together from units appearing in many Pi- and inspires him to produce an artefact, casso pictures. However, if he thus projects called representamen, which, when it is re- a Picasso paradigm onto the Velázquez lated back to the presumed object, incites syntagm, it is a paradigm only existing somebody else to produce an interpretant. post hoc, after Picasso has gone through Thus, for instance, when Velázquez looks all his different styles. at the infanta, he is motivated to produce a In conclusion, then, syntagms and configuration of shapes and colours on the paradigm are not basic principles for the canvas, which I interpret as corresponding organisation of pictorial meaning. They to the infanta. My immediate interpretant

23 (corresponding to the immediate object) from the sign relation ; indexicality, when is really something like “a small girl with they are related in some way in the real long hair adorned with a bow wearing a world, independently of the sign relation; crinoline and having a stiff corporeal atti- and symbolicity when their relationship is tude”, etc. If I had been a contemporary of the result of a specially introduced rule. Velázquez, it might have been sufficient for As we will see, these notions really have me to make some further inquiries within a much wider application than the concept the real world, to arrive at the dynamical of sign, but they remain relevant to it, and interpretant corresponding to the dynami- they are most easily understood in relation cal object “The Spanish infanta Margarita to the sign. As any elementary handbook of who was five years old in 1656”, and so or. semiotics will tell us, the picture is certain- As it is, I will have to be content with other ly an iconic sign, the reproduction, since it signs – books about Velázquez, about the is based on a photograph, is an indexical royal family, etc., other pictures, and so on sign, and the title is obviously symbolic, in – in order to reconstruct by these means the Peircean sense. some similar dynamical object. In princi- But our preceding discussion has al- ple, I can go on for ever, hoping to reach ready hinted at one way in which we will one day the final interpretant. have to make the Peircean trichotomy more What this means is that the process precise: we cannot take it to distinguish ob- of interpretation is never finished. It goes jects, but only relations between objects. In through many phases and it never really ar- this sense, there is nothing which is an icon, rives at the end. This applies to perceptual an index, or a symbol. Rather, between reality itself. If I meet the infanta herself, some objects which we identify as expres- her appearance, just as well as Velázquez’ sions and contents, relationships are to be picture, may trigger off the process of in- found which may be described as iconi- terpretation. It is in this sense we must un- cal, indexical, and/or symbolic. Thus, the derstand that to Peirce the universe is pro- works of Velázquez, Picasso, and Hamilton fused with signs. are all iconical. And while there is certainly For some purposes, such a concep- a sense in which Velázquez’ work is “more tion of meaning is useful; for others, it is iconical” than the others (in relation to the not. It will not tell us what is peculiar about most obvious content), it also has indexi- pictures. It may be elaborated, however, in cal and symbolic traits, just as the works of some rather un-Peircean ways, in order to Picasso and Hamilton, both in relation to allow for the necessary types of compar- the same content, and to other, less obvious isons. Yet there is some clues in Peirce’s contents (such as the continuity of the sur- own work for these elaborations. face or the scene). Much more will have to Everybody is more or less familiar  The term “symbol”, as it appears within with the three relationships between rep- the Peircean triad, is very problematic to use: in resentamen and object (or, as I will con- Saussurean terminology, for instance, the same term stands for a particular brand of the iconic sign tinue to say, expression and content) which (as we will see below), and in many other connec- Peirce distinguishes: iconicity when there tions it is simply used in the way we employ the term “sign”. In this lecture, however, we will fol- are some properties they share, quite apart low the Peircean usage. 24 be said about this later. Suffice it to say, for it had not been assisted and complemented the moment, that the Peircean distinctions by pictures. Nothing of this can however are useful, only to the extent that they are be clearly seen, before we have discussed complex, and can be reconstructed within the specific differences between the semi- a wider framework. otic resources offered by pictures and lan- guage, respectively. The picture as a cognitive- semiotic operation The picture as socio-semiotic The picture must be seen within much operation wider frames than those of art history. Art Nothing of what has been said above is historians will never forgive me for say- meant to suggest that pictures do not have ing so, but their discipline is reductionist, a part to play in that domain that some psy- not only for excluding pictures not having chologists would place in-between child any “aesthetic” properties, however those development and the development of the are defined, but in particular because it re- species: cultural history. Indeed, this is also mains focused on the small world of art, a way in which pictorial semiotics goes be- when pictures should really be studied yond art history (though perhaps not in the from the point of view of world-history, not sense of a general “science of pictures”). simply the human kind, but that of general In is obviously interesting to place the biology. Pictures are the result of peculiar works of Velázquez, Picasso, and Hamil- ways of thinking, of “visual thinking”, in a ton within the sphere of cultural semiotics, deeper sense than that favoured by Gestalt in which they will not simply be the crea- psychology, or rather, more specifically, of tions of famous individuals, but artefacts “pictorial thinking”. produced by means of specific, historically In this sense, the “graphic act”, by dated, semiotic operations. The point is, means of which figures are created on once again, not simply to produce an “art surfaces (giving rise to writing as well as history without names”, but a history of pictures) is a specific type of cognitive op- possible socio-semiotic operations. Again, eration, a dual kind, which could more in this is only possible once we have analysed particular be called a semiotic operation: the specific resources offered by the semi- one which requires us to separate clearly otic sphere of pictures. that by means of which the thinking is done from that about which something is Summary thought. The emergence of the graphic act The goal of this section has been to indi- is an important moment in the development cate in what way the questions asked by of the child as well as of the human species, pictorial semiotics are different from the although is has so far been somewhat ne- kind of questions addressed to the same glected by most of those who have written objects of study by traditional disciplines about the process by means of which man- such as, most notably, art history. We have kind becomes distinct from other animals. seen that the interest of pictorial semiotics There are reasons to believe that language is directed to general, rather than partic- only would not have brought us very far, if ular, facts, and that it is a fundamentally

25 comparative approach: it tries to find out ing the possibilities of semiotics for becom- in which way the resources put at our dis- ing a science; but I will do so for historical posal by picture-making are different from and social, rather than systematic, reasons those, for instance, of verbal language. We are involved then, with the different ways Out of the semiotical soup in which pictures may be used or, more shops: Semiotics and precisely, the constraints upon their usage, Philosophy as well as the possibilities they offer. From From an epistemological point of view, it our point of view, then, pictorial semiot- seems rather simple to ascertain that semi- ics is concerned with the place of pictures otics can in no way be a method or a mod- within the process making the human be- el. Not to overburden our argument, let us ing into a very particular animal indeed, as define a method as a series of operations well as with the position taken by pictures which might be applied in ordered stages in different historically given societies. to an object of study, with the goal of yield- ing information of a particular kind about 1.2. Theoretical the object studied; and let us similarly de- beginning : Semiotics cide that a model is a simplified, but still as a distinctive more or less iconic, representation of the object studied which can be more easily discipline manipulated than the real thing, and which Before we embark on the semiotic study (ideally) has the advantage of representing of pictures, we need to have some inklings classes of objects of a particular category, about what this involves. Elsewhere, I have rather than a single object, so that, when argued that semiotics cannot be considered methodological operations are applied to it, to be some kind of method, a “model”, a it yields information about the category of particular philosophical tradition, or even objets concerned. an “interdisciplinary perspective”, what- It should be obvious that semiotics ever that may mean; nor is it simply a cri- cannot offer anything of the kind – or, rath- tique of ideology or a “meta-analysis”; but er, it offers too much of it. For semiotics, it must be taken to be a science in its own just as all other sciences, contains a wealth right (cf. Sonesson 1992; 1993a, b; 1994a, of models, as well as a panoply of methods. b; 1996a). This seems to leave us with only When one particular model and/or method one way of looking upon the interplay be- is attributed to semiotics, it is obviously tween semiotics and other enterprises such being confused with one of its manifesta- as art history, literary history, general his- tions having course during some particular tory, archaeology, psychology, sociology, period, most probably the movement know and so on : that semiotics may function as as French structuralism, which was popu- an ancillary science to any of them, just as lar in the 1960:ies and 70:ies, but which has they may play the same part in relation to since lost its relevance in most quarters. It semiotics, when the latter is pursuing its may rightly be said about French structur- own aims. In the following, I will take a alism that it tried (mostly in vain) to apply somewhat more sceptical stance concern- a linguistic model (itself abusively derived 26 from the linguistic structuralism devel- and mathematics, semiotics is now well on oped, notably, by Saussure and Hjelmslev), its way to the elaboration of its proper mod- as well as to implement (but completely els (cf. Sonesson 1992a, c; 1993a, 1994a, failing to do so) the method of the same 1996a, 1998). linguistic school. Nor should we adopt the popular pre- Semiotics as such is not restricted to conception, according to which the semi- any single method, but is known to have otic field is inhabited simply by the fol- used several kinds, such as an exhaustive lowers of Peirce and Saussure. In the first analysis of concrete texts, or text analysis place, there would be no reason (more than (comparable to distributional analysis in a superficial terminological coincidence) linguistics and “explication de texte” in lit- to amalgamate two such dissimilar doc- erary studies), as well as classical experi- trines as those represented by the elaborate mental technique (well-known from psy- but fragmentary philosophy of Peirce, and chology) and imaginary variation of prop- the marginal, if suggestive, annotations of erties, or system analysis, reminiscent of Saussure. But, more importantly, in adopt- the kind of reasoning found in philosophy, ing this point of view, we would be unable most explicitly in phenomenology. In addi- to account, not only for the semiotical work tion, semiotics has employed a hybrid form accomplished well before the time of our of text analysis and imaginary variation two cultural heroes, be it that of the stoics, which I have elsewhere called text classi- Augustin, the scholastics, Locke, Leibniz, fication, notably in semiotically inspired or the ideologues, but also for much of con- rhetoric. Nor is semiotics necessarily de- temporary semiotics, some parts of which pendant on a model taken over from lin- are not particularly indebted to any of the guistics, as is often believed, although the forefathers. construction of models remains one of its In an article in which he says many peculiar features, if it is compared to most sensible things in defence of semiotics, of the human sciences. Indeed, semiotics Umberto Eco (1988: 323ff) comes up with differs from traditional approaches to hu- a very strange conception of what the latter manitas, whose domain it may partly seem is: on the one hand, he admits that there are to occupy, in employing models which certain specific semiotic sciences, such as guide its practitioners in their effort to those which study the interpretative habits bring about adequate analyses, instead of of events in verbal language, gestures, traf- simply relying on the power of the “inno- fic signs, pictures, and so on; on the other cent eye”. After having borrowed its mod- hand, he claims that there is a general sem- els from linguistics, philosophy, medicine, iotics, which simple postulates the concept of sign, thus permitting us to speak about  Bouissac (1999a, b) also talks about four superficially dissimilar things within a uni- “ways of acquiring knowledge” within semiotics and elsewhere, which partly correspond to my di- fied framework. The latter, he maintains, is vision: “experiment” and “reasoning” has obvious not a science, but a philosophical activity, parallels, “serendipity” would for me be something and this is in his view demonstrated by the occurring at certain moments within the other strategies, and “meta-analysis” is an aspect which very proliferation of different conceptions I have not mentioned, but to which I will turn be- of what semiotics is. Indeed, it is a variety low. 27 of the philosophy of language, which has among the social sciences and the humani- the particularity of going beyond the study ties, because what they have in common is of statements, to the underlying activity, that they are involved with meaning. Eco and which does not limit itself to a single (1988: 351) himself points out that while semiotic system, verbal language. the natural sciences are interpretations of It is interesting that Eco should ad- the first degree, the semiotic sciences are mit that the study of specific semiotic phe- interpretations of interpretations. The lat- nomena are sciences; but that is no doubt ter, undoubtedly, also applies to what ar- because some of these sciences existed chaeology does with artefacts left in some well before modern semiotics was in the prehistoric burial; it may not apply to the works. The study of verbal language, for radiocarbon dating of these artefacts, but it instance, has long been known as philol- certainly applies to the interpretative frame ogy or linguistics. In some cases, how- in which the resulting dates are later insert- ever, this conception would require the ed and given a meaning. More obviously, it establishment of new disciplines: there is, applies to most things done in art history, for instance no well-accepted branch of though, once again, the study of artistic knowledge involved with the study of ges- materials is only indirectly contained with- ture, which is still treated within anthro- in this description, because of the chemical pology or psychology, or under the absurd analyses being made on substances defined and misleading heading of “non-verbal for an “artistic” purpose. communication”. The semiotics of picto- But Prieto allowed general semiotics rial signs is even more in need of being to subsist and to remain a science, although established as an independent discipline, at another level of generality. Although because art history has never been inter- Prieto is not very clear about the nature ested in pictures as such, and the findings of this general semiotic theory, his own of recent perceptual psychology have to be work within the domain seems to imply brought into contact with more systematic the conviction that it should not only fur- studies, similarly to the way in which post- nish the semiotic sciences with a coherent Chomskyan linguistics has been related to framework, before the specific disciplines psycholinguistics. The rudiments of a body can accomplish their task, but that it would of knowledge corresponding to a semiotics also be called upon to compare the results of pictures already exist, as we shall see; of these disciplines, in order to determine but it can hardly be considered a well-es- how different resources for conveying sig- tablished discipline. nification may differ. Whether or not this This part of Eco’s thesis was actually common framework consists in the con- formulated well before him by Luis Prie- cept of sign, or if something different, or to’s (1975a,b), who argued that disciplines something additional, is needed, it seems such as anthropology, ethnology, sociolo- strange to say that this framework is simply gy, psychology, literary history, art history, “postulated” by a philosophical movement, history of religion, archaeology, and so on, as Eco maintains. If so, all these disciplines should more aptly be called the “semiotic would only be valid, given a particular sciences”, rather than being distributed philosophical framework, and for someone 28 not sharing this framework, all these par- selves take up a position outside of semiot- ticular domains of study would have noth- ics. Eco’s claims, however, are made from ing to contribute. In the end, then, specific within semiotics itself. A more commonly semiotics would also be given over to the voiced point of view among people closely whim of philosophy. involved with semiotics is that it is “an inter- Curiously, Eco even claims that the disciplinary perspective”. I find it difficult fact of there being different semiotical to see the point of that description. Either it points of view demonstrates that semiot- means that people representing a lot of oth- ics is a philosophical activity; but, at the er more well-established disciplines come very least, this would show that semiotics together at semiotic congresses; but, if so, is a class of different philosophical and/or it does not describe any situation which is scientific activities. Actually, a much more original to semiotics, and there is no reason natural conclusion would be that, just as for this state of facts having to determine sociology, psychology, archaeology, liter- the future of any discipline. Or it really ary history, and so on, semiotics can be means that semiotics itself is something practised from the point of view of differ- “in-between” other disciplines. If so, that ent philosophical conceptions. Thus, there is not particularly new either: from social may be a structuralist semiotics, a nomi- psychology to cognitive science, other dis- nalist semiotics, a phenomenological semi- ciplines have been born out of such an in- otics, and so on – just as there may be, for termediate space. This also means that the instance, a processural and a post-proces- phrase cannot describe the particularity of sural archaeology, a positivist and a post- semiotics: there are a lot of other “inter- modernist art history, and so on. disciplinary perspectives”. So, at the very The way to get out of the “philosoph- least, something needs to be added to this ical soup shops”, to adopt Peirce’s phrase, definition. is to bring semiotics itself out of them. A more sophisticated version of this All sciences have once separated them- description is Paul Bouissac’s (1998: 1999a, selves from philosophy – a process which b) claim that semiotics is mainly involved of course (as we shall see) always leaves a with “meta-analysis”, which “consists in residue in the tureen.10 reading through a large number of special- ised scientific publications, selected among Those who look upon semiotics as a method or a model undoubtedly them- the published literature in one or several do- mains of inquiry, and of relating the partial  Deely (2001:700ff), who seems to accept results within a more encompassing model this division, points out however that there must be an interaction between the foundational discipline than the ones that are held by the various and its different specialities, but this makes non- specialists concerned” (1999a: 4). This is sense of the division. Deely’s terms, theoretical versus applied semiotics are therefore much better. indeed something which semioticians tend to do; but so do of course a lot of people 10 Another variety of this thesis could be to present semiotics as a “critique of ideology”, working within cognitive science and a which is a conception probably first formulated lot of other purportedly “interdisciplinary be Volocinov, and more famously repeated by Kristeva, but it is still found in Angenot 1985. This perspectives”. We are still left with the self-contradictory conception will not be further question what the specificity of semiotics discussed here. 29 is. It cannot lie in that “more encompassing the point of view which we adopt on other model”, for we have seen that semiotics is objects (see Sonesson 1989a,I.1.4.). More more than a model, since it makes use of recently, Greimas has rejected the notion of a lot of them. Of course, it may contain a sign, and his followers Floch (1986a) and class of more wide-ranging models. But in Thürlemann (1982; 1990) have argued the order to contain models, it must be some- case in the domain of pictorial semiotics. thing else: a science. In a similar fashion, Eco (1976) himself, at So what, then, is the central frame- the end of his tortuous critique of iconicity, work provided by a semiotic “meta-analy- substituted the notion of sign process for sis”? Not simply the postulated concept of the traditional sign concept. sign, as Eco suggests. I would be the first to So there seems to be wide agreement agree with Bouissac (1998) that the notion within semiotics, although with somewhat of sign is insufficiently defined in semiot- different slants, that the sign (also termed ics. In fact, I have often argued that both the semiotic function) is not comprehen- the central traditions, the Peircean as well sive enough to delimit the field of semi- as the Saussurean, simply presuppose the otics: rather, the domain of semiotics is essential components of the sign (cf. Son- meaning (or “mediation”), in some wider, esson 1989a; 1992a,b; 1996a: in lecture 2 yet to be specified sense. However, since below). Contrary to Bouissac, however, I everything, or almost everything, may be think the concept of sign makes perfectly endowed with meaning, any object what- good sense, once it has been properly de- soever (or almost) may enter into the do- fined (which is precisely what I will try to main of semiotics, but only in so far as it do later in lecture 2). Itself a fruit of meta- is studied from the point of view of its ca- analysis, my definition abundantly refers pacity for conveying meaning. Semiotics, I to ontogeny, as well as to phylogeny. How- will contend, is not about what something ever, this does not mean that the concept means; it is about how it means. of sign is sufficient to define the domain of semiotics, which has to be much wider, at Two ways of defining sciences: least because signs cannot be treated inde- Semiotics vs. the History of pendently of a wider concept of meaning. Religion According to Saussure, semiotics (or So far, I have tried to characterise complex semiology as he called it) was to study “the notions such as method, model, movements, life of signs in society”; and our second and so on, in very simple terms, sufficient mythical founding-father, Peirce, as well to rule out the possibility of semiotics being as his forerunner John Locke, conceived of one of those things. Now we face the even semiotics as being the “doctrine of signs”. more daunting task of trying to determine Later in life, however, Peirce come to pre- what a science is. As a first approximation, fer the wider term “mediation” as a de- one may want to say that a science is a par- scription of the subject matter of semiotics ticularly orderly and systematic fashion for (cf. Parmentier 1985). And Saussure actu- describing and analysing or, more general- ally argued that in the semiotic sciences, ly, interpreting a certain part of reality, us- there was no object to be studied except for ing different methods and models. At this 30 point we may want to introduce a division find the same thing in the natural sciences: between natural sciences, on the one hand, chemistry and physics often appear to be and social and human (or, better, semiotic) different points of view taken on the very sciences, on the other, which, following a same matter. This is not the whole truth: in traditional hermeneutical conception ech- fact semiotics, psychology and sociology oed by Eco (1988: 351), separates the inter- only apply their points of view to the hu- pretation of facts from the interpretation of man world, or at least to the world of liv- interpretations. Normally, it is added that ing beings (in most cases, to animals, not the first kind of knowledge involves phe- to plants). So the point-of-view approach is nomena for which laws may be formulated, supplemented by a domain-approach. The while the second kind only refer to unique domain of chemistry and physics is much occurrences; and that while the second wider: its goes well beyond the human type may be understood, the first can only world. But both apply the same point of be explained. As we will see, this is largely view to the human world and what lies be- a pre-semiotic conception. hind it, which is impossible for semiotics, But there is something seriously as well as for psychology and sociology. wrong with this analysis, even at its earli- Contrary to chemistry and physics, biology est stage. Not all sciences appear to have is not just another point of view, but it is their own reserved piece of reality to study. also domain-specific: it only involves liv- It seems to me that sciences may be defined ing creatures. This may explain that there either as being preoccupied with a particu- is now such a speciality as biosemiotics but lar domain of reality, or as applying a par- not (at least I hope so) chemical semiotics. ticular point of view to the whole of reality It is impossible to establish a consen- (which is really one and the same). Thus, sus among all semioticians on what semi- French studies are involved with French otics is all about; and many semioticians language and literature, linguistics with all will not even care to define their discipline. languages (or what is common to all lan- However, if we attend less to definitions guages); similarly, the history of religions than to real research practice, and if we describes a very particular domain of real- leave out those would-be semioticians who ity, religion, as it evolves through history simply do not seem to be doing anything (and pre-history). Even within the natural very new (those who merely go on doing sciences, there are some sciences that have art history, literary history, philosophy, their particular domains, such as geogra- logic, or whatever), it seems possible to phy, astronomy, and meteorology. This isolate the smallest common denominators seems to be even more obviously true of of the discipline. such applied sciences as medicine and den- In the following, then, semiotics will tistry. be taken to be a science, the point of view But there is no semiotic domain, just of which may be applied to any phenom- as there is no psychological or sociologi- enon produced by the human race or, more cal one: rather, everything may be studied widely, by living beings. This point of view from the point of view of its semiotic, psy- consists, in Saussurean terms, in an inves- chological, or sociological properties. We tigation of the point of view itself, which is

31 equivalent, in Peircean terms, to the study use, we cannot, like the philosopher Nel- of mediation. In other words, semiotics is son Goodman (1968), reject the folk notion concerned with the different forms and of picture because of its incoherence, but conformations given to the means through must discover its peculiar systematicity. which humankind believe itself to have ac- But it does not follow, as Prieto (1975a) cess to “the world”. This is at least the way would claim, that we must restrict our I have formulated the task of semiotics in study to the knowledge shared by all users my earlier work. For many reasons (which of the system, for it is necessary to descend will be discussed more fully in the second at least one level of analysis below the ulti- lecture), it now seems impossible to limit mate level of which the user is aware, in or- semiotics only to the way the human world der to take account of the presuppositions is endowed with meaning. Even when dis- underlying the use of the system. Semiot- cussing pictures, which are peculiar to hu- ics must go beyond the standpoint of the man beings, we can only understand their user, to explain the workings of such oper- specificity in contrast to meanings handled ative, albeit tacit, knowledge that underlies by other animals. It will therefore be better the behaviour constitutive of any system of to avoid any kind of belief-predicates in the signification (cf. Sonesson 1989a,I.1.4). characterisation of semiotics. Thus, semi- Moreover, semiotics is devoted to otics should here be said to be concerned these phenomena considered in their quali- with the different forms and conformations tative aspects rather than the quantitative given to the means through which living ones, and it is geared to rules and regu- beings are observed, through their interac- larities, instead of unique objects. This tions with it, to have access to “the world”. is to say that, pictorial semiotics, like all The very term “point of view” is of semiotic sciences, including linguistics, is course a visual metaphor. Yet the point, a nomothetic science, a science which is which is a standpoint, matters more than concerned with generalities, not an idio- the sense modality. For, in studying these graphic science, comparable to art history phenomena, semiotics should occupy the and most other traditional human sciences, standpoint of humankind itself (and of its which take as their object an array of sin- different fractions). Indeed, as Saussure ar- gular phenomena, the common nature and gues, semiotic objects exist merely as those connectedness of which they take for grant- points of view that are adopted on other, ed.11 I would like to insist on this combina- “material” objects, which is why these tion here, since it overrides the traditional points of view cannot be altered without divide between the humanities and other the result being the disappearance of the sciences, postulated by the hermeneutical semiotic objects as such. Analogously, is tradition from Dilthey and Weber to Haber- has been argued that we should have to mas and Apel: even a traditional semiotical adapt the point of view of the bat, let alone 11 This is not to say that semiotic results the tick, but it is not clear that this can be must be formulated in terms of Hempel’s cover- done in the same sense. ing law (as has been claimed by some exponents of “New Archaeology”): we are referring to the Taking the point of view of the us- distinction between nomothetic and idiographic ers, and trying the explain their particular descriptions in the more general sense of Rickert and Windelbrand. 32 discipline such as linguistics, including the imagined. study of any particular language, involves In contrast, archaeology is today a the establishment of laws and regularities, well-established science, represented, as a not individual facts. Just like linguistics, matter of course, at most universities, and but contrary to the natural sciences and to occupying a pride of place among those some varieties of the social sciences, all endeavours founded by state, regional, and other semiotic sciences are concerned with even international, institutions.12 That is no qualities, rather than quantities – that is, doubt as is should be, for archaeology, in they are concerned with categories more its different avatars, has contributed a great than numbers. Thus, semiotics shares with amount of knowledge about human culture the social and natural sciences the charac- and behaviour through the ages. And yet, ter of being a law-seeking, or nomothetic, at least with the aid of the simple epistemo- rather than an idiographic, science, while logical model we have used so far, it is dif- retaining the emphasis on categories, to the ficult to say what archaeology is all about. detriment of amounts, which is peculiar If there is some particular archaeological to the human sciences. Being nomothetic domain of study, it is hard to discern it; and and qualitative, pictorial semiotics has as if there is a peculiar archaeological point of its principal theme a category that may be view, it is not easy to define. termed pictorality, or picturehood – which This is hardly surprising: after all, is not, as we shall see, simply the same archaeology emerged rather recently, not, thing as iconicity. as most disciplines, out of philosophy, but The social institution of science: from what, on first sight, may seem an un- holy alliance of travellers and adventurers Semiotics vs. Archaeology (often with diplomatic passports), on one This is as far as I have taken the argument hand, and of museologists and other spe- in earlier articles. But there is certainly cialists in classification, on the other (cf. something wrong with this reasoning. Bahn ed. 1996; Trigger 1989). According When he suggested the creation of “semi- to a common suggestion, archaeology is ology”, Saussure claimed that its place about “prehistory”, which is then charac- among the sciences was prepared before- terised as the period before the advent of hand. Ironically, a century later, semiotics writing (cf. Fagan 1998:4f). If so, is this a still does not exist as an independent disci- domain of study, or a particular point of pline, apart from a few universities such as view? It would be the former, if it meant Lund, Tartu and Bologna. It certainly has that archaeology was dedicated to the de- more of an existence as collaborative net- works between institutes and countries in 12 As will be obvious here, if not before, different parts of the world, as partial defi- the present discussion relies heavily on a contrast nitions of research positions, and of course between archaeology and semiotics, which I de- veloped in a lecture, which has not been published, in the form of numerous associations, con- that formed part of a colloquium on semiotics and gresses, reviews, and books. But, clearly, archaeology, organised at the Swedish Institute in Istanbul in December 2 to 10, 2000. The reflec- something tells us that the place of semiot- tions of archaeology here simply serve as an exam- ics was not as well-prepared as Saussure ple, but they will be more directly relevant to our discussion in lecture two. 33 scription of a world in which writing did ered that the Ancient Maya had true writ- not exist; if would be the second, if it meant ing. Perhaps we could live with this later that archaeology involves the description consequence. But there is a third objection, of the world as it appears when we only which may turn out to be more serious: it have recourse to knowledge not transmit- is not obvious that there is such a clear-cut ted through writing. The latter description difference between true writing systems undoubtedly seems most promising from a and different kind of “pre-writing” that is semiotic point of view: it would imply that allows for the distribution of the past into archaeology only describes the world as it two or more domains of study (cf. Bouissac can be recovered by means of information 1997: Rudgley 1998). conveyed by a particular kind of semiotic Fagan (1998: 4f) also offers a second vehicle. criterion: archaeology, as opposed to his- Fagan implicitly appears to opt for tory, “is, most of the time, entirely anony- the latter alternative: he says archaeology mous”. Perhaps it is some similar idea that is concerned with periods in which most explains that Renfrew & Bahn (2000: 11) people are not literate, so that much new describe archaeology as ”’the past tense knowledge may be gained from excava- of cultural anthropology’” (their quotation tion. Moreover, he proceeds to oppose marks), not, for instance, of history. If, so, “text-aided archaeology” to “prehistoric like Fagan, they are referring to a rather an- archaeology”.13 However, it seems that a lot tiquated (but still dominant) conception of of archaeology would then turn out to be of history, pre-dating the preoccupation, initi- the hybrid kind: in fact, “text-aided archae- ated by the Annales school, with enduring ology” would not only include the study of, structures and long-time developments. for instance, the castles erected during the Once again, this conception seems to have Middle Ages, but also, within the domain the effect of allowing archaeology itself to of “industrial archaeology”, of near-con- transform something into the subject mat- temporary buildings (cf. Renfrew & Bahn ter of another science. To pick the same ex- 1991).14 In the second place, the very ef- ample, as long as everybody accepted the fect of archaeological work may then be opinion of Sir Eric Thompson, according to to make some phenomenon cease being an which Mayan writing was only concerned object of study for archaeology and being with astronomical events, Mayan studies transformed into a subject matter of some were part of archaeology, but now that we other branch of learning, such as history. know that the inscriptions on many stelae Thus, Mayan studies would suddenly stop concern highly individual “lords” of differ- being archaeological, once it was discov- ent Mayan cities, as well as the wars the waged on each others, the subject would 13 From our point of view, this prompts the question: should we also add a “picture-aided cease being archaeological. Quite apart archaeology”? And perhaps even further hybrids from this embarrassing situation, not only such as a “trace-aided archaeology”? do recent, “post-processural” thinkers such 14 It has been argued that, although intro- as Hodder (1991) argue that archaeology ductory text books give a different impression, most archaeology is really involved with historical should be more concerned with individu- times, that is, it is “historical archaeology” (An- als, but Renfew & Bahn (2000:9f) claim to drén 1997: 12f). 34 discover some convergence with classical In some respects, however, I might have “processural” archaeology on this point. been much more on the right track in an Let us suppose, just for a moment, earlier publication (Sonesson 1989a,I.1.), that what Fagan really wants to say, is that where I suggested that semiotics could per- archaeology is some particular method, haps best be viewed as a series of entan- which can be used when the more direct gled strains of problem areas making up a method of simply “reading the text” is not continuous discussion extending through available: that is, a way of recovering old the centuries, and that is was only by tak- artefacts, usually by means of excavation. ing a retrospective view of (some restricted As a complete characterisation of archae- part of) this mesh, that semiotics could be ology this will of course not do. Digging defineda posteriori, for instance as I did in is simply the first (or perhaps some middle earlier paragraphs. From an epistemologi- part) of the endeavour called archaeology. cal point of view, nothing would change. It involves a lot of other methods, for recon- This research tradition would still be char- struction as well as analysis, from carbon acterised by its peculiar point of view. And dating to experimental archaeology. Yet it it would not be equivalent to a “doctrine of seems to me that it is the only operation signs”. It would be much more like a dis- that really singles out archaeology. With- cussion: a network of problems branching out excavation archaeology is not archaeol- out ever further through the centuries. In ogy. As an enterprise, archaeology is not the following, when I talk about semiotics defined by any particular domain, nor any as a science, it should be understood in this point of view. It is defined by a method, sense. Indeed, I would like to claim that a which is not even a method of analysing, science is simply a research tradition, in but of acquiring the artefacts that it turns the above-defined sense, which has been into its object of study. institutionalised within society. For the moment, it does not matter Meta-analyses in our time: whether I might convince any archaeologist about the correctness of my observations. Semiotics and Cognitive But the preceding discussion shows clearly Science that something being a science is much To say that something becomes a sci- more a social than an epistemological fact. ence because of social reasons is not to Saussure’ idea of there being some kind suggest that those reasons are necessar- of system of the sciences within which the ily superficial, the result of power games position of semiotics is already prepared and nepotism. In the case of semiotics, it now seems rather ingenuous, and so does may simply be the case that semiotics has my own division of the sciences into those so far failed to demonstrate it usefulness based on divisions into domains and those to wider groups within society. However, adopting a particular point of view. society as such is certainly also at stake: I still think it would be useful for for some reason, the fortune of semiotics semiotics to be recognised as a science. has been very different in Latin, and in It would then gain some of the coherence particular Latin American, countries, than and the focus it still seems to be lacking. in the Anglo-Saxon, and more generally 35 Germanic, world. People in the latter part suffers by being endowed with meaning. of the world would no doubt tend to think Indeed, in an earlier phase, cognitive sci- that this is so because Latin culture is more ence seemed more susceptible of being susceptible to intellectual fads. There may described by a simple model: the mind as be some truth in this, if semiotics is iden- computer. At present, however, even cog- tified with intellectual fashion statements nitive science has several models, one of such as structuralism, post-structuralism, which could be described as involving the and post-modernism. But this is a very mind as brain. limited, and uninteresting, way of looking The disciplinary history of these two at semiotics. approaches has been very different. Cogni- It might be useful here to contrast tive science is often described as the result semiotics with another brand of “meta- of joining together the knowledge base of analysis” which has met with more luck rather disparate empirical disciplines such in the contemporary world, at least in the as linguistics, cognitive psychology, philos- sphere under Anglo-Saxon influence: cog- ophy, biology, and computer science. Thus, nitive science. Like semiotics, cognitive sci- instead of one research tradition connected ence is often conceived as an interdiscipli- through the ages, cognitive science repre- nary perspective that sometimes (no doubt sents a very recent intermingling of several more often than semiotics) has gained the research-traditions having developed sepa- position of an independent discipline. Cu- rately until a few decades ago. Semiotics riously, it might be argued that cognitive has, in a more classical way, developed science and semiotics cover more or less out of the amorphous mass of philosophy, the same domain of knowledge – or rather, and still has some problems encountering to apply the observations made above, take its empirical basis. It might be suggested a very similar point of view of the world. that the basic concept of semiotics is the This in itself is controversial, since semi- sign, whereas that of cognitive science is otics and cognitive science offer very dif- representation – even though there is a long ferent characterisations of their domain (or, tradition in semiotics of rejecting the sign strictly speaking, the point of view taken concept, and recent cognitive science has on the domain). In some sense, however, marked its distances to the notion of repre- both are concerned with the way in which sentation.15 From the point of view of meth- the world described by the natural sciences ods, semiotics is generally speaking stuck appears to humans beings and perhaps also between the analysis of single ”texts “ and to other animals and some robots. Cogni- theory construction, whereas cognitive sci- tive science puts the emphasis on the place ence is closer to relying on experimental of the appearance of this world, the mental methods (including, of course, computer domain (although some of its exponents simulation). These differences partly may would not even recognise the mind as such, explain why semiotics and cognitive sci- but would rather talk about the brain and/or ence rarely are on speaking terms. the computer), and on its characteristic op- eration, cognition; and semiotics insists on 15 If this seems a paradoxical statement, I the transformations that the physical world must refer the reader to Lecture 2 for its elucida- tion. 36 On the other hand, there have recent- levels of meaning involving “mimesis” and ly been some encouraging developments language, from more basic ones, character- within cognitive science which, no doubt istic of all biological systems (life forms), with some exaggeration, may be qualified such as “cues” and “associations. as a “semiotic turn”: an interest in meaning In the end, then, what we do need, as such, in particular as it has developed, is some kind of meta-analysis: as Bouissac ontogenetically and, in particular, phylo- (1999a: 4) put is so well, we need a pro- genetically, in the human species and, to cedure which “consists in reading through some extent, in other animals and animal- a large number of specialised scientific like machines. Terrence Deacon (1997) is a publications, selected among the published researcher in neuroscience whose work has literature in one or several domains of in- been particularly acclaimed within cogni- quiry, and of relating the partial results tive science. Yet he has chosen to express within a more encompassing model than some of his main arguments in a terminol- the ones that are held by the various spe- ogy taken over from Peirce, who is perhaps cialists concerned”. No doubt cognitive the principal cultural hero of semiotics.16 science, by definition, has been better at Not only Deacon, both other scholars in- this than semiotics, because it is character- terested in the specificity of human nature ized by the confluence of various earlier now put their emphasis on the concept of research traditions, whereas semiotics has sign (which they normally term “symbol”, too long been hampered by the autonomy using this word is a sense in which we will postulate, taken over from Sassurean and not employ it here). This is true, in a very Chomskyan linguistics. For my part, I have general sense, of Donald’s (1991) stages of always sided with cognitive science in this episodic, mimetic, mythic and theoretical respect, even before it was invented. What culture. It seems to apply even more to To- cognitive science needs, however, is to masello (1999), less, in the end, because of take into account even more research tradi- his epigraphs taken from classical semioti- tions, one of which is no doubt semiotics. cians such as Peirce and Mead as well as However, meta-analysis taking a semiotic Bakhtin and Vygotsky, than because of the as well as a cognitive point of view might general thrust of his analysis, which consists perhaps better be called semiotics. In the in separating true instances of interpreting end, there may be no meaning without cog- actions as intentional from those which nition, and no cognition without meaning, may merely appear to be such. Building on at least given the wide definition of cogni- the aforementioned works, Jordan Zlatev tion characteristic of cognitive science.17 It (2002, 2003) is explicitly concerned with the conditions for the emergence of higher 17 According to the “language of thought” hypothesis (first formulated by Fodor) even cate- gorical perception and other elementary perceptual 16 Without trying in any way to diminish operations are based on cognition. Contemporary Deacon’s contribution – in fact, I find him very representatives of cognitive sciences such as La- convincing whenever he is not having recourse to koff and Johnson (on which more will be said in semiotic terminology -, I have earlier expressed later lectures) would seem to think that also think- serious misgivings about his way of using Peircean ing in a more traditional science may be reduced to terms, because this serves to obscure both the cen- very simply operations, in which case “cognitive tral issues of semiotics, and those introduced by science” becomes a misnomer. “Cognition” ap- Deacon. Cf. now Lecture 2 and 3. pears to have changed meaning, standing simply 37 might perhaps be said that semiotics differs tellectual scene, also had to define itself in from cognitive science simply by putting relation to (Husserlean) phenomenology, the emphasis on meaning rather than cog- at least in its French, subjectivist, variety, nition. known as Existentialism. At least the early Specialities such as pictorial semiot- work of such well-known French structur- ics can of course not be reduced to meta- alists as Greimas, Barthes, and Foucault analysis, because, in a fundamental sense, contains explicit phenomenological refer- they have to start from zero (even though ences. None of them really reflected on pictorial semiotics cannot do without per- the epistemological incompatibility of ceptual psychology). So much may be ac- phenomenology and positivism (though at corded to Eco (and Prieto). But even though least Foucault clearly marked his distanc- general semiotics must feature meta-anal- es later). Some later semioticians, such as ysis in an essential way, it should not be Jacques Fontanille and Jean Peititot, have viewed as simply a tradition within phi- later returned to the phenomenological tra- losophy. As Peirce said, we have to get out dition. At the same time, however, semiot- of the philosophical soup shops. But some ics generally has largely grown out of the philosophical residue will no doubt be left structuralist strait-jacket (although I feel in the tureen. rather lonely having tried to take into ac- count the rich intellectual yield of structur- From Phenomenology to alism, as well as showing what was wrong Ecological semiotics with this epistemological stance). Just like (French) structuralism was semiot- From this point of view, cognitive ics with a particular epistemological slant, science still seems to remain at the stage of cognitive science so far often has been a structuralist semiotics. It is a meta-analy- study of cognition equipped with a particu- sis determined by the computer-metaphor, lar epistemology. Basically, French struc- both as a way of constructing models, and turalism was characterised by a positivistic (probably less) as a method of analysis conception of the world and of scientific known as simulation. No doubt, while ear- method, taken over less from Saussure than ly cognitive science was entirely depend- coming out the subsequent development of ent on the idea of the mind as a computer, linguistics prior to the advent of Chomsky functioning on the model of extant compu- and forming the background of distribu- ter programs, recent decades has seen the tionalism and behaviourism. As all French advent of computer programs, called “neu- intellectual fads at the time, Structuralism ral networks”, constructed so as to function (in this sense) obviously also had to take as models of the mind, identified with the Freud and Marx into account, which could brain, or at least as models of some aspects only be done by tempering the positivist of brain functioning. This is perhaps the conception, or rather, concomitantly ren- sense in which Pinker (1997; 2002) sug- dering it rigid and inoperant. Something gests that the idea of the mind as computa- which is less well-known, however, is that tion is wider than the “computer metaphor”. Structuralism, appearing on the French in- At least for some thinkers within this tradi- tion, this has prompted the question of how for that which may be simulated on a computer. 38 the mind relates to the brain, or, in other about the subjective one. It is often forgot- terms, the problem of explaining the “per- ten that Husserl not only inspired but also sonal level” from the “subpersonal level”. himself was inspired by the Gestalt psy- Some neurologists within the cognitive chologists. Close followers of Husserl such science framework have seen the necessity as, most notably, Gurwitsch (1957, 1966), of accounting for “qualia”, i.e. the mind were as much involved with phenomeno- as experienced by a subject (Edelman & logical psychology as with philosophy and Tonini 2000), and philosophers operating discussed the findings not only of the psy- within the same frames have tried to map chology of perception but of contemporary commonsense psychology to brain func- contributors to neurobiology such as Gelb tioning, often in terms of computer models and Goldstein. Also the early Merleau– (Bermúdez 2005). This should really bring Ponty was, in this respect, an exponent of cognitive science closer to phenomenol- phenomenological psychology. ogy, even though such as rapprochement Being a neurologist, Edelman (1992) has only been suggested in rare instances clearly does not discover the body from the so far (Gallagher 2005). The lack of input horizon of consciousness, like a phenom- from phenomenology and other philosoph- enologist, but quite the opposite, he implies ical traditions current during the turn of that the mind cannot be divorced from the the 19th century is clearly apparent in the body. In a sense, this is hardly controver- discussion between “simulation theory” sial: unlike those hypothetical angels pos- and “theory theory” concerning the rela- tulated by Max Scheler, human beings can tion between Ego and Alter.18 only boast a mind as long as they have a The task of phenomenology, as Hus- body. But, if this is true in the order of ex- serl saw it, was to explain the possibility istence, it is not necessarily so from the of human beings having knowledge of the point of view of investigation. After all, world; as a philosophical endeavour, phe- Brentano (1885) did not use a scalpel, much nomenology is about the way the world of less fMRI, to discover the property of in- our experience is “constituted”. As a con- tentionality (in the sense of directedness), trast, psychology is not about the world, which Edelman recognises as an irreduci- but about the subject experiencing the ble characteristic of consciousness; nor did world. However, every finding in phenom- James (1890) find any of those “Jamesian enological philosophy, Husserl claims, has properties” of consciousness repeatedly a parallel in phenomenological psychology, mentioned by Edelman in such a way. In- which thus could be considered a tradition deed, far from being “a deliberately non- within psychological science (cf. Husserl scientific set of reflections on conscious- 1962; Gurwitsch 1974). If consciousness ness and existence” (Edelman 1992: 159), is a relation connecting the subject and the phenomenology started out from the fact of world, then phenomenology is concerned intentionality and attempted to probe ever with the objective pole and psychology is deeper into its ramifications, in order to re- discover and amplify those very Jamesian 18 For a discussion of this tradition, which is useful even for those who are not able to agree properties of consciousness mentioned by with the Heideggeran conclusions (curious in this Edelman. Husserl and Gurwitsch may have author), see Gurwitsch 1979 (written in 1931). 39 been wrong to think of phenomenology subject matter. Even so, common sense as a discipline completely separate from gives rise to an apparent contradiction: my biology and psychology, but the relative body is necessarily experienced through disconnection of phenomenological reflec- my consciousness, but in my conscious- tions, like those of Brentano and James, ness it is experienced as being outside of from biological knowledge has no doubt it.19 All post-Cartesian mediations, from borne rich intellectual fruit. If “a biologi- those of Husserl to those of Merleau-Ponty, cally based theory of mind” can in some have been concerned to account for this respects “invigorate” phenomenology, the paradox. To do so, it is necessary to ac- opposite is certainly just as true. complish a painstaking analysis (of which It is, no doubt, phenomenology on there can by no better example than the the sense of phenomenological psychol- posthumous papers of Husserl himself, to- ogy which is of relevance here: it is in this gether with the – also largely posthumous sense that I think that, together with semi- – works of Peirce) of all those structures of otics, phenomenology should participate in the mind that are normally at the margin the confluence of research traditions mak- of consciousness. Since this is a question ing up cognitive science. Personally, I also of discovering the way in which that which have some sympathy for phenomenology has a meaning means, there is no other way as a philosophical stance (though not in the of achieving it, even if it is an extremely sense of transcendental idealism), but this fallible enterprise. is less essential in the present context. In this sense, as I have argued else- Phenomenology, like semiotics, where (Sonesson 1989: 27ff), all human takes as its point of departure the way and social sciences which aspire to dis- things make sense to us, that is, how they cover regularities, such as linguistics and mean. In this, very broad sense, phenom- other semiotic sciences, necessarily start enology accomplishes a semiotical reduc- out from phenomenology – and we should tion: things are considered only from the be happy if those phenomenological inves- point of view of their having meaning to tigations sometime manage to be as metic- us (which might be people of a particular ulous as those of Husserl and Gurwitsch. culture or subgroup, or humankind in gen- This common sense world from eral). From a phenomenological point of which all analysis of meaning must start view, there is, in a sense, no way of over- out was characterised by Husserl as the coming the divide between the mind and Lifeworld, paraphrased by the later phe- the world formulated by Descartes, for nomenologist Alfred Schütz as the world Descartes did not invent it: it is intrinsic to taken for granted. The Lifeworld, in this that phenomenon which, in Descartes’ own sense, must comprise both what, in recent words, is the most widely distributed in the cognitive science, is known as “naive phys- world: common sense. ics” (what we, as members of the human Common sense is not notorious for 19 Strictly speaking, this is not the problem being right: but if we ask ourselves how of our own body, nor of the other, but the more the body (and the rest of the world) makes general one of the external world, as pointed out by Gurwitsch (1979: 26f): but it is quite sufficient sense to us, then common sense is our very for us to note that it also applies to the body. 40 pool of knowledge: abductions, norms, sign systems Rhetoric pole: how should adaption to adaption to Ego express himself in order target source to get the adherance of Alter

Coding Inner reinterpretation

CON TEXT Concretisation Source A r t e f a c t Percept Target

Semiotic pole: what Hermeneutic pole: resources are at hand what was Alter trying to Transport for the mediation convey (or what is his between Ego and Alter work conveying)

Fig. 4. The communication model, as revised in Sonesson 1999b race, mpt as students of the natural sci- course only possible from within World 2. ences, believe about the physical world) And yet World 3 has its own existence, just and “common sense psychology” (what we as well as World 1.20 believe about ourselves other persons). The psychologist James Gibson, who some- The three sciences of times repeated Husserl’s very words in communication: rhetoric, describing what he called “ecological psy- hermeneutics, semiotics chology” (what we must take for granted Like semiotics, rhetoric and hermeneutics about the environment in order to be able may be viewed as age-old research tradi- to perceive the world as we do), is more ob- tions in search of scientific status. The case viously concerned with the naive physics of rhetoric is certainly somewhat different parts. Taking my clue from Gibson, I have from the others: it was a kind of epistemé called this kind of study, prefigured by (that is a science, not simply an art, a tech- both Husserl and Gibson, ecological semi- né), already in Antiquity, and it remained otics (cf. Sonesson 1993a, 1994a, 1996a, so through the Middle Ages, when it was 1997a, 2999a). However, as semioticians part of the Trivium, and beyond that at we still feel that something is lacking here: least until the Enlightenment, when Vico, the world of signs itself. In the terms of for instance, was a professor of rhetoric. Karl Popper (prefigured by many other au- More recently, is has again become a disci- thors), our experience is not only made up pline occupying a position at the university, of subjects with consciousness (World 2) though, like in the Middle Ages, it tends to and the world of physics (World 1), but also be reduced to a techné, in the Aristotelian of organism-independent structures such 20 The nature and origin of organism-inde- as languages and other semiotic resources pendent artifacts (which are of course independent (World 3). All access to these worlds is of of both matter and mind, both only in a limited sense) will be further discussed in Lectre 2. 41 sense, that is, merely a set of practical pre- I want to suggest that this sisterhood cepts which are taught to the students and could more conveniently be expanded to as not questioned, rather than a science trans- to comprehend also a third member, which forming this “art” into explicit knowledge, is of course semiotics. The resulting trin- as suggested by Aristotle. Only in the work ity is not the one proposed by Popper (and of the two Belgian schools of new rheto- even less that of Peirce): both hermeneutics ric, that of Perelman, and that of Groupe µ, and rhetoric are concerned with the Pop- which is semiotically informed, has rheto- perean World 2, that of the subjects, but ric taken on some of the trappings of a sci- semiotics is of course mainly concerned ence.21 with World 3, that of organism-independent As for hermeneutics, when it first artefacts. Instead, rhetoric, hermeneutics, emerged in Late Antiquity, and then again and semiotics occupy different positions in during the Middle Ages, it was certainly relation to the process of communication. an art in the Aristotelian sense, featuring They may be said to be different sciences a series of guidelines for the interpretation of communication taking different points of religious texts, to which later the canon of view on the communication process. But of Antiquity was added. Hermeneutics in in order to see this, we have to go beyond the Renaissance sense, on the other hand, the model of communication propounded remained a techné, but one which was ali- by the mathematical theory of information, mented by historical and critical research, which is too often taken for granted outside permitting the restitution of the true text. and inside of semiotics. At least since Schleiermacher and Dilthey, Even today all semiotic theory relies, however, hermeneutics rather seems to more or less explicitly, on the communica- have been transformed into a philosophi- tion model derived from the mathematical cal tradition, particularly tainted by the theory of information, which was designed Heideggerean conception of thinking, al- to describe a few, by now rather old-fash- though it also has been given its own status ioned, technological means of communica- within the social sciences, mainly thanks to tion, telegraphy and radio, and in particular the contribution of Habermas. Interesting- to devise remedies to the loss of informa- ly, within the recent hermeneutic tradition, tion often occurring during transportation. hermeneutics and rhetoric have appeared Largely because of the influence of Jakob- as some kind of sister sciences (most ex- son (1960) and Eco (1976; 1977), this model plicitly perhaps in the work of Gadamer), has been used inside semiotics as a model though very little is generally said about of all communication, all signification, and the latter.22 of all kinds of semiosis. This practice has produced at least 21 For the history of rhetoric, see, for in- stance, Conley 1990. More will be said about two symmetrical, equally negative, conse- the two Belgian schools on new rhetoric in later quences: by reducing all kinds of semiosis lectures. An important theoretical contribution to rhetoric is of course also Ricœur (1975), although to the mass media kind, in particular to the author is more well-known form his phenom- that employed by radio and telegraphy, we enological and hermeneutical inspirations. become unable to understand the peculiar- 22 An extensive overview of the hermeneuti- ity of more direct forms of communication; cal tradition is found in Ferraris 2002. 42 and by treating all semiosis as being on a is the task of the sender to recover the part par, we deprive ourselves of the means to of the message possibly being understood understand the intricacies added to direct by the receiver. This is an elementary rhe- communication by means of different va- torical operation. Of course, rhetoric is also rieties of technological mediation. Beyond about recovering such elements of the mes- this, we may even discover a third, even sage that may possibly be not only under- more serious consequence: by projecting stood but also convincing to the receiver, the communication model onto each and but conviction is based on understanding every form for conveying meaning, we (in some sense). There could also be a cor- lose sight of that which is really common responding hermeneutical operation, if to all kinds of semiosis hermeneutics is taken to include the dis- I will here suggest a model of com- covery of the most favourable interpreta- munication which takes the basic operation tion.23 of communicating to be, not transference More simply, looked at from this an- in space, or translation into another code, gle, rhetoric is concerned with the way of but the act of interpretation, which suppos- creating the message, so as to win the ad- es an active contribution on the part of the herence of the other, and hermeneutics is receiver, as well as on that of the sender, involved with the task of recovering what the receiver being sometimes more, and the other wanted to say (or what a partic- sometimes less involved than the sender ular work really may be taken to mean). (See Fig. 4). Indeed, the first result of a In between the position of Ego and Alter, process of communication is to produce a semiotics has to elucidate what resources task of perception for the receiver, who has are at the disposal of both participants in to have the means of accomplishing this the process of communication. If rheto- task. We could start be separating the proc- ric, semiotics and hermeneutic are sister ess of communication into three acts which sciences, it makes sense, however, to ask do not have to take place at the same time semiotics to take care of the other sisters. A and/or in the same space: the act of pro- semiotically informed rhetoric would ask ducing the artefact, the act of presenting what means are at our disposal for gain- into to the receiver, and the act of receiv- ing the adherence of the other. Similarly, a ing it. The whole of this process supposes semiotically informed hermeneutics would there to be some pool of knowledge held in ask what recourses there are for construing common by the sender and the receiver, or the meaning of a certain sender and/or a rather, one pool of knowledge of the sender certain work. and one pool of knowledge of the receiver Could there then also be something which, even in the most favourable case, like a hermeneutically and/or rhetorically can only partly overlap. If the message is informed semiotics? I think not. As re- sender-adapted, it is the task of the receiver search traditions, hermeneutics and rheto- to recover the part of the message that is not given in his pool of knowledge. This is 23 The conception of communication pre- sented here is very much indepted to the theories a typically hermeneutic task. On the other of the Prague and Tartu schools of semiotics, in hand, if the message is receiver-adapter, it particular as discussed in Sonesson 1999b. More will be said about this in later lectures. 43 ric has a lot to teach semioticians. But they cal perspectives, and it is just one of the cannot redefine semiotics in an essential many enterprises which may be seen as oc- sense, in the way in which semiotics rede- cupying the space between the traditional fines hermeneutics and rhetoric. This is be- sciences. Rather, we suggested, semiotics cause, as against traditional hermeneutics must be considered to be a science in its (at least what is sometimes called the sub- own right, defined by a particular point of jectist tradition), semiotics observes that view, rather than a domain of reality. But nobody can be understood but through the after realising that a science is at least as semiotic resources offered by a given so- much a social institution as an epistemo- ciety (including the very structures of so- logical position, we finally decided that, at ciality). In the same way, semiotics points the time being, semiotics must be seen as a out, against rhetoric as it is often practiced research tradition, containing many issues nowadays (though it would concord, in worthy of discussion as well as a few provi- spirit at least, with the so much criticised sional solutions. As such it is in many ways rhetoric tradition of the 16th to the 19th comparable to cognitive science, although century, which was restricted to the tax- from the point of view of the sociology of onomy of expressive devices, or elocutio) sciences, cognitive science has had more that nobody can express himself, except by success. Cognition in the wide sense of means of the semiotic resources given in cognitive science appears to be co-exten- a particular society. This means that, not sive with meaning. A distinct advantage of only is communication only possible as an cognitive science is that it is by definition a interaction between sender and receiver, confluence of different research traditions, addresser and addressee, but that even this whereas semiotics has long been hampered interaction cannot take place but through by the autonomy postulate. On the other the intermediary of signs and other mean- hand, cognitive science stills seems to be ings. But these meanings are really part of stuck in an epistemological impasse, just as the world going beyond both addresser and semiotics was at the time of structuralism. addressee, the World 3 of Popperian ontol- If we look upon semiotics as a research trad- ogy: that which is only given through con- tion of meta-analysis putting the emphasis sciousness, but has an existence independ- on meaning, while cognitive science insists ent of consciousness. on cognitive operations, semiotics may also be compared to rhetoric and hermeneutics, Summary which look at different ends of the proc- The present section has been taken up by ess of communication, whereas semiotics a discussion about the nature of semiotics is concerned with the intermediary space. in general, as related to other disciplines. We may therefore learn a lot from herme- We pointed out that such familiar charac- neutics and rhetoric, if we integrate them terisations of semiotics as being a method, into a semiotical perspective, which insists a model, an interdisciplinary perspective, on all expression and all interpretation of or a philosophical movement could not be meaning only being able to take place by sustained. Semiotics itself comprehends means of the semiotic resources offered in many models, methods, and philosophi- a particular society and situation.

44 1.3. System and to which the object to be studied was the individual, purportedly unique, work of History: Beyond art. Many scholars have merely searched Barthes and Eco in for a practical way of mapping an indi- Pictorial semiotics vidual picture onto a verbal description, while retaining a minimum of confidence Pictorial semiotics involves, of course, the in the objectivity of the procedure. Al- study of pictures as particular vehicles of though some scholars developed models of signification. Pictorial semiotics, I submit, analysis which embodied hypotheses about is that part of the science of signification wide-ranging regularities found in pictori- that is particularly concerned to under- al semiosis, there has been little awareness, stand the nature and specificity of such until recently, that pictorial semiotics, if it meanings (or vehicles of meaning) that are is to be a part of general semiotics, must be colloquially identified by the term “pic- concerned with all kinds of pictures, and ture”. In other words, pictorial semiotics is formulate principles applicable to all em- the science of depiction, as a peculiar mode pirically occurring picture kinds, and even of information and communication. Thus, to all objects potentially recognisable as the purview of such a speciality must in- pictures. Such a conception, although ex- volve, at the very least, a demonstration of tended to the wider domain of visual semi- the semiotic character of pictures, as well osis, is implied (but not explicitly stated) in as a study of the peculiarities which differ- Saint-Martin’s (1987) recent work. Argu- entiate pictorial meanings from other kinds ments to the effect that pictorial semiotics of signification, and a assessment of the should be a general science of depiction, ways (from some or other point of view) or of visual images, are only presented in which pictorial meanings are apt to dif- in the recent books by Groupe µ (1992: fer from each other while still remaining 11ff), Sonesson (1989a:9ff), and O’Toole pictorial in kind. In differentiating pictorial (1994:169ff). meaning form other meanings, we should in fact be particularly interested in know- To elucidate the meaning of pictorial ing how they are distinguished from other semiosis must mean, among other things, kinds of visual signification, such as sculp- to find out in what respects pictures are ture, architecture, gesture, and even writ- like other signs, and how they differ from ing; or how they differ from other iconic them, most notably perhaps how they are signs, that its, from other signs motivated differentiated from other signs of such sign by similarity or identity. categories to which they undoubtedly be- long: the category of visual signs, and the The domain of pictorial category of iconic signs. Such as task, and semiotics even the very specificity of pictorial semi- otics, obviously dissolves itself if we accept In the work of the pioneers, pictorial semi- the idea of the Greimas school, according otics, even when it concerned itself with to which all meaning is of a kind, or is advertisement pictures, tended to make its identical in nature as far at it is pertinent to own the traditional conception of art his- tory and literary history alike, according 45 Fig. 5. Pan- zani pasta, analysed by Barthes

semiotic theory (cf. Floch 1986b).24 described as “socio-cultural”, but should Curiously, Floch (1984: 11, 1986a: instead attend to the minute details of an 12f) who defends this theory, also argues, individual picture. Groupe µ (1992:12) on the other extreme, that semiotics should follows suit in denying the pertinence of not concern itself with middle-range cat- these same categories, which the group egories like “photography” and “painting”, conceives of as being “sociological” or “in- stitutionalised”. Whatever the sociological 24 Thürlemann 1990, on the other hand, con- ceives of pictorial semiotics merely as an ancillary status of photography and painting, how- of art history. 46 ever, it seems to me that they are also, and picture has been curiously blended into a primarily, particular varieties of the picture publicity picture. A more abundant source sign, embodying a particular principle of of the rupture of the norm is, nevertheless, pertinence, which serves to rely expression the expectations, which we entertain that and content, and as such they should be of there will be certain correlations between interest to semiotic theory. categories of construction, categories of A division of the pictorial signs function and categories of circulation (or founded on everyday language may thus perhaps also categories of organisation). A result in four categories of picture catego- great part of Modernism (as well as Post- ries (Sonesson 1992a): construction cat- modernism) has consisted in breaking, in egories, defined by what is relevant in the ever new forms, with the prototype of the expression in relation to what is relevant art work that was current in the XIX centu- in the content, which, among others, dif- ry: an oil painting (construction category) ferentiates the photograph from the paint- with aesthetic function (function category) ing; function categories, that are divided that circulates through galleries, museums according to the social effects anticipated, and exhibition halls (circulation category). for example, the publicity picture which In this sense Modernism has been a gigan- has as its goal to sell products, the satirical tic rhetorical project, as Postmodernism picture which serves to ridicule somebody, was later to be. the pornographic picture, which is sup- However, even the very history of posed to produce sexual stimulation; and mass media and sign systems serves to the categories of circulation characterised undo the anticipated connections between by the channels through which the pictures pictorial kinds. This is valid also on a more circulate in a society, which makes the general level: xylography already implies bill-board into something different from that the pictorial sign stops being absolute- the newspaper picture or the postcard into ly bound to manual distribution; but only something different from poster; and, fi- the computerised picture consummates nally, organisation categories, which de- the rupture with a construction realised by pend on the conformation of the configura- hand. tion occupying the expression plane of the Pictorial semiotics, then, could well picture. be conceived as that particular branch of This is of course a primary source of semiotics which is concerned to determined visual rhetoric: by means of the mixture of in which way the picture sign is similar and different construction categories, function different from other signs and meanings, in categories, or circulation categories, a rup- particular as far as its relationship to other ture of our expectations is produced (Son- iconic and/or visual meanings are con- esson 1993;1994a; 1996a). Among well- cerned; and which is also called upon to known blends of construction categories analyse the systematic ways in which signs may be counted the Cubist collages, whose which are pictures may yet differ form each materials are heterogeneous. A mixture of other, thus, for instance, as to construction, function categories is present in the well- socially intended effects, channels of cir- known Benetton publicity, in which a news culation, and configurational kinds.

47 From Panzani pasta to the article marks a real breakthrough in pic- theory of iconicity torial semiotics. There could be some in- trinsic reasons for this, for the article may Although pictures are mentioned, and well constitute the first attempt to employ compared to verbal language, already by a simple model permitting to fix the recur- such precursors of semiotics as Lessing ring elements of pictorial signification. Yet and Degérando, and in spite of the fact that the importance of the work is mainly due Saussure, and even more Peirce, refer to to the influence it was to exercise on al- pictorial signs repeatedly, pictorial semiot- most all later analyses, either directly, the ics must be considered a recent discipline Barthesian terms being applied as a matter indeed: the Russian formalists have little to of course, or by way of reaction, when the say about pictures, and the Prague school authors took pains to dissociate their ap- merely uses them to illustrate general prin- proach from that of Barthes. ciples of semiosis. Only with the advent of French structuralism did a body of knowl- Not only did Barthes and his follow- edge particularly geared to the elucidation ers try to reduce all meaning to the lin- of general principles underlying the organ- guistic kind, employing a model inspired isation of the picture sign start to emerge. in structuralist linguistics, but in so doing, they unfortunately misunderstood the im- 25The history of pictorial semiot- port of most linguistic terms. What is con- ics begins with two false starts, which are fused in Barthes’ work tends to become nevertheless immensely important, since even more so in that of his followers, who, everything that has happened since has moreover, inherit his exclusive attention emerged from the criticism of those two in- to the content side of the pictorial sign, itial attempts: Barthes’ article on Panzani or more exactly, to the extra-signic refer- publicity, and Eco’s critique of iconicity. ent and its ideological implications in the First and foremost among the pio- real world, even to the point of ignoring neers of pictorial semiotics must be men- the way in which the latter are modulated tioned Roland Barthes, whose article “La within the sign. The socio-critical strand rhétorique de l’image”, stands at the origin of the Panzani analysis gave rise to several of two diverging developments within the national traditions, differently integrated semiotic field, the semiotics of publicity, with other scholarly conceptions, which represented by George Péninou and many have seemed to be fairly immune to later others; and the semiotics of visual art, rep- developments in semiotics, such as, in the resented by, among others, Louis Marin, sixties and seventies, the work of Hermann Hubert Damisch and Jean-Louis Schefer. Ehmer and others in Germany, that of Pe- In spite of the confusion to which Barthes ter Larsen in Denmark, and that of Gert testifies in his employment of linguistic Z. Nordström in Sweden, as well as, to terms, and although the usage to which he some extent, the more recent publications puts these terms is in itself incoherent, his by Gunther Kress, David Hodge, and Theo 25 What follows is only a historical over- van Leeuwen in Britain. The same errors of view. The theoretical problems raised by the dif- ferent models will be discussed in later lectures. It linguistic understanding are also found in is thus not necessary to understand all the concepts Damisch’s (1979) refutation of the linguis- and terms invoked. 48 tic model, identified with semiotics tout Actually, neither Barthes, nor his followers court, which, moreover, testifies to a much makes any real attempt to analyse the pic- more serious confusion in comparing the ture: they are discoursing all the time on merely intuitive, pre-theoretical notion of the referent, that is, on the depicted scene. the picture with the concept of language as Lindekens (1971) already recognised that a reconstructed in linguistic theory (just as “rhetoric of the referent”, not of the picture Metz did in the case of the notion of film; sign, was at stake in the Panzani article. cf. Sonesson 1989a,I.1.2.). Another fundamental parti pris of Starting out from a few general obser- the Panzani essay, which has left its im- vations, Barthes’ pioneering article rapidly print on pictorial semiotics, is the idea of turns into a regular text analysis concerned no picture being able to convey informa- with one particular photograph (Figure 5, tion by itself or, alternatively, containing defined both as to its means/ends category so much contradictory information that a (publicity) and, somewhat more loosely, its verbal message is needed to fix (or “an- channel division (magazine picture). The chor”) its meaning. No matter which in- photograph under analysis shows samples terpretation we choose (and the latter one of Panzani products, i.e. spaghetti, Italian may have more support in the text), picto- tomato sauce, and grated cheese, together rial meaning is supposed to depend on lin- with a selection of vegetables presented in guistic meaning. Pictures certainly offer a string bag, which is held up by an invis- much less linguistic information than ver- ible hand outside the picture. The brand bal texts, except in those cases in which the name is to be read on the Panzani products, picture itself contains the reproduction of and there is also a short text below the de- written messages, as is the case of the Pan- piction of the string bag. Barthes first com- zani publicity; but it could be argued that ments on the importance of the linguistic the picture much better conveys another part of the message, and then, in the main kind of information that resembles the one part of the essay, goes on to specify a se- present in the perceptual world (cf. Sones- ries of “connotations” supposedly appear- son 1989a: 114ff). ing partly in the verbal text and partly in The second most influential figure the picture. in early pictorial semiotics was no doubt It is here (as well as in Barthes 1961) Umberto Eco, who defined two of the ba- that Barthes proclaims his famous paradox, sis issues of the domain, and whose resolu- according to which the picture is a message tion of these issues was hardly contested deprived of a code. The term “image” in until recently. Probably because only con- fact alternates in the same paragraph with ventional signs, according to Saussure, the more particular term “photograph”, as were of interest to semiotics, Eco set out if this were the same thing, but later on the to show that pictures are as conventional photograph is opposed in this respect to as linguistic signs. In terms well-known the drawing. Yet many followers of Bar- within semiotics, Eco claimed that there thes retain the wider interpretation, using were no iconic signs, that is, no signs mo- it to defend the inanalysability, or ineffa- tivated by similarity. Pursuing even fur- bility, of paintings and other works of art. ther the analogy with linguistic signs, Eco

49 went on to suggest that pictures could be ing something similar to the conception of analysed into elementary signs, which, in the early Barthes, according to which the turn, could be dissolved into features hav- expression and content of (at least some) ing no meaning of their own. Although iconic signs are “tautologous”. Eco himself was to quality this latter idea Less influential than Barthes and ever more through the years, one or other Eco, but certainly as important for the version of his conception continues to be development of pictorial semiotics, René accepted by many scholars in the field. Eco Lindekens in his two early books (1971; himself thus ended up rejoining the argu- 1976) discusses questions pertaining to the ment of the philosopher Nelson Goodman basic structure of the pictorial sign (e.g., (1968), who thought pictures were similar conventionality and double articulation), to verbal language in being conventional, using photography as a privileged exam- but different from them in not being made ple. His theoretical baggage is complex: up of smaller units. Hjelmslevian semiotics, of which he has a During the late eighties and nineties, much more solid knowledge than Barthes, Eco’s conception of the picture sign was combined with an inkling of the Greimas heavily criticised by a number of schol- school approach; phenomenology, which, ars, notably by Paul Bouissac, Groupe µ, however, affected him in the subjectivist and the present author. My own argument, reinterpretation of the existentialists; and which relied both on evidence from per- the experimental psychology of perception, ceptual psychology, and on a refutation mainly derived from the Gestalt school. of logical arguments, showed that, quite Yet, the different theoretical strands re- contrary to what had been claimed by Eco main badly integrated, and much knowl- and Goodman, pictures were not funda- edge present in these perspectives is insuf- mentally conventional, whereas they were ficiently exploited (cf. Sonesson 1989a) indeed analysable into features, albeit of a In order to demonstrate the con- very different kind from those found in lin- ventionality of pictures, and to show how guistic signs. My conclusion was that there they are structured into binary features, were both a primary iconicity, found main- Lindekens (1971) suggests, on the basis of ly in pictures, in which it is the perception experimental facts and common sense, the of similarity between the item serving as existence of a primary photographic oppo- expression and the item serving as content sition between the shaded-off and the con- which is one of the condition for the postu- trasted; at the same time, he also turns to lation of the sign character of the sign, and experiments involving geometric drawings a secondary iconicity, in the case of which which have the function of brand marks, in it is our knowledge about the sign charac- order to discover the different plastic mean- ter which first permits us to discover the ings (which Lindekens calls “intra-iconic”) similarity between the two items involved of elementary shapes. In fact, Lindekens (Sonesson 1989a; 1994b; 1995a; 1997a; would seem to argue for the same conven- 1998b, c; 2000). Taking stock of this strain tionalist and structuralist thesis as the early of criticism, Eco (1997) in his latest books Eco (1968), but while the latter tends to ig- seems to pass to the other extreme, embrac- nore the photograph as the most embarrass- 50 ing counter-example, Lindekens attacks its linguistic and non-linguistic meanings im- frontally from the beginning. possible. Moreover, Floch and Thürlemann agree with other Greimaseans in taking all The emergence of pictorial knowledge about the object of study to be semiotics irrelevant to semiotics, so that they must In the late seventies and in the eighties, pic- refrain from using the knowledge base of, torial semiotics made something of a new for instance, perceptual psychology. start, or, rather, produced several fairly dif- The interest of this approach resides ferent, new beginnings: one, which is asso- not only in the fact that it involves the ap- ciated with the Greimas school, and whose plication of a model having fairly well-de- main representatives are Jean-Marie Floch fined terms, which, at least to some extent, and Felix Thürlemann, and more recently recur in a number of text analyses, but also also Jacques Fontanille; another, which is due to the capacity of this model to ac- comes out of the “general rhetoric” defend- count for at least some of the peculiarities ed even earlier by the Liège group known of pictorial discourse. Thus, for example, as “Groupe µ”; and, finally, a development Floch and Thürlemann have noted the pres- centred around Fernande Saint-Martin and ence of a double layer of signification in the her disciples in Montréal and Québec. To picture, termed the iconic and plastic levels. this could be added, following the distinc- On the iconic level the picture is supposed tion made by Saint-Martin and Carani, “the to stand for some object recognisable from Swedish” or “ecological school”, and also, the ordinary perceptual Lifeworld (which in my view, another recent tradition (with is of course a much more restricted notion two brands) rooted in the “social semiot- of iconicity than that found in the Peirce ics” of M.A.K. Halliday. tradition); while concurrently, on the plas- Jean-Marie Floch, Felix Thürle- tic level, simple qualities of the pictorial mann, and their followers accept the basic expression serve to convey abstract con- tenets of the Greimas school, and make use cepts. Floch, it is true, has tried to gener- of its abundant paraphernalia, albeit with alise these notions to other domains, most unusual restraint. Thus, like all contribu- notably to literature, but they seem much tions from the Greimasean camp, their ar- better adapted to pictorial discourse. ticles employ an array of terms taken over A second, more controversial aspect from the linguistic theories of Saussure, of, in particular, the work of Floch, is the Hjelmslev, Chomsky, and others, but given idea that pictorial meaning is organised into quite different meanings. The real problem contrasts, i.e. binary terms, one member of resulting from this approach, therefore, is which is an abstract property and the other not, as it is often claimed, that it deforms its opposite (“continuity” vs. “discontinui- pictures and other types of non-linguistic ty”, “dark colours” vs. “light colours”, etc.), meanings by treating them as being on a both of which are present in different parts par with language, but that, in attribut- of the same picture. Indeed, each analysis ing quite different significations to terms starts out from an intuitive division of the having their origin in linguistic theory, it picture into two parts, which may then be renders any serious comparison between repeated inside one or both the division

51 blocks. The remaining task of the analyst building-blocks of the theory. At this stage, is thereafter to justify this segmentation, Groupe µ seems heavily dependent on a set setting up long series of oppositional pairs, of Hjelmslevian concepts (which they may the members of which are located in the not interpret quite correctly; cf. Sonesson different division blocks resulting from the 1988,II.1.3.7., and 1989a,II.3-4.), as well as segmentation. Although Floch shows con- on the notion of isotopy as conceived by siderable ingenuity for discovering a bina- Greimas (which in itself may be incoher- ry division in all pictures studied, one may ent, cf. Sonesson 1988,II.1.3.5). wonder whether such an analysis is equally In spite of being general in import, adequate in all cases, and whether it re- the theory to begin with was mostly con- mains on the same level of abstraction. cerned with figures of rhetoric as they ap- Thürlemann appears to have been pear in verbal language. In a short study very little active in semiotics in recent of a coffee pot disguised as a cat, Groupe years, and Floch died a few years ago. µ (1976) tries to implement the theory also Jacques Fontanille, who now is the princi- in the pictorial domain. Over the years, the ple exponent of Greimasean pictorial semi- theory has been continuously remodelled, otics, has tried to introduce a phenomeno- so as to account better for the peculiarities logical tinge to the models inherited from of pictorial meaning. Recently, Groupe µ Greimas. The Greimas school is still very rhetoric appears to leave behind at least influential in France and Spain and, in par- part of the linguistic strait-jacket inherited ticular, in Latin America. from Hjelmslev, in order to incorporate “a Equally of seminal importance to certain amount of cognitivism”. Yet, the pictorial semiotics, the Groupe µ, or Liege theory still seems far from integrating the school has consisted of different members perceptual and sociocultural conditions through the years, the most constant of that constitute the foundations of all rhe- which are the linguists Jean-Marie Klinken- torical modulations. berg and Jacques Dubois, the chemist Fran- Like the Greimas school, Groupe cis Edeline and the aesthetician Philippe µ recognises the difference between the Minguet. Starting in the late sixties, this iconic and plastic layers of the picture sign Belgian group of scholars produced a book (again using a notion of iconicity which is of “general” rhetoric, in which they ana- much more restricted than that of Peirce). lysed in a novel way the “figures” appear- In this conception, iconic figures can be in- ing in the elaborate taxonomies of classical terpreted because of the redundancy of the rhetoric, using linguistic feature analysis iconic layer, and plastic figures acquires inspired in the work of Hjelmslev, as well their sense thanks to a corresponding re- as the mathematical theory of amounts. As dundancy of the plastic layer (thus, for in- in classical rhetoric, a figure is taken to ex- stance, we recognise the bottles substituted ist only to the extent that there is a devia- for the eyes of Captain Haddock as a figure, tion from a norm. The latter is understood because of the context of his body; and we as redundancy, and thus identified with identify the geometrical shape substituted the Greimasean concept of isotopy, which for the circle in one of Vasarely’s works, henceforth becomes one of the essential because of the environment of repeated cir- 52 cles). More recently, Groupe µ (1992) also is to have systematised a series of analyti- recognises iconico-plastic figures, which cal conceptions familiar from earlier art are produced in the plastic layers, while history and Gestalt psychology. the redundancy occurs in the iconic one, or Much of the importance of the Que- vice-versa (a comic strip personage which is bec schools resides in its explicit criticism like a human being but has blue skin would of the Greimasean approach, most clearly be of this kind, the bodily shape permitting spelled out by Marie Carani, who is also recognition while the blue colour creates the author of important studies concerned the deviation). Norms may be either gener- with pictorial abstraction and perspective, al, valid for all pictures, or local, if they are respectively (Carani 1987; 1988). As com- created in a particular picture in order to be pared to the binary opposition, which is the overturned: thus, the repetition of identical regulatory principle of the Greimas school geometrical shapes in Vasarely’s works is approach, as well as to the norm and its de- the backdrop on which another geometri- viations, which determines the conceptual cal shape stands out as a deviation. economy of Groupe µ rhetoric, the Que- Contrary to the Greimas school, bec school offers a much richer tool-kit of Groupe µ has never formed a closed move- conceptual paraphernalia, more obviously ment. Instead, the group has inspired iso- adapted to the analysis of visual phenom- lated followers in many parts of the world. ena. Yet this very richness also appears to The third conception of importance constitute the basic defect of the theory: it in the domain of pictorial semiotics is the is not clear whether it offers any restric- one propounded by Fernande Saint-Martin tions on what may be taken as relevant in and her collaborators, sometimes termed the picture sign, which means that no ana- the Quebec school. In a number of pub- lytical direction has been presented. lications (1985, 1987a), Saint-Martin has The constraints imposed by the grid been elaborating a theory of visual semi- taken oven from the linguistic theory of otics, which is based on the conviction M.A.K. Halliday by, notably, Michael that a picture, before being anything else, O’Toole (1994), are, in this respect, much is an object offered to the sense of visual more enlightening. According to this con- perception. Visual meaning, according to ception, every work realises some alter- this conception, is analysable into six vari- native from among the ideational, inter- ables, equivalent to a set of dimensions on personal and textual “macro-functions”, which every surface point must evince a renamed by O’Toole the representational, value: colour/tonality, texture, dimension/ modal and compositional functions. The quantity, implantation into the plane, ori- first function is involved with the relation- entation/vectorality, and frontiers/contours ships between the participants and process- generating shapes. The surface points, es in the real world, the second concerns the specified for all these values, combine with way in which this world is presented by the each other, according to certain principles, creator of the sign, and the third has to do notably those of topology, and those of with rules of internal patterning applying Gestalt theory (cf. Saint-Martin 1980 and to the work as such. It is not clear why the 1990). The principle merit of this approach functions are given other names, if they are

53 really analogous to the functions Halliday commentators, to the “Swedish school” finds realised in language; and indeed, one (Saint-Martin 1994) or the “ecological may doubt that they actually correspond school” (Carani 1998). In my main work to these function in any very interesting (Sonesson 1989a), which is devoted to an sense. It is also very unfortunate that, in critical review of earlier accomplishments trying to specify by means of a cross-clas- in pictorial semiotics, I emphasise the basi- sification the different options available for cally perceptual nature of the picture sign, the realisation of the different functions, and expound some of the consequences of O´Toole often employs traditional art-his- this observation, invoking the testimony of torical terms, which are notoriously vague contemporary perceptual psychology, and and ambiguous, without O´Toole offering of philosophical and phenomenological any specification of his own. theories of perception. Contrary to, most The pictorial semiotics proposed by notably the Greimas school, “the ecological Hodge & Kress (1988) and later by Kress school” thus shuns the autonomy postulate & van Leeuwen (1996) also refers back to of semiotics, admitting that pictorial semi- the linguistic theories of Halliday, but, in otics has a lot to learn from psychology and this respect truer to their master, they put other sciences, while claiming that their much more emphasis the social framework results must be inserted into a specifically of picture use, and thus on pictures used semiotic framework, which has evolved in society generally, more than in the art from the age-old tradition of this science. world. In the first book, the invocation of Critically reviewing the use of many lin- Halliday seems largely vacuous: the book guistic and otherwise semiotic concepts, is really inspired by Barthes’ Panzani anal- such as sign, feature, connotation, iconic- ysis, and shares all its defects. The second ity, and so on, I argue that these are use- work, however, is interesting for having re- ful only to the extent that their import are course to other aspects of Halliday’s work, clearly spelled out, so that the specificity of his semantic (as opposed to syntactic) pictorial meaning can emerge. This work analysis of sentence structure, thus point- has later been extended, by myself as well ing out parallels, as well as their absence, as by some students and collaborators, to between linguistic and pictorial organisa- pictorial rhetoric, photographic semiotics, tion, without necessary falling pray to the cultural semiotics, and much else. In the customary linguistic imperialism. In the following lectures, I will have occasion to end, however, it seems to me that Kress & go deeper into this work, as well as into the van Leeuwen fail to make use of the most lessons offered by the other schools of pic- interesting contribution offered by Hal- torial semiotics, when critically acknowl- liday, the analysis of thematic structure, edged.26 and thus are unable to notice the really im- portance differences between pictures and Some contributions from verbal signs. In this context, my own contribu- 26 Nothing have been said here about the orthodox Peirceans, who only recently seem to tions to pictorial semiotics constitute a take an interest in pictures, mostly, however, in the forth (or fifth) strand, attributed, by certain spirit of a simple application of Peircean catego- ries to a new domain of reality, pictures. 54 philosophy and psychology torial consciousness” elaborated by Hus- At is present stage pictorial semiotics may serl. Both contribute to our understanding well have less in common with Barthes’ of the specificity of the picture sign within Panzani analysis than with that “linguis- the general category of iconicity. tics of the visual image” invoked by the art Within perceptual psychology, Ge- historian E.H. Gombrich, or that “science stalt psychology has often had recourse to of depiction” called for by the psychologist pictures in order to illustrate certain gen- James Gibson; as well as with the studies eral principles of perception, and the same of pictorial meaning initiated in philosophy is true of many social constructivists, no- by, for instance, Edmund Husserl, Ludwig table with reference to so-called “impos- Wittgenstein, Nelson Goodman, and Rich- sible pictures” (similar to the work of Es- ard Wollheim. The most relevant reference, cher and Reutersvärd). Only James Gibson however, as will be seen in later lectures, realised the necessity of a particular study may well be that to Gibson, who, together of pictorial perception, originally because with such disciples and colleagues as Julian he wanted to protect ordinary perception Hochberg, John Kennedy, and Margaret from the abusive generalisations suggested Hagen, has started to elaborate a psychol- by pictorial examples. To perceive a pic- ogy of picture perception – but psycholin- ture is very different from the perception guistics cannot do without linguistics, and, of the real, three-dimension world, already by the same token, we need to establish a because the former is actually a surface, more general, theoretical, framework for masquerading as part of the world of our the study of the picture sign (cf. Sonesson experience. The work of Gibson, Kennedy, 1989a). and Hochberg has been very important in From the point of view of semiotics, pointing out the particularities of the pic- the problem with the work of Goodman and ture as a sign. his followers is that it is explicitly norma- Contrary to the other research tradi- tive: not only does Goodman not attend to tions in pictorial semiotics, the ecological the common sense notion of picture, which school does not defend any autonomy of he claims is contradictory, instead of which semiotic knowledge. In the following lec- he introduces his own definition, but he tures, we shall therefore delve deeper into compares pictures, not with real-world lan- the heritage of those philosophers and psy- guage, but with the logically reconstructed chologists who have taken an interest in language of analytical philosophy. Philoso- pictures. But we will replace their findings phers of the phenomenological school, as within a specifically semiotic framework. well as some ordinary language philoso- phers have made more directly relevant Summary contributions, since they start out from our In the light of the earlier discussion of sem- experience of the common sense world. iotics general, we first studied the special- As we shall see, Wollheim’s characterisa- ity of pictorial semiotics, claiming that it tion of the picture as being a kind of “see- must be concerned with the specificity of ing-in”, inspired by Wittgenstein’s work, is the picture sign, which involves relating it very similar to the earlier analyses of “pic- to those higher order categories of which it 55 is a part, such as meanings, signs, iconical University Press. signs, visual signs, and so on, as well as Barthes, Roland (1961): Le message photographique, in Communications 1,. scrutinising the different categories which 128-130. it contains, such as construction types, – (1964a). Éléments de sémiologie, function types, circulation types, and or- Communications, 4, 91-135. Also in Barthes, Roland, Le degré zéro de l’écriture/ ganisation types. After referring to the pio- Éléments de sémiologie, Paris: Gonthier. neering work of Barthes and Eco, which – (1964b). Rhétorique de l’image, has been important mainly because of the Communications, 4, 40-51. Also in Barthes, critical reactions which they have spurred, Roland, L’obvie et l’obtus., 25-42. Paris: Seuil 1982. this section characterised the most impor- Bermúdez, José Luis (2005) Philosophy tant research traditions within contempo- of psychology. New York & London: rary pictorial semiotics, such as the Grei- Routledge. Boudon, Pierre, (1979): Logique spéculaire et mas school, the Liege school, the Quebec iconicité: Les Ménines, i Canadian Journal school, two approaches developed out of of research in semiotics, VI,3VII,1, ss. 43- the linguistics of Halliday, and, finally, 55. the ecological school, associated with the Bouissac, Paul (1997) New epistemological perspectives for the archaeology of writing. work of the present author. We also noted In Archaeology and Language I, Blench, the important contributions of some phi- Roger, & Spriggs, Matthew, (eds.), 53-62. losophers and psychologists, who are not London & New York: Routledge. – (1998) Converging Parallels. Semiotics and normally identified as semioticians, but to Psychology in evolutinary perspective. In which we will turn more fully in the fol- Theory & Psychology, 8, 6, 731-753. lowing lectures. This historical parts thus – (1999a) Semiotics and the science of finishes the first lecture, in which we have memory. Paper presented at the conference on Semiotics and the European Heritage, been involved with a characterisation of Dresden, February 1999. semiotics in general, and pictorial semiot- – (1999b) Le savoir sémiotique. In Degrés, ics in particular, first with the help of some Vingt-septième anée, no 100, c1-c15. Brentano, Franz (1885) Psychologie von simple pictorial examples, with led on to empirischen Standpunkt. Leipzig: Mener. an epistemological discussion, and finally Carani, Marie (1987) Sémiotique de to this last historical part. l’abstraction picturale, in Semiotica , 67:12, 1-38. – (1988) Sémiotique de la perspective picturale, in Protée, 16:1-2, 171-181. – (1998) Bilan des sémiotiques visuelles References contemporaines. Introduction intempestive Andrén, Anders (1997) Mellan ting och au débat organisé sur les sémiotiques text. Stockholm/Stehag: Brutus Östlings visuelles, Séminaire intersémiotique de Bokförlag Symposion. Paris (Jacques Fontanille et Denis Bertrand Angenot, Marc (1985). Critique de la raison resp.), Institut Universitaire de France, 18 sémiotique. Montréal: Presses de l’Université décembre 1998. Available online at http:// de Montréal (English translation: Critique www.fl.ulaval.ca/hst/visio/bilan.htm of semiotic reason. New York, Ottawa & Conley, Thomas (1990) Rhetoric in the Toronto: Legas. 1994) European tradition. Chicago: Univerist of Bahn, Paul G. (1998) The Cambridge Illustrated Chaicago Press. History of Prehistoric Art. Cambridge: Damisch, Huber (1979) Sur la sémiologie Cambridge University Press. de la peinture. In A semiotic landscape/ – ed. (1996) The Cambridge Illustrated History Panorama sémiotique. Actes du premier of Archaeology. Cambridge: Cambridge 56 congrès de l’association internationale de Floch, Jean-Marie (1978). Quelques positions sémiotique, Milan, juin 1974. Chatman, S, pour une sémiotique visuelle/Roland Eco, U, & Klinkenberg, Jean-Marie, éds; Barthes: Sémiotique de l’image, Bulletin du 128-136. The Hague & Paris: Mouton. Groupe de recherches sémio-linguistiques, Deacon, T. (1997). The Symbolic Species: The 4/5, 1-16, 27-32. Co Evolution of Language and the Brain. – (1984). Petites mythologies de l’œil et NY: Norton. l’esprit. Paris: Hadès. Deely, John (1982). Introducing semiotic. Its – (1986a). /entries in/ Sémiotique. Dictionnaire history and doctrine. Bloomington: Indiana raisonné de la théorie du langage. Tome II. University Press. A.J. Greimas, & J. Courtès, J., (eds.). Paris: - (2001) Four ages of understanding. Toronto: Hachette. University of Toronto Press. – (1986b). Les formes de l’empreinte. Donald, M. (1991). Origins of the Modern Périgueux: Pierre Fanlac. Mind. Three Stages in the Evolution of – (1990). Sémiotique, marketing et Culture and Cognition, Cambridge, Mass.: communication. Sous les signes, les Harvard University Press. stratégies. Paris: PUF. Edelmann, Gerald (1992): Bright Air, Brilliant Foucault, Michel, (1966): Les mots et les Fire. On the Matter of the Mind. London: choses. Paris: Gallimard. Penguin Books. Gallagher, Shaun (2005) How the body shapes – & Guilio Tonini (2000) A Universe of the mind. Oxford: Clarendon. Consciousness. New York: Basic Books. Gibson, James (1966). The senses considered Eco, Umberto (1968). La struttura assente. as perceptual systems. Bonston: Houghton Milan: Bompiani. Swedish translation: Den Mifflin Co. frånvarande strukturen. Lund: Cavefors – (1971). The information available in pictures, 1971. Leonardo, 4:1, 27-35. – (1976). A theory of semiotics. Bloomington: – (1978). The ecological approach to visual Indiana University Press. perception of pictures, Leonardo, 4:2: 227- – (1977) The influence of Roman Jakobson 235. on the development of semiotics, in D. – (1980). A prefatory essay on the perception of Armstrong, and C. H. van Schooneveld surfaces versus the perception of markings (eds.), Roman Jakobson: Echoes of his on a surface. In The perception of pictures, scholarship, 39-58. Lisse. Peter de Ridder. Volume I: Alberti’s window., Margaret – (1984). Semiotics and the philosophy of Hagen, (ed.), xi-xvii. New York: Academic language. Bloomington: Indiana University Press. Press. – (1982). Reasons for realism. Selected – (1988) De los espejos y otros ensayos. essays of James J. Gibson, Edward Reed, Barcelona. Lumen. [Spanish translation of & Rebecca Jones, (eds.). Hillsdale, New Suglie specchi e altri saggi. Milan: Fabri Jersey, & London: Lawrence Erlbaum Ass. 1985]. Gombrich, E.H. (1960). Art and Illusion. – (1992) Los limites de la interpretación. London: Phaidon Press. Second edition Mexico: Lumen. [Spanish translation of 1962. I limiti dell’intepretazione. Milan: Fabri Goodman, Nelson (1968). Languages of art. 1990]. London: Oxford University Press. – (1998) Réflexions à propos du débat sur Greimas, A.J. (1970). Du sens. Paris: Seuil. l’iconisme (1968-1998). In Visio, 3, 1, 9- Groupe µ (Dubois, J., Dubois, Ph., Edeline, Fr., 32. Klinkenberg, J.M., Minguet, Ph, Pire, Fr., – (1999) Kant and the platypus. New York: & Trinon, H.) (1970). Rhétorique générale. Harcourt Brance & Co.. Paris: Larousse. Fagan, Brian M. (1998) Peoples of the Earth. – (1976). La chafetière est sur la table, An Introduction to world prehistory. 9th Communications et langages, 29, 36-49. edition. New York: Longman – (1977). Rhétorique de la poésie. Bruxelles: Ferris, Maurizio (2002) Historia de la Editions Complexe. hermenéutica. Mexico: Sigle XXI editores – (1978). Douce bribes pour décoller en 40 000 (Italian original 1988). signes, Revue d’ésthétique, 3-4, 11-41.

57 – (1979) Iconique et plastique: sur un structures. In The perception of pictures, fondement de la rhétorique visuelle, Revue Volume II: Dürer’s devices, Margaret d’ésthétique, 1-2, 173-192. Hagen, (ed.), 47-93. New York: Academic – (1980). Plan d’un rhétorique de l’image, Press. Kodikas/Code, 3, 249-268 Hodge, R, & Kress, G (1988) Social semiotics. – (1985). Structure et rhétorique du signe Cambidge: Polity Press. iconique. In Exigences et perspective de Hooder, Ian (1992) Reading the past. Second la sémiotique. Recueil d’hommages pour edition. Cambridge: Cambridge University A.J. Greimas, H. Parret, & H.G. Ruprecht, Press. (eds.), Volume I, 449-462. Amsterdam: – ed. (1996) On the surface. Çatalhöyük 1993- Benjamins Publishing Co. 95. London: British Institute of Archeology – (1992). Traité du signe visuel, Pour une at Ankara. rhétorique de l’image. Paris: Seuil Husserl, Edmund, (1962) Phänomenologische Gurwitsch, Aron (1957) Théorie du champ de Psychologie. Husserliana IX. The Hague: la conscience. Bruges: Desclée de Brouver. Nijhoff – (1964) The Field of Consciousness. Jakobson, Roman (1960) “Linguistics and Pittsburgh: Duquesne University Press. poetics”, in: Sebeok, Thomas, (ed.), Style – (1966) Studies in Phenomenology and in language, 350-377. Cambridge, Mass.: Psychology. Evanston: Northwestern MIT Press. University Press. Kennedy, John M. (1974). A psychology of – (1974) Phenomenology and the Theory of picture perception. San Francisco: Jossey- Science. Evanston: Northwestern University Bas, Inc. Press. Kress, Gunther, & van Leeuwen, Theo van – (1977) Human encounters in the social world. (1996), Reading images. The grammar Pittsburgh: Duquesne University Press. of visual design. London & New York: – (1985) Marginal Consciousness. Athens, Routledge. Ohio: Ohio University Press. Larsen, Peter, (1976) Billedanalyse, in Halliday, Michael A.K. (1967-68): ”Notes Massekommunikation, Olivarius, P., on transitivity and theme in English”, in Rasmussen, O., & Rugholm, P., (eds.), 71- Journal of linguistics, III:1, ss. 37-81; 118. Copenhagen: Dansklærerføreningen. III:2, ss. 199-244; IV:2, ss. 179-215. Lindekens, René (1971). Eléments pour une Hagen, Margaret (1979). A new theory of sémiotique de la photographie. Paris & the psychology of representational art. In Bruxelles: Didier/Aimav. Perception and pictorial representation, C. – (1976). Eléments de sémiotique visuelle. Nodine, & D. Fisher, (eds.), 196-212. New Paris: Klincksieck. York: Praeger Publishers. Lombardo, Thomas (1987). The reciprocity of – (1980). Generative theory: a perceptual perceiver and environment. Hilldale, New theory of pictorial representation. In The Jersy, & London: Lawrence Erlbaum Ass. perception of pictures, Volume II: Dürer’s Lotman, J.M., (1990): Universe of mind. A devices, Margaret Hagen, (ed.), 3-46. New semiotic theory of culture. London & New York: Academic Press. York: Taurs & Co, Ltd. Hermerén, Göran (1983). Aspects of aesthetics. Mallery, Garrick, (1880-81): Introduction to Lund: Gleerups. the study of sign language among the North Hjelmslev, Louis (1959). Essais linguistiques. American Indians and A collection of Copenhagen. Second edition: Paris, Minuit gesture signs. In Aboriginal sign languages 1971. of the Americas and Australia I., Umiker- Hochberg, Julian (1972). The representation of Sebeok, J., & Sebeok, Th., (eds.), 1-76; 77- things and people. In Art, perception, and 406. New York & London: Plenum Press reality, E.H. Gombrich et al., (eds.), 47- 1978. 94. Baltimore and London: John Hopkins – (1881). Sign language among North American University Press. Indians. Reprinted: The Hague & Paris: – (1978). Perception. Second, revised edition. Mouton 1972. Englewood Cliffs, Prentice Hall, Inc. Marin, Louis, (1971) Etudes sémiologiques. – (1980). Pictorial functions and perceptual Paris: Klincksieck.

58 Neisser, Ulric (1976). Cognition and reality. constitution du champ textuel, Protée 16, San Fransisco, Freeman & Co. no 1-2, 202-213. – (1987). From direct perception to conceptual – (1990). La théorie de la Gestalt et l’art structure. In Concepts and conceptual visual. Québec: Presse de l’Université du development: ecological and intellectual Québec. factors in categorization, Ulric Neisser, – (1994) Avant-propos: pour en finir avec la (ed.), 11-24. Cambridge & London: mutité de la peinture. In Nouveaux Actes Cambridge University Press. Sémiotiques, 34, 35, 36, 1-3- O’Toole, Michael, (1994) The language Saussure, F. (1968/74). Cours de linguistique of displayed art. London: Leicester générale. Edition critique par Rudolf Engler. University Press. Tome 1-II. Harrossowitz, Wiesbaden. Parmentier, Richard J. (1985). Signs’ Schaeffer, Jean-Marie (1987). L’image place in Medias Res: Peirce’s concept précaire, Paris: Seuil. of semiotic mediation. In Semiotic Searle, John, (1980): ”Las Meninas” and the mediation. Sociocultural and psychological paradoxes of pictorial representation, i perspectives, R. Mertz, & R. Parmentier, Critical Inquiry 6:3: The language of (eds.), 23-48. New York & London: images, ed. by W.J.T. Mitchell; ss.477-497 Academic Press, Inc. Sieveking, Ann (1979). The cave artists. Peirce, Charles Sanders, (1931-58) Collected London: Thames & Hudson Ltd. Papers I-VIII. Hartshorne, C, Weiss, P, Scruton, Roger (1974). Art and imagination. & Burks, A, (eds.). Cambridge, Mass. London: Meuthen & Co. Harvard University Press. Sonesson, Göran (1988). Methods and models Pinter, Steven (1997) How the mind works. in pictorial semiotics. Report 3 from the New York: Norton. Semiotics project. Lund: Institute of Art – (2002) The blank slate. London: Penguin History (mimeographed). Prieto, Luis (1966). Messages et signaux. – (1989a). Pictorial concepts. Inquiries into Paris: PUF. the semiotic heritage and its relevance for – (1975a). Pertinence et pratique. Essai de the analysis of the visual world. Lund: sémiologie. Paris: Minuit. Aris/Lund University Press. – (1975b). Essai de linguistique et sémiologie – (1989b). Semiotics of photography. On générale. Genève: Droz. tracing the index. Report 4 from the Reed, Edward S. (1988). James J. Gibson and Semiotics project. Lund: Institute of Art the psychology of perception. New Haven History (mimeographed). & London, Yale University Press. – (1992a) Bildbetydelser. Inledning till Renfrew, Colin (1982). Toward an archaeology bildsemiotiken som vetenskap. Lund: of mind. Cambridge: Cambridge University Studentlitteratur. Press. – (1992b). The semiotic function and the genesis Renfrew, Colin, & Bahn, Paul, (1991). of pictorial meaning. In Center/Periphery Archaeology. Theories, methods, and in representations and institutions. practice. London: Thames and Hudson Proceedings from the Conference of The Ltd. International Semiotics Institute, Imatra, – (2000). Archaeology. Theories, methods, and Finland, July 16-21, 1990., Tarasti, Eero practice. Third edition. London: Thames (ed.), 211-156. Imatra: Acta Semiotica and Hudson Ltd. Fennica. Ricœur, Paul, (1975) La métaphore vive. Paris: – (1992c). Comment le sens vient aux images. Seuil. Un autre discours de la méthode. In De Rudgley, Richard (1998) Lost civilisations of l’histoire de l’art à la sémiotique visuelle. the Stone age. London. Century. Carani, Marie(ed.), 29-84. Québéc: Les Saint-Martin, Fernande (1980) Les fondements éditions du Septentrion/CÉLAT. topologiques de la peinture. Montréal: – (1993a). Pictorial semiotics, Gestalt Hurtubise. New edition 1989. psychology, and the ecology of perception. – (1987). Sémiologie du langage visuel. Review of Saint-Martin, Fernande, La Québec: Presse de l’Université du Québec. théorie de la Gestalt et l’art visuel. In – (1988). De la fonction perceptive dans la Semiotica 99: 3/4, 319-399.

59 – (1993b) Die Semiotik des Bildes, Zeitschrift From mechanical reproduction to digital für Semiotik 15: 1-2, 131-164. production. In Visio 4, 1, 11-36. – (1994a). Prolegomena to the semiotic analysis – (1999b) The life of signs in society – and out of prehistoric visual displays. In Semiotica, of it. In Σηµειωτικη, Sign System Studies, 100: 3/, 267-332. 47-64. - 27, 88-127. – (1994b) Sémiotique visuelle et écologie – (2000a) Iconicity in the ecology of semiosis, sémiotique, in RSSI, 14, 1-2, printemps In Iconicity - A Fundamental Problem in 1994, pp. 31-48. 47-64. Semiotics. Troels Deng Johansson, Martin – (1995a) On pictorality. The impact of the Skov & Berit Brogaard (eds.), 59-80. perceptual model in the development of Aarhus: NSU Press. visual semiotics, in The semiotic web – (2001a) De l’iconicité de l’image à l’iconicité 1992/93: Advances in visual semiotics, des gestes, Actes du congrès ORAGE 2001 Sebeok, Th., & Umiker-Sebeok, J., eds., ORAlité et Gestualité, Aix-en-Provence, 67-108. Mouton de Gruyter, Berlin & New Juin 18-22, 2001. Cave, Ch., Guaïtella, I, & York. Santi, S. (eds), 47-55. Paris : L’Harmattan. – (1995b) Livsvärldens mediering. – (2001b) La retorica del mondo della vita. In Kommunikation i en kultursemiotisk Modo dell’immagine, Pierluigi Basso (ed), ram. In Mediatexter och mediatolkningar. 93-112. Bologna: Esculapio. Holmberg, Claes-Göran, & Svensson, Jan, Thürlemann, Félix (1982). Paul Klee. Analyse (eds), 33-78. Nora: Nya Doxa. sémiotique de trois peintures. Lausanne: – (1996a) An essay concerning images. From L’Age d’Hommes. rhetoric to semiotics by way of ecological – (1990). Vom Bild sum Raum. Beiträge zu physics. Review of Groupe µ, Traité du einer semiotischen Kunstwissenschaft. signe visuel. Semiotica, 109-1/2, 41-140. Köln. DuMont. – (1996b) Le silence parlant des images, Protée Tomasello, M. (1999). The Cultural Origins 24:1, 37-46. of Human Cognition, Cambridge, MA: – (1996c). De la retórica de la percepción a la Harvard University Press. retórica de la cultura/Från varseblivningens Trigger, Bruce G. (1989) A history of till kulturens retorik, in Heterogénesis 15, archaeological thought. Cambridge: 1-12. Cambridge University Press. – (1996d) Iconicité de l’image – Imaginaire Uspenskij, Boris, A, (1976a): The semiotics of de l’iconicité. De la ressemblance à la the Russian icon. Lisse de Ridder Press. vraisemblance. Actes du Premier Congrès de – (1976b): La ”droite” et la ”gauche” dans l´art l’Association internationale de sémiologie des icônes, i Lotman, Y.M., & Ouspenski, de l’image, Blois, novembre de 1990. In [Uspenskij] B.A., eds., Travaux sur les Visio, 1, 1, pp. 23-34. systèmes de signes. Bruxelles: Editions – (1997a) Approaches to the Lifeworld core of Complexe; ss. 168-174. Också på engelska visual rhetoric. Visio 1, 3, 49-76. i Semiotica 13, 1975, ss. 33ff . – (1997b) The ecological foundations of – (1976c): The language of ancient painting, i iconicity, in Semiotics Around the World: Dispositio. Revista hispanica de semiótica Synthesis in Diversity. Proceedings of literaria 1, 3, ss. 219-246 the Fifth International Congress of the Vanlier, Henri (1983). Philosophie de la IASS, Berkeley, June 12-18, 1994. Rauch, photographie, Laplume: Les cahiers de la Irmgard, & Carr, Gerald F., eds., 739-742. photographie. Berlin & New York: Mouton de Gruyter. Zlatev, Jordan (2002) “Mimesis: The ‘missing – (1998a) Metonymy, in Encyclopaedia of link’ between signals and symbols in Semiotics. Bouissac, Paul, ed., New York & phylogeny and ontogeny: In Mimesis, sign, London: Oxford University Press. and the evolution of language. Pajunen, – (1998b) That there are many kinds of pictorial Anneli (ed.), 93-122. Turku: University of signs, Visio, 3, 1, 33-54. Turku. – (1998c) The concept of text in cultural — (2003) Meaning = Life (+ Culture). An semiotics, in Σηµειωτικη,, Sign System outline of a unified biocultural theory of Studies, 26, 83-114. meaning. Evolution of Communication, – (1999a) Post-photography and beyond. 4/2: 253-296.

60