Lecture 2: the Psychology and Archaeology of Semiosis. Contents

Lecture 2: The Psychology and Archaeology of Semiosis. Pictorality as a Semiotic Function

In this lecture, we will discuss the emergence of the semiotic function, both ontogenetically and phylogenetically, and we will consider the part played by the picture sign in this development. In order to demonstrate that pictures are indeed signs, we will explore the basic elements of the sign presupposed but never put into focus neither by Saussure nor by Peirce. Indeed, explorations in the psychology and phenomenology of perception will turn out to be necessary, in order to characterise the sign in opposition to more elementary meanings, such as those given to us in the common sense world, variously characterized as the “lifeworld”, the “natural world”, or the world of “ecological physics”.

Contents 2.1. The emergence of the picture sign – individual and generic histories 4 The original picture interpretation situation: epitomic aspects 4 Further views from the playground: pigeons, apes and men 9 Pictures in the sand: Anati’s “prayer” and the native eye 13 Terrestial and Interstellar Archaeology 20 Summary 34

2.2. Signs in the Human Lifeworld 34 Two classical views of the sign – and beyond 37 From pebbles to feathers. The notion of differentiation 42 Different ways of “connecting significations” 49 Signs and mediations : The Fonseca-Peirce connection 53

Summary 58

2.3. Meaning before the sign: the commens of perception 59 The ecology taken for granted: the Lifeworld 59 The affordances of a game of chess 64 Von Uexküll on how it feels to be a tick 70 From Umwelt to Lebenswelt : the thematic field 72 Common sense psychology and intentionality 80 Summary 84

2.4. Steps to the cultural world 85 Signs as portable memory 86 The schemes of perception and memory 91 Collective memory and the “tragedy of culture” 95 From mimesis to theory in evolution and development 99 The three world circulations: Mates, money and love 108 Two types of types, with iconicity and mimesis 115 Summary 118

References 118

Semiotics of pictures, it might be argued, can only exist if the picture is indeed a sign. There have been those who have denied this; others, instead of demonstrat- ing the proposition, have simply taken it for granted. Even if somebody would take upon himself the task of showing that the picture is a sign, it would naturally be suspected that the assignment must be different for those who follow the Peircean and the Saussurean tradition. In fact, in both conceptions, the notion of sign is really basically taken for granted instead of being defined, and thus there is no non-arbitrary way in which the task may be accomplished. A more explicit concept of sign is needed in order to begin answering the question. Some elements of such a definition may be gathered from the notion of semiotic function charactised by Jean Piaget; others, as we shall see, can be bor- rowed from the phenomenology of Edmund Husserl. This is a concept of sign which supposes there to be other meanings than signs – more elementary meanings, such as those given in ordinary perception. In this sense, the domain of semiotics is wider than the sign: it is some more general property which might be described as “meaning”. There could thus be a semiotics of pictures even if pictures were no signs. However, the present lecture is designed to show that the picture must indeed be a sign, in the precise sense which we are going to introduce. In order to do so, we will have to attend to the place of the picture in the development of the semiotic function. There is of course no real evidence in phylogeny, except for the indirect way of comparing human beings with other animals; and thus the facts have to be searched out in child development as well as in the comparison between cultures.

2.1. The emergence of the picture sign – individual and generic histories There can be no doubt that the ability to interpret pictures is as unique a property of human beings as is language. However, it is normally taken for granted that the picture sign is more simple, at least in the sense of being evolutionary older, than language. Thus, for instance, those who have tried to teach language to apes have had recourse, at a preparatory stage, to the mediation of pictures. However, there are now reasons to think that, at least in some respects, the picture sign is more complex than language – it appears, it seems, later in ontogeny, if not also in phylogeny. James Gibson (1971, 1980) has claimed that, while all animals perceive surfaces, only humans are able to see surfaces as having reference. In other words, pictures have ”referential meaning”; they contain invariants for surfaces but also for the objects referred to. Gibson thus appears to have a somewhat implicit concept of the picture as being a sign. Julian Hochberg showed that a child 19 months old who had never seen a picture could readily interpret it if he/she were familiar with the objects depicted (Hochberg & Brooks 1962). But Hoch- berg did not investigate whether the child saw the picture as a picture or as an instance of the category of the depicted object — a picture of a bird as a bird, etc. For the picture to be a sign, both similarity and difference have to be involved.

Sonesson 1989a used this as an argument (together with logical ones) against the conventionalist critique of iconicity formulated by Eco, Goodman, and others. Cf. Lecture 3.

The original picture (1972:70f) himself observes that interpretation situation: there either must be an innate capac- epitomic aspects ity for interpreting pictures, or that According to a famous anecdote such an ability must develop at an kwown to us from the historian early stage, and then not from pic- Herodotus (1954: 102f), the pharao torial experience itself, but from the Psammeticus designed an ingenious ordinary experience of the world. experiment to find out which lan- This result, and Hochberg’s guage were the original tongue of conclusions, are remarkable. To be- humankind: he took two newly born gin with the former, it is obviously infants from an ordinary family and imcompatible with any theory, such had them brought up under strict or- as that of Goodman or Eco, accord- ders that no-one should utter a word ing to which a picture acquires its in their presence. As a result, Hero- meaning simply by being “appoint- dotus (himself originally from Phry- ed” to be the sign of an object (as not- gia) reports, Phrygian was found to ed in Hochberg 1978b:235ff). What be the original language. is interesting about Hochberg’s con- Two and a half centuries af- clusions is that the most “obvious” ter the time of Herodotus, the same alternative is not even considered, type of experiment was at last car- i.e. that no interpretative capacity at ried out in the study of pictures. all would be needed, because the ob- But Hochberg & Brooks, who per- ject and its picture are simply “simi- formed this experiment, were not in- lar”. But of course this is no serious tent on finding out in which style the alternative since there is no similari- child would execute its first drawing ty between the picture and its object, if left alone (if indeed there would except from the point of view of a ever be any; cf. Gardner 1980); in- very superficial phenomenology. If stead, their experiment bore on the lines on paper are taken as equiva- interpretative capacities of the child. lent to the edges of the object, Hoch- Thus, they raised a child to the age berg (1978b:236) notes elsewhere, of 19 months, impeding it from hav- this is a fact about the viewer, not ing other than incidental experience about the light at the eye. of pictures, and then exposed it to Apart from the observations on outline drawings and later to photo- children, research into origins has graphs of objects with which it was always employed a second kind of already acquainted, finding that the investigation, that of peoples reput- child had no trouble to recognize edly less civilized than ourselves. the objects. Commenting on this ex- This method was also first used in periment in a later text, Hochberg the study of verbal language. How- ever, at least from the last century Fig. 1. “La clef des champs”, by René Magritte – an illustration of the problem of separating sign and reality.

onwards, explorers and travellers, are Herskovits’s story about the puz- and later anthropologists and social zled woman who turns the photo- psychologists, have reported on the graph of her own son over again and difficulties experienced by members again, without being able to under- of “savage tribes”, principally in Af- stand what it is, and Muldrow’s de- rica, when they were confronted with scription of the Me’ tribe, whose pictures for the first time and asked members smell and taste the pic- to explain their content. Essentially, tures, but do not think of looking at these reports would seem to testify them. The second series of anecdotes to two very different, and apparent- may be illustrated by the tale of the ly contradictory, obstacles to an ade- tribe panic-stricken to the point of quate pictorial understanding: for ei- running away at the sight of a slide ther the hero of the story is unable to projection showing an elephant; and make out what kind of object the pic- by the report of another tribe treat- ture is, and what function it serves, ing photographs of white women as or he fails to distinguish the picture if they were real people. from what it represents. Typical in- Here, then, we encounter in their stances of the first kind of anecdotes practical form the very same theoretical issues that have been central to Cf. Deregowski 1972; 1973; 1976; the discussion of iconicity (to which also for the following anecdotal material.

we turn in the next lecture): the prob- ty (Deregowski 1984: 20). Epitomic lems of relating the picture to its ob- aspects of depiction may be present ject, and of distinguishing the former in outline drawings and silhouettes; from the latter (cf. Fig. 1.). Different- and eidolic aspects may appear in ly put, iconicity theories must expect depictions the object of which cannot all human beings to discover the re- be identified because it cannot exist, latedness of the picture and its ob- that is, in “impossible pictures” such ject immediately, but some tribes fail as the devil’s turning fork. to do that; and, rather more implicit- With the exception of such sin- ly, these same theories must suppose gular instances, the failure to see that we are all able to tell the picture depth normally testified to be abili- and its object apart, but this too, it ty to see pictures as pictures. Hud- seems, is something some tribes fail son performed a number of tests us- to do. ing perspectival pictures, which But the experimental literature were repeated by Deregowski with is really concerned with a third prob- some modifications, and both of lem: our ability to discover, not that them found a lesser ability on the something is a picture, but what it part of native Africans, particular- is a picture of. Moreover, most of ly unschooled ones, to interpret cor- the experiments have been devoted rectly the depth cues of otherwise to an investigation of the extent to ambiguous pictures. But Kennedy which Non-western people are able (1974:65 ff) gives a number of rea- to decode the depth cues inherent in sons for questioning these results: for Western linear perspective, whereas instance, the drawings were often so the logically primary task, the study unclear that the answers given by the of their willingness to take pigment Africans seem as plausible as the ex- patterns on paper to represent three- pected ones; and the social conse- dimensional objects of the world, has quences of having, in South Africa, been seriously neglected. This state of a white experimenter posing ques- the art has repeatedly been regretted tions to black people were ignored, by Deregowski (1973:165; 1976:19), although these are evident from the who went on to distinguish the prob- fact that some persons waited an lem a identifying the percept corre- hour before making their reply. Jones sponding to a objects in a picture, the & Hagen (1980:203 ff) observe that epitomic ability, and the problem of white people never get 100 % right recognizing depth, the eidolic abili- at the Hudson test either, and that New York children have been clas- Cf. the reviews of this literature in sified as two-dimensional perceivers Deregowski 1972; 1973; 1976; Kennedy according to the criteria of this test. 1974a; Pick & Pick 1978; Jones & Hagen 1980. Cf. Lecture 3. However, from our point of view it is dow behind a woman’s head looks important to note that, even if Hud- to the East African like a four-gallon son is right, the fact that the Africans tin carried on the head in question were able to go through with the test, (which supposes a Necker cube type seems to suppose that they recog- of perspectival reversal). But consid- nized the picture as such, and as dis- ering the great amount of different tinct from what it was a picture of, picture types, and their different lev- and that they identified the pigment els of complexity, almost nothing is patterns as standing for the antilope, really known about the limits of ob- the elephant, and the tree. Thus, they ject recognition in pictures. were certainly superior to the Afri- We can now return to the is- cans of the anecdotes. sues raised at the beginning of this In fact, Hudson’s subjects were section: the difficulties of relating probably familiar with pictures, the picture to its object and distin- though not with perspectival ones. guishing the two. Referring to Her- However, Kennedy & Ross found that skovits’s puzzled woman, Kennedy the Songe of Papua, who have no pic- (1974:68) points out that being puz- tures, could identify well-known ob- zled over something is very different jects on outline drawings in 90 % of from seeing it as “mere daubs on a the cases, while less well-known ob- surface. Indeed, mere daubs on a sur- jects were identified by 10-20 years face would hardly puzzle anyone.” It olds 97 % of the time, and by those is conceivable that the woman does over 40 years of age 68 % of the time. recognize her son, but that it seems Also Deregowski found that Ethio- unbelievable to her that a mere piece pians over 40 years old were slow- of paper is capable of suggesting the er at identifying depicted objects (cf. appearance of her son. But now what Jones & Hagen 1980:198). In general, about Muldrow’s story? Members of the identification of objects on col- the Me’ tribe, we are told, smell the our photographs occasions no prob- pictures, taste them, bend them, and lems, according to Jones & Hagen so on, in short behave like a Piage- (1980:196); but black-and white pho- tian child exploring his world. Ac- tographs may cause trouble, particu- cording to Deregowski (1973:167; larly so if, as in one of Deregowski’s 1976:20) not only pictures, but ma- tests, pictures of unknown animals terials like paper are unknown to the have to be matched with the cor- Me’; therefore, when Deregowski had responding models. In the case of pictures printed on coarse cloth, an- somewhat more complex drawings, imals well-known to the tribe could Deregowski (1973:165) has noted the be identified, although the recogni- importance of cultural expectations: tion was still not immediate. In the what to a Westerner seems a win- case recounted by Muldrow, it seems

the Me’ were so busy trying to dis- Cabe (1980:335), who makes this ob- cover the fundamental properties of servation, tells us he has taken pains the paper as an object in itself, that to ascertain that the pigeons of his the iconic properties, those making experiments possess the later capac- it a pictorial sign of something else, ity (p 313f). So far, then, it seems that were not noted; other attributes be- even pigeons are superior to our an- came dominant in their experience ecdotical Africans in the art of read- of it. It therefore seems (as I suggesting pictures. ed in Sonesson 1989a) that for some- According to Howard Gardner thing to be a pictorial sign of some- (1982:105), American children aged thing else, it must occupy some rel- 4 to 7 tend to confuse the motive atively low position in the particular and the picture; however, when at- Lifeworld hierarchy of “things”. Be- tention is called to the medium, they fore returning to this question, how- are able to understand the point. Per- ever, it will be convenient to consid- haps, then, the distinction just seems er the second of the issues mentioned to them to be too obvious or too un- above: the distinctiveness of picture important to be mentioned. The mo- and object. ment after having taken to flight at the sight of the pictured elephant, Further views from the the members of the tribe visited by playground: pigeons, apes and the explorer Lloyd discovered their men mistake and returned laughingly to The Ancient Greek painter Zeuxis is the front of the screen. Of course, famous for having depicted a bunch the difference between the elephant of grapes in so illusory a manner, that and its picture was neither unim- even the birds were fooled. Comment- portant nor obvious to them; but in ing on Pliny’s well-known story, Go- a moment of potential threat, they mbrich (1963:5f) claims this was no were certainly wise to react on in- great feat of Zeuxis’s since, as ethol- sufficient evidence. Since perception ogy has shown, animals react to very seems to start relatively high up on gross similarities. However, it seems the ladder of abstraction, it is indeed that Cabe’s piegons would not follow probable that, in a moment of stress, suit as the other birds launch their only very gross similarities will be attack on Zeuxis’s grapes. Most ex- noted, even those which are not or- periments purporting to demonstrate dinarily category-defining. The oth- the ability of some animal species to er story, where photographs of white interpret pictures have neglected to women are treated as real people, is investigate whether the animals are rather implausible; if not some mag- also able to tell the difference be- ical equivalence is meant, then per- tween the picture and its object; but haps this behaviour must be under-

stood as a kind of social deference to torial signs, are made, was not obvi- the white men who showed the pic- ous to Herskovits’s puzzled woman; tures. Again, more research would and to the Me’, this material was so be needed to go beyond these anec- interesting in itself that it absorbed dotes. all interest; coarse cloth, however, More recent experiments have was easier to conceive in this hum- shown that even children 5 months ble part, though even now, time was of age look longer at a doll than at its needed to discover what was depict- picture (DeLoache & Burns 1994). ed, perhaps because the sign func- However, it does not follow that the tion itself had to be discovered. If children see the picture as a picture. we suppose the Hochbergian child to Indeed, 9th months olds, but not 18th understand, not only that given pig- month olds, try to grasp the object ment patterns on paper have some- depicted as if it were a real object thing to do with the shoe, the doll, (DeLoache 2004); whatever the dif- and the Volkswagen of the real world, ference they perceive, then, it does but also that the former are signs for not seem to involve signs as opposed the latter, and not the reverse, then to objects. It seems to me that, just it will not be enough for the child to as in the case of the doves, this may have learnt from his experience with simply show that the picture and its objects of the world that the edges object are seen as being different, but of objects have properties which are not necessarily as being a sign-vehi- shared by contours drawn on paper, cle and its referent. The real doll is or to be innately predisposed to re- perhaps seen as a more prototypical act to these common properties (cf. instance of the category; or, alterna- Hochberg 1978a:136). He must also tively, the real object may be more have acquired, probably from expe- interesting because of having more rience in his particular Occidental perceptual predicates. Lifeworld, some notion of the rela- Sonesson (1989a) argued that tive low ranking on the scale of pro- once we know that something is totypical Lifeworld things of a mate- a sign, and, specifically, a pictorial like paper, which directs his at- rial sign, the particular “similari- tention, not to what the pigment pat- ties” will take care of themselves. terns on the paper are as “selves”, but If we are not told that some particu- to what they stand for. And perhaps lar thing is a sign, and iconic at that, he must also possess some idea of a then we may perhaps be aware of it meaningful organization, which re- because of general facts derived from lieves him from the task of finding our experience of the common sense a meaning in ink blots, in the dirt on world. That paper is the kind of stuff the road, in the stains he makes with of which signs, and in particular pic- his dinner on the tablecloth and in

the clouds. scribe the necessity for the child to Familiarity with paper or cloth attend both to the picture and the ob- are facts of particular cultures. Pa- ject depicted. This is a misleading per, which is too prominent to the term, for there is only one representa- Me’ to serve as a sign-vehicle, tradition, that is, one sign function. Rath- tionally carries this function in West- er, in Gibson’s more enlightening ern culture. But Sonesson (1989a) terms, there are invariants for both suggested that there probably also the surface and the referent in the ob- would be universals of prominence: ject, and the task is to tell them apart, thus, for instance, two-dimensional and decide which is most prominent. objects are felt to be less prominent In fact, the problem only arises be- than three-dimensional ones and cause there is at the same time a sign may thus more readily serve as ex- function and iconicity. This means pressions. In this sense, it is not true that the term ”double representation” that the object is its own best icon, as is not only misleading: if fails to exis ordinarly claimed – at least if icon- plain why pictures are easier to inter- ic means iconic sign. Indeeed, ico- pret than scale models. nicity stands in the way of the sign In all Deloach’s experiments, function. The objects of the common the task is, in one way or other, to sense world are three-dimensional: find a hidden object by using infor- much less is required for a two-di- mation contained in a picture or a mensional object to be able to repre- scale model. According to the stand- sent one of these objets than for an- ard procedure, the experimenter and other three-dimensional object to do the child are at first outside the room so. This is precisely what is suggest- in which the child is to search for the ed by DeLoache’s more recent ex- toy. The child cannot see the picture periments with children: not only is or scale model and the room at the the picture understood later than lan- same time. The experimenter tells guage in these experiments, around the child that she will hide the toy 2 1/2 years (Deloache & Burns 1994, in the room and then come back and etc.), but scale models are under- ask the child to search for it. She re- stood even later, at 3 years of age, turns to the child and points out the half a year after pictures (DeLoache appropriate location in the picture/ 2000). As noted also by DeLoache, scale model telling it “This is where this contradicts what is expected by Snoopy is hiding in his room, can common sense. But it is reasonable, if the issue is separating the sign and Perhaps DeLoache talks about its referent. “representation” in the sense in which the term is often used in cognitive science, DeLoache (2004) employs the but then this is precisely the problem, as term ”double representation” to de- we shall see later in this lecture. 10 you find him?”. If the subject fails in scaffolding: not only are the objects the first search it is once more shown pointed out by the experimenter in the picture and given more explicit the picture or the scale model, but prompts. 24 month old do not pass the latter are even placed on the real the retrieval test, but 30 month old objects, creating an artificial neigh- do; there is no difference in perform- bourhood relation. ance using photographs or line draw- Another one of Deloache’s ex- ings. However, when the whole pro- periments seems to indicate that the cedure is conducted verbally, chil- sign function is at least part of the dren pass the test already before 24 problem. When the experimenter, in- months old; and when a scale model stead of talking about a model and a is used, only 36 months old pass it. real room, tells the children that the This way of investigating the search has to take place in the same picture function may be criticised room, which has shrunken since from two diametrically opposed it was last seen, the task is accom- points of view. First, it could be ar- plished much more easily (DeLoache gued that the task involves much & al. 1997). The difference, clear- more than the recognition of the pic- ly, is that the two instances are here ture as picture - it requires an action, connected by a narrative chain rath- which is no doubt difficult in itself, er than by a sign relationship. In an- namely, to search for the hidden ob- other experiment, DeLoache (2000) ject. It remains, however, that even places the scale model behind a win- this task is differently accomplished dow-pane, in order to make it more if the instructions are given entire- similar to a picture, with the expected ly in verbal form, or if they involve results. In fact, however, two things pictures or scale models. On the oth- happen here which would have to be er hand, even when the instruction separated: the object becomes less for the task features pictures or scale prominent, because it has less the models, at lot of verbal and indexical appearance of three-dimensionality; scaffolding also takes place, without and it is put into a frame, which cre- this being taken into account in the ates a centre of attention. interpretation. It has been argued by DeLoache’s work experimen- Callaghan & Rankin (2002) that pic- tally investigates the central issues tures would be interpreted even lat- broached in Sonesson (1989a). As er if such verbal scaffolding had not always, the investigation engen- taken place. More fundamental, how- ders new problems. However, if un- ever, may very well be the indexical derstanding pictures is as difficult for children as DeLoache and, even I owe this suggestion to my stu- dent Sara Lenninger, who is preparing a Clearly without knowing Sonesson dissertation on this subject. 1989a.

11 more, Callaghan, suggest, then we It is clear, however, that, in or- should not expect animals to be able der to understand the peculiarity of to do so. We have already proposed the picture, we need a concept of sign some alternative explanations for the which can account for the difference behaviour of Cave’s pigeons. On the and similarity between perception other hand, primatologists, as men- and pictures, on the one hand, and of tioned at the beginning of this section, pictures and scale models on the oth- tend to take for granted that the apes er. to which they are trying to teach language already understand pictures. Pictures in the sand: Anati’s There are only a few regular inves- “prayer” and the native eye tigations of apes looking at pictures Archaeology should ideally be able and scale models. Itakura (1994) re- to tell us something about the origin ports that enculturated chimpanzees of pictures in the prehistory of hu- can interpret line-drawings; Kuhl- man beings. However, those artefacts meier & al (1999: 2001; 2002) have which clearly are pictures, such as the even shown their chimpanzees to un- well-known Ice Age rock carvings, derstand scale models. It is difficult are products of a very recent prehis- to know what to make of these re- tory indeed, and this even holds true, sults, already because these apes are in view of the length of prehistory, all enculturated, which is to say that of those artefacts which, perhaps less they are trained in many of the semi- convincingly, are claimed by some otic resources which in ordinary cir- archaeologists to be pictures or other cumstances are peculiar to the human kinds of man-made artefacts, such lifeworld. Moreover, it should be not- as, notably, sculptures and calendars ed that, while the children were in- (such as the Berekhat Ram figure troduced to a model of a room which and Marshack’s putative calendar; they had never seen before the train- cf. Bahn 1998; White 2000; Elkins ing-phase, the apes were confront- 1996; 1997). No matter how early ed with a model of their own famil- such artefacts are in the end shown iar environment. In addition, a lot of to be, however, there is no way of facts about the subjects and the ex- establishing that no pictures existed perimental procedure are not clear before them. The first drawings may from the articles. At present, it would not have been made not on rocks, but therefore be premature to draw any perhaps on sand, on clothing, or on conclusions about the abilities of the great apes in this domain. none of these articles taken into account to different between the picture and the More about picture perception depicted object. in pigeons, and a little in apes, can be See the next subsection for a dis- gathered from Fagot, ed. 2000. However, cussion of these artefacts. 12 human skin, and on other highly per- ishable materials. The figures appearing in many prehistoric rock carvings curiously resemble our present-day traffic signs, or the tadpole men of contemporary children’s drawings, or the logograms found on the doors of men’s and women’s washing rooms, Blissymbolics used to communi- Fig. 2. One variety of Anati’s ‘prayer’ cate with those suffering from dif- (from Anati 1976:46, passim) ferent kinds of speech-impairment, the Alchemic symbols of the Mid- same configuration, may thus serve dle Ages, the Hobo signs still em- as the expression of quite different ployed by tramps and vagabonds un- contents, that is, may form part of til the Second World War, or signs very different signs. What is materi- stemming form many other, mutual- ally identical, is not semiotically so. ly divergent, repertories. Thus, dif- To describe a sign, or a sign system, ferent varieties of the anthropomor- we have to recover the point of view phic figure, which Anati (1976) de- of its user. This is what was meant scribes as a “prayer” (see Figure 2), when it was said in Lecture 1 that may be compared to a sign denoting semiotics should describe meaning the golden number 18 in the clog al- as it appears to those who employ the manacs of the Middle Ages, or one signification system. In verbal lan- of the letters of an Rumanian alpha- guage, this is illustrated most clear- bet used around the year 1000, the ly by the case of the “same” sound alchemical signs for test, for essence, which forms more or fewer differ- or for mix, or the astrology signs for ent phonemes, or only a variant of a Pisces or the fixed star Spica, Nep- phoneme, according to the language; tune’s or Jupiter’s staff, and so on (cf. thus, for instance, the same physi- Liungman 1991: 117; 118, 155, 156, cal sound which is English forms the 434). In none of these cases the fig- phoneme /r/, and is opposed, among ure represents a person: indeed, in other things, to the phoneme /l/, is most of them, it is not even a pictori- only a variant of the latter in Japa- al sign. Yet it is easy to imagine that nese; while, on the other hand, free the same figure may stand for a hu- variants of the English phoneme /r/, man being also in the drawing made with one or more slaps of the tongue, by a contemporary child. form two different phonemes in Spanish. The very same material lines, i.e. what is, on the face of it, the Renfrew (1982:11) implicitly re-

13 curs to the same analogy, when he for us make up the remains of his so- quotes the linguist Kenneth Pike ob- ciety and of his world of thinking, serving that the archaeologist’s ob- and for this reason, the creation of a servations relate to the “etic” (as in social archaeology, let alone a cogni- “phonetics”) rather than the “emic” tive one, constitutes a difficult, if not (as in “phonemics”). Phonetics is an impossible, task. On the face of concerned with the sounds as such, it, it also heavily constrains the pros- but phonemics (or “phonology”) de- pects for prehistoric semiotics. scribes the sounds as they are con- In the most common, or most ceived by the speaker of a particular commonly reproduced, variety, Ana- language (that is, in relation to other ti’s “prayer” could probably be de- sounds appearing in that language). scribed, in Liungman’s (1991) terms, In the terms of other linguistic the- as a sign which is single-axis sym- ories better known to semioticians, metric, both soft and straight-lined, Saussure and Hjelmslev, phonetics open and with crossing lines (with al- is concerned with substance, where- lowance made for the little, scooped as phonology investigates form. A circle, that appears on the top, which single substance underlies the Japa- should make the figure into a sign nese /l/, the English /l/ and /r/, and which is both open and closed). Ana- the Spanish /l/, /r/, and /rr/. Anati’s ti certainly does not think in these “prayer” would share its substance terms: even the groups in which he with, but have a differing form from, puts the different figures (1976:46) those other signs culled from Liung- show that he conceives of them im- man’s book. mediately in terms of what they “rep- Interestingly, even Tilley (1991), resent”, not as spatial configurations. who often rhetorically insists on In semiotic terms, in would appear the materiality of material culture that Anati passes too rapidly from (which, to him, includes rock carv- the plane of expression to the possi- ings and rock paintings), is enough ble corresponding plane of content. of a structuralist to realise that mean- Indeed, Bednarik (1991:1) seems to ing (that of the rock pictures, for in- accuse archaeologists in general of stance) is only there to be seen for doing just that. As we shall see lat- those who have the capacity, that is, er on, this charge will be particularly the members of some particular cul- serious, if Bierman, Eco, Goodman, ture (which may, of course, be as Lindekens, and others, are justified wide as humankind itself). As Ren- in their critique of iconicity. Even if frew (1982:11) notes, there can be no they are mistaken, however, we are direct access for us to the meanings still, as Jarl Nordbladh (1973; 1977) which were once projected by pre- observes, faced with the task of re- historic man onto the artefacts which covering the vanished context of pre-

14 historic pictures, not only their de- too simplistic an interpretation of the cayed tactile, auditory and olfactory term, an iconical sign. structures, but the particular socio- Before we can even begin to ask cultural Lifeworld in which they oc- ourselves whether Anati’s “prayer”, curred generally. or rock carvings generally, are icon- It is, however, not only the vari- ical signs, we have to take care to ety of the contexts in which what is, avoid two, fairly trivial, but com- in a way, the “same” drawing, may monly made, confusions pertain- appear, which should surprise us ing to the import of iconicity. To be- here. What is perhaps more remark- gin with, iconic signs are often erro- able is that, in spite of all the diverg- neously taken to be the same thing ing contexts, there appears to be a as visual signs (for instance in cog- common background, a human envi- nitive psychology, when discussing ronment which we unavoidably take “iconic codes). And, in the second for granted, which may in part ac- place, iconicity mostly tends to be count for the fact that Anati can hope identified with picturehood (which to perceive the same thing as prehis- may happen, in a more surrepti- toric man. This world of background tious way, as we shall see, even in- experience is known in Husserlean side semiotics), when in actual fact, phenomenology as the Lifeworld, in if we rely on Peirce’s definition, pic- James Gibson’s ecological psycholo- tures only constitute one variety of gy as the world of ecological phys- iconicity. When considering the first ics, and in Greimas’ semiotic theo- appearance of “iconic structures (en- ry as the natural world (in the sense gravings, sculptures and ultimate- in which we talk about “natural lan- ly cave art)”, Foley (1991:114) ap- guage”, the language which seems pears to be guilty of at least one, or natural to its users). Indeed, if there perhaps some compound form, of can be such a thing as a semiotics of these confusions. As for Bednarik Culture (as propounded by the Tartu (1991:1), he clearly associates “icon- school, by Koch, Posner, etc.), it must ic intent” with the belief, which he be elevated on the foundations laid censures, that the researcher has the by a semiotics of Nature — which is ability to identify the objects depict- of course, in a very general sense, a ed in prehistoric pictures. And even culturized Nature. But, before we go Chesney (1991), when claiming that in to discuss the eventuality of Ana- much more early production than is ti’s “prayer” being interpretable on commonly believed is non-represen- the basis of common, anthropologi- tational and non-iconic, appears to cal universals, let us ponder the pos- take the latter term to mean simply sibility of it being simply perceived, non-pictorial.10 that is, of its being, on one, probably 10 Similar confusions are found in

15 Since the discussion of pictoral- here as Figure 3. As we have suggest- ity as a specific variant of iconicity ed (relying on Scruton 1974:204), al- will have to wait until Lecture 4, I though its expression plane is quad- will here simply substitute the term rangular, and no actual faces are, “pictorial” for “iconic”. Even so, Fig. 3a is naturally seen as a face; yet there remain many problems with Fig. 3c should be even more inevi- putative prehistoric pictures. In the tably be identified as representing a case of a proper picture, we are im- face, although Hermerén (1983:101) mediately able to “see into” the ex- claims that this is so, only because pression plane, and project as its of “the limitations of human imagi- content, some part of the perceptu- nation”, since the same pattern may al world, without receiving any fur- equally well be perceived as “a jar ther indication on how it should be from above, with some pebbles and taken. In this sense, pictures are dif- broken matches on the bottom, and ferent from “droodles” (in the terms a stick placed across the opening”. of Arnheim 1969), which require a Even such an elementary stick fig- “key”. That which defines a droodle ure as Fig. 3b, was immediately de- is not the presence of multiple inter- clared to be a chair by a child one pretations, but the fact that the app- year and eleven months of age (Stern resentation in sparked off, and mean- 1914:159); and we could easily agree ings distributed to the parts, only with von Däniken (1973) that Fig.3d once a verbal label has been attrib- represents a wrist-watch, until we uted to the figure. Clearly, in this learn that it is found on prehistoric sense, Anati’s “prayer” is a picture, rock paintings. not a droodle, as are many other rock There is nothing accidental, I paintings and engravings, discussed submit, to those “limitations of hu- in books by Anati, Burenhult, Tilley, man imagination” invoked by Her- and others; yet it may be suggested merén: they are imposed by the that, while being indeed a picture of a Lifeworld hierarchy of prototypi- man, Anati’s “prayer” is only a droo- cal things. Indeed, there must be an dle of a person making his prayers!11 infinity of objects whose light pat- Consider some limiting-cases tern, in a static view, fit much better of pictures and droodles reproduced to the square pattern on Fig.3a. than the psychological literature concerned a face, and yet we cannot help see- with Blissymbolics and other alternative ing it. And although it is possible to communication system for the speech-im- impose the jar reading suggested by paired, e.g. Muter 1986. Hermerén on Fig. 3c, it is only there 11 Or, put into Panofsky’s terms, in the droodle fashion, once a key which we will discuss in Lecture 3 and has been given, and it is all the time 4, it is a pre-iconographic picture, but an iconographic droodle. being disturbed, and in fact overrid- 16 d

Fig.3. Pictures and droodles: a) Quadrangular face (from Scruton 1974: 204); b) chair (from Stern 1914:159); c) face or jar (from Hermerén 1983:101); d) wrist-watch or something else (suggested by von Däniken 1973). den, by the more “natural” face in- daggers and other arms, is that the terpretation. It seems, then, that we latter, but not the former, may derive come the task of picture interpreta- some justification in his labelling of tion equipped with certain expectan- the figures from an indexical abduc- cies to encounter those objects which tion, i.e. the presence, in the same or are normally close at hand in our eve- similar archaeological sites, of ob- ryday Lifeworld, such as faces and jects having a form which may be fit- human bodies and, in our culture, ted to the pattern. Garrick Mallery, chairs and wrist-watches. No doubt who may have been the first to con- most or all objects and scenes may duct what he already termed a sem- be depicted, but if they rank below iotic study of iconical signs, nota- the apex of the hierarchy built of our bly of American Indian rock paint- Lifeworld expectancies, many more ings and manual gestures, observed, details are necessary, for the object in the case of the latter that many of or scene to be recognisable. At some the manual gestures were “reasona- point in human history, chairs be- ble”, because the similarity between came such familiar objects in the or- the sign relata could be observed by dinary Lifeworld that just three lines a person acquainted with the cul- were required to make them recog- ture, or once the sign had been ex- nisable; very much later, the same plained to him (Cf. Mallery 1881:94f destiny befell wrist-watches, astro- and Kroeber’s introduction, p xxiv). naut’s helmets, space-crafts, etc. — Thus, Mallery’s manual gestures, which is why von Däniken’s obser- like Anati’s dagger, are iconical in a vations are off the mark. way, but only secondarily, once a key Many of the figures identi- has been furnished. Then the shape, fied by Anati (1976:223ff) are real- or the outline, may even be reseman- ly droodles: indeed, the difference ticised. between von Däniken recognising The case of the “prayer”, how- wrist-watches, and Anati identifying ever, is different. No key is needed

17 to see the man. And, the evidence for ecological physics. the man on the rock being so scant, I This kind of argumentation can think we must conclude that the will- be taken much further: indeed, it ingness of human beings to perceive might be argued that even iconic de- other human beings, wherever pos- tails are picked out, not because they sible, is great indeed. It is not clear, depict so well, but for conventional form reading Anati’s (1976) book, reasons and/or because they serve a however, why we should take the metaphorical level of meaning, for man to be in a position of praying. instance so as to separate men and Perhaps Anati has some evidence for a women, for which a weapon or a this, comparable to that for the dag- plough may be as good, given con- ger, though remains of decayed ac- temporary circumstances, as an tion sequences are certainly more erect penis, and a cup mark as elo- difficult to come by; or perhaps there quent anywhere as between the stick- is really a kind of anthropological figures legs.12 universal of praying which may be First, however, it will be con- profitably invoked here (which would venient to discuss the difficulty of indeed be universal, then, for figures interpreting pictures from prehistor- of the same general kind are found ic cultures in analogy with the prob- also in China, cf. the illustrations lems raised by the interpretation of to Li Fushun 1992). It seems to me, pictures from other culture, notably however, that a much less risky hy- those which are maximally foreign, pothesis, with more general validity, stemming from other planets or so- may be proposed to explain the posi- lar systems. Since we are aware of tion of the arms of the “prayer” fig- no pictures from other solar systems, ure: it may be suggested, that in the however, it might be better to the re- context of a fairly limited set of other verse the formulation of our conun- motifs, including many animals, the outstretched arms are there to signi- 12 Since we need to introduce some fy “humanity” in an emphatic sense, more semiotical terminology before being that is to say, prototypically: to sin- able to discuss these issues, they will be gle out the peculiar feature which taken up in Lecture 3 and 7, respectively. The present criticism involving Anati’s marks off human beings form other “prayer” may with as much reason be di- animals, and the discovery of which rected to the presently fashionable reinter- was a decisive step in the process of pretation of later Scandinavian rock car- hominisation: the erect posture; and vings as showing a type of vessel which was used in the Mediterranean, and thus thus to indicate the horizontal direc- reflecting, not Scandinavian culture, but tiveness which remains a determin- that of a higher foreign culture. Just as in ing characteristic of the human Life- case of Anati’s “prayer” the interpretation world, the terrestrial environment of may well be correct, but the justifications for it seems completely arbitrary. 18 pool of knowledge: abductions, norms, sign systems incommensurability

adaption to adaption to target source

Coding Inner reinterpretation

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

Transport

Fig. 4. The place of ”incommensurability” in the ordinary communicational model drum, taking up the problem, set to At the heart of semiotics, as Martian and other green little men, well as the problem of communi- of interpreting human pictures. cating with extraterrestrials, is what Douglas Vakoch (1999) has called Terrestial and Interstellar “the incommensurability problem”: Archaeology the models constructed by scien- As we have seen, there traditionally tists on earth vary considerably, in are two or three ways of investigating part because of their different so- the constraints on the specificity of cial and historical backgrounds, so it the (human) semiotic function: stud- would be surprising if such a varia- ying child development, scrutiniz- bility were not augmented by the fact ing the capacities of apes, monkeys of the scientists coming from differ- and other animals; and analysing ent planets, in which case biology cultures which are not familiar with may also be different. This issue is some kinds of semiotic resources not only relevant to scientific mod- (“primitive” cultures), such as, most els, but applies to the transmission of classically, pictures. In this sense, different planets or, more reasonably, munication with extraterrestrial intelligen- solar systems, would be maximally ce (which the researchers at the SETI call foreign cultures.13 CETI) in itself. Rather, for me, thinking about communication with extraterrestri- 13 Unlike people at the SETI insti- als is a testing case (imaginary so far) for tute, I am not basically interested in com- the constraints imposed on semiosis.

19 any kind of messages. Indeed, in my for the latter situation. Philosophical own version of the communication hermeneutics, on the other hand, of- model (Fig. 4), which - deriving its ten envisions some kind of combi- inspiration from the Prague school nation of the two processes: a “fu- of semiotics - takes into account the sion of horizons”, in Gadamer’s fa- active construal of the message on mous phrase. The incommensura- the part of the receiver, the pool of bility problem, in its extreme forms, knowledge, including norms, abduc- suggests the opposite case: when the tions, and sign systems, held in com- overlap between the two initial pools mon by the protagonists of the com- of knowledge approaches to zero. munication process, is - following It is important to recognize that, the parallel suggestions of Lotman in a situation of communication, the and Moles - supposed to overlap only first problem is not to find out what in part at the beginning of the proc- the messages means: it is to realise ess (Sonesson 1999b). that there is a message. That is, it in- If the act of communication volves the recognition of the message may still succeed, this must either as such – as a message, rather than be because the sender takes pains to a message about something in par- adapt his pool of knowledge to that ticular. Even those theories of com- of the receiver, or because the receiv- munication which insist on the act er does so with respect to the knowl- by means of which meaning is pro- edge of the sender, or owing to some duced and conveyed, on the enuncia- combination of both approaches. In tion rather than the utterance, are not the first case, we have what the Tar- very clear about this issue. “Speech tu school calls a receiver-culture; it act theory” (Austin, Grice, Sear- is, as I have formulated it elsewhere le, etc.) separates the content of the (Sonesson 1999b), a culture in which message (“locution”) from how it is it is felt to be the task of the send- to be taken (“illocution”) and even er to recover the norms and interpre- the effect it may have or not have de- tations characteristic of the receiver. pending on circumstances (“perlocu- The classical case is the pedagogical tion”), but it is very vague about “up- situation. In the case of a sender-cul- take”, the necessity for the message ture, on the other hand, the receiv- to be attended to as such. In Jakob- er is assigned the task of recuper- son’s model of communication, one ating the part of the pool of knowl- of the functions, called the “phatic edge peculiar to the sender that does function”, is supposed to assure that not overlap with his own. High art, the message gets through, but Jakob- as well as mystery cults, are of this son has very little to say about the kind. Hermeneutics, as a science way this is brought about, apart from with practical goals, was developed giving the commonplace example of

20 Fig. 5a. Stretches of forest cleared in the shape of geometrical figures in order to send messages to other planets (from Vakoch)

checking whether the telephone line be recognized as such. is open by saying “Hello”. Luis Pri- The first incommensurabili- eto has been much more insistent on ty problem thus concerns the recog- the difference between the message nition of the message as a message. (“indice”) and the information that Such recognition requires us to share somebody is sending a message (“in- some common presumptions about dication notificative”). But somehow the shape of possible messages. This the essential question gets lost in the is nicely illustrated by the examples discussion of intentions. After all, quoted by Vakoch (1999: 2003) of even an unintended message has to messages which a mathematician in

Fig. 5b. Kerosene-filled channels created with the purpose of sending messages to other planets (from Vakoch).

21 the 1820s suggested could be formed pose somebody had the intention… by clearing massive stretches of for- etc. If it is impossible that some- est in Siberia producing geometrical body was around, then, miraculous- figures; and which others hoped to ly, God, some ghost or ET must have obtain by digging geometrically ar- been doing the writing.14 The as- ranged channels in the Sahara which tronomer Richard Hoagland says he could be filled with kerosene and set has discovered, on pictures from the aflame during the night (Fig, 5a-b). planet Mars, a sculpture of a mon- Even if the Martians or the inhab- key’s head, together with some other itants of the Moon could see these strange constructions, which must be shapes, and recognize them for what traces of an ancient Martian civiliza- we think they are, they would only tion. For obvious reasons, other as- learn anything about us, to the ex- tronomers think this is as absurd as tent that they understood that these affirming that the man in the moon are messages send by us – and, even has been painted by intelligent beings. more fundamentally, messages, period. 14 This was my criticism of Searle in Sonesson (1978). Interestingly, in more Searle (1969) claims we can recent publications, Searle (1995) clearly only see patterns in the desert sand goes along with this criticism, both in as writing if we suppose somebody postulating a “we intentionality” and in positing a “background” which seems to intended that we should understand be another name for the world taken for somebody had the intention… etc. granted. Unfortunately, I cannot even have But the opposite is of course true: it the illusion of having influenced Searle is only because we see something as here, since my criticism was only published in Swedish. For my new quarrels with being (typically) writing that we sup- Searle, see further sections of this lecture!

Fig. 6. Cros’ and the Niemans’s codings and the resulting figure

22 However, what Hoagland presents us with is actually an iconical sign of another putatively iconical sign, i.e. a photograph of the monkey’s head. If his claim were borne out by direct observations, then we would have to admit that von Däniken’s space gods, with their superior technological resources, have landed on Mars and ed- ified the monkey’s head, just so as to bewilder us.15 This is parallel to, but more complex than, a case considered by Arnheim (1966:93ff): a prototypical picture should possess con- figurational and other holistic properties not found in ink blots which, in their natural state, are all too ir- regular and, in their Rorschach version, too symmetrical. In the end, then, what is needed, are criteria for some shape being a message. One such criterion is no doubt “ruleboundedness”: regularity, repetition, etc. – that is symbolicity, in a Peircean sense. Interestingly, Fig. 7. The so-called Berekhat Ram figure as we shall see, this is what is found icality, on the other hand, as found in Gros’ and the Niemans’ schemes in traces, for instance, could easi- (Fig. 6.), as well as in Drake’s later ly suggest no intention to commu- proposal: in the first case, the same nicate, that is, messages involuntar- number for each line, and in the sec- ily produced. Thus, we recognize the ond “551 = 19 x 29” – although the same interplay of iconicity, index- same clue has to do service a second icality, and symbolicity as in inter- time as a signifier of “mathematical- stellar messages.16 ness”. Another such criterion is similarity, that is, iconicity (But this may Interstellar communication pro- lead to projection, as in Hoagland’s jects into space such issues that have monkey face and van Däniken’s long preoccupied archaeology in wrist watches and helmets). Index- their temporal manifestations. Thus, archaeologists are wont to ask: Is the 15 Quoted from a newspaper article in Sonesson (1989a: 254f).s 16 On these terms, see Lecture 3.

23 Berekhat Ram figure (Fig. 7), an object dated to between 233-800 000 BP (according to Bahn 1998:86), the likeness of a woman? But before this question can be formulated another question must be posed: do the traces of abrasion left on it show regularity sufficient and, at the same time, not too extensive as to suggest “anthropogenic” movements (that is, intentional manipulation by human beings)? Although it has never been claimed to be a picture, Marshack’s “calendar”, if it were indeed a calendar, i.e., another kind of artefact with a cultural imprint, would have to evince some kind of regularity in the very way its traces are disposed. Indeed, Marshack uses a microscope to detail the sequences of differently disposed strokes which are found on the Bâton from Le Placard, Charente, arguing (as quoted by El- kins 1996: 189; 1997: 60ff) that the strokes must have been purposefully made, since the sequence of figures appears odd, deviating from a near-regularity, and thus, he supposes, they cannot by purely ornamen- Fig. 8a. The result which should be tal, but must be some kind of nota- obtained when construction Drake’s tion representing a lunar calendar. figure If there is some justification for this claim, it can never come from the us, the sequences of lunar phases. If scrupulous observation by means of such as correlation between the in- a microscope realised by Marshack, scription on the bone and the lunar contrary to what the latter claims, system is successfully made, there is but must stem from the comparison every reason to suppose the inscrip- of the configuration of the strokes on tion to be purposefully created (cf. the bone with another system of or- Sonesson 1996b). The problem, how- ganization, independently known to ever, is that the only reason for tak-

24 ing the scheme of interpretation corresponding to the lunar phases to be known to man during the Upper Pal- aeolithic is the very success of this correlation. Two, otherwise unjusti- fied suppositions thus rely on each other for their substantiation. There are actually two problems here: one is that Marshack claims to observe something without the aid of any scheme of interpretation; the other is that the scheme he eventually introduces does not account for his putative observations. In fact, in spite of his microscope, as Elkins (1996) has shown, Marshack has failed to observe numerous details of the configuration appearing on the bone, which makes it less probable that a correlation may be made to the lunar calendar, and thus that the in- scriptions are intentional. It is of course possible that Marshack’s lunar calendar is identical to the principle of pertinence used by prehistoric man, however implausible that may seem from his observations.17 From the point of view of pictorial semi- Fig. 8b. Drake’s figure otics, von Däniken’s claim that cer- inverted tain pre-technological images show scheme or Anati’s “prayer” (Sones- wristwatches seems at least as well son 1994a). substantiated as Marshack’s lunar In his history of interstellar 17 Unfortunately, Elkins (1996) uses messages, Vakoch also tells us about this case study to argue that the post- some ingenious ways of constructing structuralist point that “close readings” messages invented by Charles Cros are impossible, which is trivially true, if this is taken to mean that all details can be in 1869, by the Niemans in 1920, and observed, using no system of relevances by Drake in the 1960s (Fig. 6. and at all, but is disproved, on a more reaso- Vakoch 1999; 2001; 2003). Cros sug- nable interpretation, by Elkins own work, gested that several series of num- producing a “closer reading” than that of Marshack (Cf. Sonesson 1996b). bers should be sent, each one of

25 them having the same final sum. with their electrons (Fig.8). The re- When the numbers where translat- sult of the reconstruction, then, may ed into strings of beads of two dif- be said to be more of the same gener- ferent colours, and these strings were al kind as the better-known Pioneer aligned one over the other, a figure plaque (Fig. 9). would appear. According to the pro- The idea if of course that, if posal made by the Niemans, dots and these extraterrestrials beings are in- dashes would be used instead, again telligent, they will be familiar with corresponding to beads of differ- the same kind of mathematics as we ent colours, and the dots and dash- are, and they will know the same es of each string making up the same chemistry (and also, as I will insist sum. Drake’s proposal is of the same on below, that they would represent general tenor, but more complex: the them in the same way). Even grant- message sent consists of 551 bits of ed that, however, these proposals beg information, which is a number the the question: why would these hypo- only primes of which are 19 and 29. thetical extraterrestrials scientists When these numbers are taken to be believe, in the first place, that these the length and the side of the mes- are messages – which is the primary sage, respectively, the result will be requisite for them setting out to re- a pixelated pattern, which could be construct them. interpreted to be a “stocky biped”, Although these codings are placed beside the star and the nine much more complicated, they re- planets of our solar system, as well mind me of a parable constructed by an oxygen atom and a carbon atom Arthur Bierman, with the purpose of

Fig. 9. The famous Pioneer plaque, messages to other planets

26 Fig. 10a. A classic comic strip, ”The Upside-Downs” by G. Verbeck.

proving the impossibility of iconic- expected, it is indeed necessary to ity. This story, I submit, is instruc- have it “anchored”, to use the clas- tive in a different way from what sical Barthesian term. When open- Bierman intended. A man receives ing the parcel, the man will note a by mail a parcel, which turns out to number of things: it contains icon- contain something the man, takes ical signs, rather than writing or to be a blueprint. Using metal piec- scribbles, etc.; the particular style es, he sets about constructing a ma- of the pictures connotes “blueprint”; chine according to the blueprint, but and the shapes given to the figures when he switches it on, he is electro- suggest they depict machine parts. cuted. The next morning, his wid- These observations determine the ow receives a letter, explaining that use to which the man puts the gift: the figures marked on the paper have since it appears to be a blueprint, he to be cut out and put together, to ob- sets about constructing something; tain a paper machine. But is the mor- since the shapes of the pictures sug- al or this story really that there are no gest machine parts, and since ma- iconic signs? chine parts are usually made of some I think not. Like all activities sort of metal, he makes his construc- taking place in the Lifeworld, the in- tion out of metal pieces. Apparently, terpretation of pictures depend on there must also be some kind of sign, certain things being taken for grant- probably iconic or indexical, which ed, but not necessarily on any par- tell the man how to relate the differ- ticular conventions: “normal” con- ent pieces to each other. But Bier- ditions are thought to obtain. When man has been pulling the man on. a sign differs from what might be What seems to be a blueprint is re-

27 Fig. 10b. The story continued when the strip is turned upside down

ally a cutout sheet; instead of being as such, can explain depiction. Not pictures, the figures are self-presen- depiction, but the function of depic- tations; and what seem to be their tion, is at issue.18 borders are really indexical signs for In the present case, however, where one has to cut. incommensurability is much great- Interestingly, while instructions er. We have no reason to suppose the would be needed to discover that the sender and the receiver of an inter- sheet of paper could be seen as a stellar message to share such un- self-presentation (a secondary iconi- derstandings that permit the man in cal sign, as we shall say below), none Bierman’s story to make an interpre- was necessary for the man to take tation, even if it happens to be the it for a picture. If the sheet, consid- wrong one. Here it is true, in a much ered as an expression, were ambigu- more acute sense, that normal condi- ous between two readings, then one tions do not obtain. In fact, if depic- of them, which happens to be incor- tion, on the face of it, stands at the rect here, would seem to suggest it- beginning of Bierman’s story, it only self more readily. It should also be emerges as a result at the end of the noted, that there is no hint in the sto- coded messages aimed at extrater- ry that the man put the pieces togeth- restrials. This is, I think, a decisive er incorrectly: thus, something was difference. apparently read off from the picture Both Vakoch (1999) and Arbib iconically (and indexically). In this sense, Bierman’s parable presuppos- 18 For the parable, see Bierman es the truth of the very thesis it is (1963:249), for the argument above, cf. Sonesson (1989a:220ff and 1998; 2000; supposed to disproof, that similarity, 2001). 28 (1999) locate the problem of the Drake man beings, then they will certain- kind of message on the depiction ly not have any more problems find- end: the extraterrestrials would not ing what is upside down in the pic- be able to interpret it, they contend, ture, than perceiving the picture as if they tried to read it upside down, such. Nothing permits us to con- with the legs of the biped pointing clude, however, that extraterrestrials skywards. Indeed, Arbib even pro- share the ecological world character- poses a possible, but obviously er- istic of human beings. But all this ig- roneous, interpretation of the invert- nores the primary problems, which is ed image. Interestingly, the philoso- anterior to the depiction: why would pher Edmund Husserl long ago en- the extraterrestrials think there is a countered the same problem, without message at all? having to take the perceptual habits As we have seen, the pictures of extraterrestrials into account: he making up the blueprint are really suggested that pictures were essen- the givens of Bierman’s story: it is tially non-arbitrary, but that a con- the machine which is constructed in vention was needed for telling us their image or, as is happens, out of what was up and down. In as rejoin- them. Bierman’s formal arguments, der to Husserl (1980), I long ago re- however, rather go to prove that pic- futed the last part of this affirmation tures as such are constructs of our (Sonesson 1989a: 276ff): it is suffi- perception. As is well known, Nel- cient to turn a picture slowly around, son Goodman later on gave more fa- and at some point the configuration mous formulations to those same ar- giving rise to a depiction will emerge guments. The messages conceived of itself. This is nicely illustrated by by Cros, the Niemans, and Drake the comic strip “The Upside Downs” are really better illustrations than (Fig. 10ab), created by G. Verbeck that of Bierman’s story of this con- in 1903: at the end of each strip, you structionist theory of picture percep- have to turn the whole strip on its tion. Indeed, if human beings really head in order to follow the rest of the have to construct each picture before story. Thus, each drawing has a dou- perceiving it, then it is perhaps not ble interpretation, in which what was so strange to think extraterrestrials a hat may, after inversion, appear as would be able to do the same thing. a skirt, and so on. When you turn the Three schools of perceptual figure around, not only a new con- psychology are commonly distin- figuration Gestalt( ) appears at some guished. The most venerable one is given point - but also a new repre- known as constructivism and goes sentation. At least, so it is for human back to Helmholz, but has in recent beings. times most famously been repre- If extraterrestrials are like hu- sented by Gregory, who claims that

29 impoverished stimuli only can give against this, Gibson has claimed that rise to percepts thanks to social con- no conclusion about the real world structs. Neisser, who was an impor- can be derived from pictorial exam- tant representative of this school, ples. Although he never says so in so later on embraced ecological psy- many words, Gibson clearly suppos- chology.19 According to the second es the picture to be a sign. All an- school, Gestalt psychology, repre- imals can understand the meaning sented by Köhler, Koffka, Arnhe- of surfaces. But, according to Gib- im, etc., innate mechanisms organise son, only human begins can interpret perception, again based on impover- markings on a surface, that is, have ished data, into configurations. Ec- indirect perceptions. ological psychology originates with To see the picture as a picture the work of James Gibson, which clearly requires the capacity to per- has been pursued by Reed, Neiss- ceive wholes (Gestalts) as such; to er, Kennedy, Hochberg, etc.: accord- take contours to be equivalent to the ing to this conception, the principles sides of objects; and to accept the of “ecological physics” explain how 2D surface as a surrogate for a 3D percepts emerge from stimuli. Thus, world. The picture supposes a simi- it supposes human perception to be larity on the background of a funda- a function of the human ecological mental difference. But the problem niche or Umwelt, that is, in phenom- may very well be to see the differ- enological terms, of our lifeworld. ence rather than the similarity. Gib- Only ecological psychology, son (1978:231) observes that, besides however, seems to have anything rel- conveying the invariants for the lay- evant to say about pictures. The par- out of the pictured surfaces, the pic- adox of perceptual psychology is that ture must also contain the invariants ecological psychology is alone in at- of the surface, which is doing the tending to the difference between picturing: those of the sheet of paper, perceiving the real world and those the canvas, etc., as well as those of signs of it called pictures. Gestalt the frame, the glass, and so on. The psychology and constructivism of- difficulty, clearly, consists in seeing, ten use pictorial examples (configu- at the same time, both the surface rations and illusions, respectively) to illustrate real world perception.20 As mensional surface into a three-dimensional experience from the observation of 19 As for the brand-new version of pictures, without pausing to consider the constructivism proposed by Hoffman fact that what he then observes would be (1998; 2004), it seems to abandon all te- a two-dimensional surface representing a nets of the classical tradition and is hard to two-dimensional surface, or that pictures distinguish from ecological psychology. are precisely not experienced, but only 20 Indeed, Hoffman (1998) derives interpreted, as three-dimensional, diffe- all his laws for transforming a two-di- rently from the world of our experience. 30 and the thing depicted. come signs of themselves, of some The relative part played by ico- of their properties, or of the class of nicity and conventionality in a sign which they form part: a car at a car may be used to distinguish primary exhibition, a stone axe in the muse- and secondary iconicity. A prima- um showcase or a tin cane in a shop ry iconic sign is a sign in the case window, an emperor’s impersonator of which the perception of a similar- when the emperor is away, and a uri- ity between an expression E and a nal (if it happens to be Duchamp’s content C is at least a partial reason “Fountain”) at an art exhibition. for E being taken to be the expres- When Man Ray makes a picture of a sion of a sign the content of which billiard table, we need no convention is C. That is, iconicity is really the to recognise what it depicts. Howev- motivation (the ground), or rather, er, if Sherrie Levine’s (real, three-di- one of the motivations, for positing mensional) billiard table is to repre- the sign function. A secondary icon- sent Man Ray’s picture, there must ic sign, on the other hand, is a sign in be a label inverting the hierarchy of the case of which our knowledge that prominence of the Lifeworld. This E is the expression of a sign the con- shows that among the properties de- tent of which is C, in some particular termining the probability of an ob- system of interpretation, is at least a ject functioning as the expression of partial reason for perceiving the sim- an iconic sign is to be found three- ilarity of E and C. Here, then, it is the dimensionality rather than the oppo- sign relation that partially motivates site. the relationship of iconicity. If our capacity to experience That pictures are instances of pictures directly, as opposed to sec- primary iconicity is shown by the ondary iconical signs, depend on the child’s capacity for interpreting pic- particular lifeworld we are inhabit- tures when first confronted with ing, that is, on the ecology typical of them at 19 months of age (as demon- human beings as it has evolved on strated in Hochberg’s famous exper- the planet Earth, then there is eve- iment); as well as by the ease with ry reason to suspect that extrater- which pictures are employed by pop- restrial beings, however intelligent, ulations whose own culture ignores would not share this capacity with them – at least, as long as the cul- us: what are for us primary iconical ture in question is within the bounds signs would be secondary to them. of our own earth. On the other hand, While we function according to eco- we do have to learn that, in certain logical psychology, they would have situations, and according to partic- to follow the precepts of constructiv- ular conventions, objects which are ism. No doubt there are other phe- normally used for what they are be- nomena which are primary iconic-

31 ities to them, but which we would are right who think that our concep- could only hope to interpret, if ever, tion of mathematics, as well as our according to the regime of second- contemporary theories of physics, ary iconicity. astronomy, and chemistry, must be In the case of the biped of known to extraterrestrial beings, ei- Drake’s picture, once it is recon- ther because they accept the same structed as a picture, or the more ex- theories, or they have entertained plicit man and woman of the Pioneer them at some earlier stage of their plaque, the problem is not so much development (as we would recognize that the characteristic body shape of Newtonian physics in other intelli- human beings must be recognized. gent beings). This fact would only Even in a normal picture, we can be relevant to the content side of the only recognize objects of the world sign. Even in the case of natural sci- with which we are already famil- ences, the expression side of the signs iar, at least as to their general type. are wholly within the limits of our Thus, if the extraterrestrials have human lifeworld. Suppose that the different body shapes from ours, and extraterrestrials are very well aware have never seen human beings, they of hydrogen transitions, pulsars and obviously cannot recognize the hu- the layout of our solar system. It is man shape. But the more general is- still very improbable for them to use sue involves the possible embodie- the same markings of the surface to ments of signs themselves. As I not- convey them to others as we would. ed above, the faculty to interpret pic- Their lifeworld would most certainly tures at least presupposes the ability predispose them differently. to perceive wholes as such, to take It is still possible that iconici- contours to be equivalent to the sides ty, in a wider sense than pictorality, of objects, and to accept 2D forms may be of some help. Peirce pointed as stand-ins for 3D objects. There is out that iconic signs convey more in- no particular reason for supposing formation than is contained in them, that this forms part of the ecology of thus, “with two photographs you can extraterrestrial beings. make a map”. This property, which In a more general sense, these Greenlee called “exhibitive import” observations are also valid for mark- (Greenlee) depends on our knowl- ings on a surface that are not pic- edge of the lifeworld: because of our tures. If our ability to interpret pic- familiarity with the layout of the life- tures is part of our competence as world, we are able to fill in the blanks inhabitants of the human lifeworld, in the representation. We can “see then all other sign system may well in” what we know should be there. by dependant on the same particular Therefore, if the extraterrestrials live ecological niche. Suppose that those in a different lifeworld, which they 32 most certainly do, they would be un- peculiar human lifeworld. able to derive any help from exhibitive import.21 Summary But perhaps there is another The picture must be understood as type of iconic surplus: something a sign, which implies that it is both which we might call “introversive similar to what it represents, and semiosis”, echoing a term used by different from it. This is where it Jakobson for signs referring to oth- becomes problematic: even though er signs rather than to the world. pictures are not conventional (to any Peirce’s favoured example of iconic- large extent), contrary to what has ity was mathematical expressions. been argued by many semioticians, Jakobson discovered an iconicity in some experience is needed to be able grammar. Such projections of the se- to interpret them as such. We know lection axis onto the axes of combi- that children need some time to gain nations, in Jakobson’s phrase, is rem- this knowledge, and other animals, iniscent of those messages with a with the possible exception of some regular structure which Arbib (1999) of the great apes, never acquire it. suggests should be used in commu- Archaeology gives of very little help nicating with extraterrestrials. More in understanding the origin of pic- importantly, perhaps, what would tures, because some artefacts which be needed are expressions that mir- have como to our knowledge cannot ror the system character of the sys- be reliably shown to be pictures or tem. This might be feasible if there is other kinds of meaningful displays, what Deacon (2003) has called “sem- and some artefacts which are clear- iotic constraints”: generalities of all ly pictures cannot be interpreted to “conceivable” semiotic systems. Of show all what they are usually taken course, like earlier philosophers such to show, because of the lack of an as Husserl, Deacon is generalizing appropriate knowledge of context. from the case of logic and mathemat- The latter observation also goes for ics to less tightly organized system of the interpretation of objects stem- the kind of verbal language (cf. Son- ming form other cultures, notably esson 2003a; forthcoming a, b). where we cannot suppose the presence of any anthropological univer- In conclusion, we have seen salia, which would be the case with that pictorial iconicity is dependant beings from other planetary systems. on the peculiar human lifeworld, but Moreover, if some picture could be that more abstract kinds of iconicity shown to be the earlier one of those may stand a greater chance of giving which we are aware of, this does not rise to messages going beyond the mean that it is the earliest of the pictures made by humankind, not only 21 Cf. Lecture 3.

33 because there may be earlier pictures good reasons for retaining the notion to be found, but also because the first of sign (or representation) for some pictures may have been made on kinds of meanings, while denying its sand, or some other highly precari- application to other instances. So be- ous surface. All this poses, in addi- fore we even ask ourselves whether tion, the question of how something there truly is such as thing as the sign, is recognised as being a sign. If the we have to be clear about what it is. sign is a stage in the development of This involves not only deciding the semiotic resources, then the char- criteria for analysing a phenomenon acterizations of the sign by Peirce of meaning into two separate parts, as well as Saussure are insufficient. but also those allowing us to posit an We need to understand how the sign asymmetrical relation between these is different from more elementary parts: not only does the expression kinds of meaning given in the per- have to be separate from the content, ceptual world. This is the task of the but the former should stand for the following sections. latter, not the reverse. It should be clear by now why 2.2. Signs in the Human we need such a concept of sign: the Lifeworld picture has been shown to be some- It is true of both the main traditions thing difficult to grasp, both to small of semiotics, the Saussurean and the children and to non-human animals, Peircean one, that they have never re- because it supposes the conscious- ally offered any definition of the sign; ness of a difference as well as of a and the same thing no doubt applies similarity. Perception and other di- to the notion of representation in rect acts of consciousness are not cognitive science.22 This goes a long difficult in this way: they appear to way to explaining why many semi- be fairly straightforward to children oticians (such as Greimas, Eco, etc.) and animals alike, rather early on in have rejected the sign, without much the development of the former. This of an argument, and why the second also applies to some unconscious or generation of adepts to cognitive sci- semi-conscious conclusions drawn ence (such as Lakoff, Johnson, etc.) from perceptual premises, as we shall now seems to be doing the very same see. In the concept of representation thing with respect to the notion of rep- of classical artificial intelligence, as resentation. There might however be well as of a lot of contemporary cognitive science, simple acts of percep- 22 A more interesting interpretation tion and sign consciousness are in- of Peirce, however, may be that he was not extricably confused. Although Saus- really interested in the sign in our sense. sure’s concept of sign was no doubt We will turn to that view in section 3 below. unambiguously restricted to mean- 34 ingful entities comprising two rela- ever, is that this conceptual tighten- ta which were clearly differentiating of the sign concept is a clear gain ed form each other and related by an coming out of latter-day philosophy, asymmetrical relation, French struc- although it must be regretted that the turalists such as Barthes and Grei- reasons for narrowing down the sign mas later on applies semiotical terms concept were never clearly brought to objects of meaning which could out. This should in no way be con- hardly be conceived to fulfil these strued as a nominalist stance, as it requirements, such as food, clothing, might have appeared during the Lat- and the world of perception. As John in Age, as Deerly shows: as a vieux Deely (2001) has shown, philosophy combattant of the critique of nomi- written in Latin during the Middle nalism, I would certainly not opt for Ages and in the following centuries such a solution.24 On the contrary, it long hesitated between a restricted is precisely because signs and per- definition of the sign, derived from cepts are so different, although they the works of the church father Au- also have something in common, that gustine, and a much broader one, ac- they must be terminologically sepa- cording to which the contents of con- rated. sciousness should be considered sig- This is why it will be neces- nifiers for which the things of the per- sary to immerse ourselves not only ceptual world were the signifieds, -fi into what I will call the semiotics nally opting for the latter solution in of the Saussure-Piaget tradition but the work of Jean Poinsot in the early also in that of the Augustine-Hus- 17th century. Deely thinks the disso- serl tradition. Saussure merely pos- lution of this all-embracing concept ited two units making up the sign, of sign was a serious failing of early but Piaget introduced the criterion Modern philosophy.23 My view, how- of differentiation in order to sepa- 23 It might sound here as if classical rate signifier and signified. Saint -Au cognitive science has brought to fruition gustine, who has often (as so many the “post-modern” view re-establishing others) been hailed as the first sem- the broad sign definition of the Latin Age, iotician, defined the sign as “a thing as anticipated by Deely (2001). However, representation, which is a term with a which, over and above the impres- long history in philosophy and psychology sion it makes on the senses, causes taking on many different senses, is largely something else to come into thought an undefined term in cognitive science. as a consequence” (as translated by Deely would probably criticize cognitive 25 science making the same reproach as he Deely 1982: 17f). Husserl’s defini- makes to Locke and the British empiricists said below). generally, that they treat the whole domain as being that of “ideas”. In so doing, I take 24 Cf. Sonesson 1989a, 1995, and it, they fail to see the relational character Lecture 4 below. of this domain (on which more will be 25 Deely (2001: 221) renders

35 tion of the sign, which describes the as different as they may seem; but I expression as something which is di- will also submit than none of them, rectly perceived but not in focus, and on their own, is able to resolve our the content as being indirectly per- problem. ceived while at the same time be- There are several ways to read ing the focus of the relation, could be Peirce and, conceivably, Saussure: taken as a way of specifying the Au- one, very common one, consists in gustinian suggestion. It implies that looking upon these writings as a de- the sign is asymmetrical in a double vout Christian approaches the Bible, sense: one part of it is more in focus as the source of all truth, even that than the other, and the other of its discovered since the time of writing, parts is more directly accessible than using some often very subtle opera- the first one. tions of interpretations to extract it. A procedure similar to this one may Two classical views of the sign actually be justified, if the aim is not – and beyond to develop an adequate semiotic the- I have argued that a well-defined ory, but simply to establish what the concept of sign is needed, in order to teachings of the founding-fathers re- understand the specificity of the pic- ally were. Another approach, which ture, both in child development and is not the one I am going to preconize in human phylogeny. In semiotics, it either, is, of course, to read Peirce often seems as the only game in town and Saussure as that rival potentate, consists in showing that the concept the Devil, is supposed to read the Bi- of sign needed is provided by Peirce ble, by inverting the meaning of eve- but not Saussure, or perhaps some- ry line: this may at first appear to be times the reverse. For those who want a purely fictional possibility, but I do to go on playing this game, what fol- think a procedure very much like it lows will be doubly disappointing: was applied by the French structur- not only will I claim that the concep- alists as well as Eco in the sixties tions of Saussure and Peirce are not and the seventies of the last century, less perhaps to Saussure and Peirce, Augustine’s definition somewhat differently: “a sign is anything perceived which but more to one of the most eminent makes something besides itself come into followers of the first, Hjelmslev (cf. awareness” (but he also quotes another Sonesson 1989a). definition more similar to the one referred to above). Perhaps “perceived” is the same If we cannot read our classics thing as “impression made on the senses”. like true converts, nor like the Dev- As we will see, it is not the sense charac- il, there remains, of course, the poster that we will retain here, but the divi- sibility of reading them like God (or sion into two items clearly separated from each other, one of which is more directly the Pope): and while this may seem accessible. a much too presumptuous alternative 36 Expression substance

Fig.11. The Saussurean sign: a) Expression form the double division of the sign into expression and content and form and substance, resulting in double articulation; Content form

Content substance to be seriously entertained, it comes metaphors, we will say that Peirce, close to what I think we should ac- Saussure and the others were wise tually do, if we are able to conceive men, great scholars, whose thinking of a eminently Peircean God, not, of today still is worth-while taking seri- course, the one in which Peirce hap- ously; but they were also very much pened to believe, but one which func- fallible, and so, in our own extremely tions according to the Peircean mod- fallible way, we may sometimes be el of the mind; a very much fallible able to do a little better than they did, God who is always still trying to ap- often because we have access to the proach the truth, without ever getting work of others scholars they did not there, yet always approaching it a lit- know about. It should be added that tle more, seeing a little further, be- the intrinsic fallibility of all work, cause he is standing on the shoulders even that of giants, is compounded, of giants. Our giants are, of course, in the case of Peirce and Saussure, Saussure and Peirce, Hjelmslev, Pri- buy the fact that almost none of their eto, and many others. And so, in or- works were ever published in their der to start entangling our chain of lifetime or even made ready for pub-

Fig.11. The Saussurean sign: b) how the sign is related to reality by means of the two forms.

37 Thing considered as the Thing considered as the expression content

Iconical or indexical ground

Semiotical or sign function

Fig.12. The Peircean sign: a) the sign as a mapping between two things taken to be the expression and the content, respectively, and related by a “ground” singling out properties in those both objects, which are either similar (iconicity) or connected (indexicality), independantly of the mapping operated by the sign function; lication, and, especially in Peirce’s cise and revise them accordingly. On case, by the fact that his thinking the other hand, there can be no doubt evolved during the long spate of time about Peirce being a very profound he was working on semiotic issues, thinker (though perhaps not in every and that he appears to have made a paragraph he wrote), so I really think lot less close reading of his own ear- we should try to do him full justice. lier work than his latter-day com- When there are several possible in- mentators do. terpretations of his works, and when It should be clear, then, that we different passages contradict each cannot be interested here in discov- other, we should choose the one most ering “what Peirce really said”; rath- favourable to him — from the point er we will be making use of his con- of view of present-day semiotics. Al- cepts to the extent that they fit in with though I love Peirce very much, I what has since then been established love truth even more: so while some by semiotical reasoning and psycho- things I say in the following may be logical findings, and we will criti- false as interpretations of Peirce, I

38 Fig.12. The Peircean sign: b) the sign as consisting of different phases of interpretation, making up six phases, from the representamen to the final interpretant still think they are valid as compo- is also triadic. But it may reasona- nents of contemporary semiotic the- bly be maintained that reality is sim- ory. ply that which is excluded from the This brings us to the notorious Saussurean sign as being irrelevant issue of Saussurean binarity as op- (although Saussure never was as exposed to Peircean triadity, which is plicit about this as the early Eco). a point of contention which will be However, it might be argued that the completely absent in the following. referent is important in the Saussure In spite of Peirce’s explicit denial, I conception, as being that which is do think he was something of a tri- divided differently by different lan- adomaniac. But that is not the real guages and other semiotic resourses. issue. It may often be convenient to From another point of view, the do- order things in rows of threes. But main may be said to be the signifi- the whole question whether there er, the signified, and the relation be- are two or three of something has no tween them, which would definitely sense whatsoever, before we know make the sign triadic. And this is a what kind of entities we are talking more valid point, since the sign as a about. The question whether some- unit of signifier and signified is very thing has two or three parts has no important to Saussure. Then again, meaning before determining the do- the Saussurean sign might really be main for which the model is valid, as claimed to be polyadic: to Saussure, well as the criteria (the relevant prop- as is well-known, even the sign is a erties) according to which the divi- superficial manifestation of the mul- sion is made. titafious interrelationships making If the domain is the sign, made up the sign system, in which every- up of signifier and signfied, plus re- thing determines everything else. ality, the Saussurean sign definition On the other hand, there is cer-

39 tainly no denying that the Peircean content at it appears to the addressee sign is triadic, but these triads are (cf. discussion in Sonesson 2003a). If then subdivided, where that which object and interpretant correspond to is of the nature of Secondness has something akin to speaker’s mean- two parts, and that which is of the ing versus listener’s meaning, then nature of Thirdness has three parts. the communication models (notably If all these distinctions are criterial, that of the Prague school) also ac- Peirce’s definition actually has six count for it. If the interpretant has levels. If the triadity of the Peircean something to do with the notion of sign really had involved some- “ground” appearing in Peirce’s early thing like the expression, the contexts, then it figures prominently in tent, and the real world (as many the Saussurean tradition in the form have been fooled by Ogden & Ri- of the distinction between form and chards to think), then it would have substance, mentioned below. This been present also in the Saussurean last interpretation is favoured, in my conception, the third item appear- view, by Peirce’s (1998: 269) conten- ing as that which is explicitly exclud- tion that “Thirdness [e.g. interpre- ed from consideration (and which is tants] is found whenever one thing then reintroduced by most post-Saus- brings about a Secondness between sureans). It rather seems as if the dis- two things [e.g. the relation between tinction between the content and the representamen and object].” referent were mimicked in Peirce’s For our purpose then, we will work by that between the immediate say that the Saussurean sign is made and the dynamical objects, so when up of expression and content (signifi- we add the interpretant, we end up ant/signifié) which both can be sepa- with four objects. However, just as rated into form and substance - and there are two objects, there are three it is separated from reality (the ref- interpretants (but only one repre- erent). “Form” here is that part of the sentamen), so there are really six in- expression which cannot be changed stances of the sign altogether. Using without giving rise to another con- another kind of reasoning, one may tent, and vice-versa; “substance” is instead add the utterer and the inter- all the rest (cf. Fig 11.). The Peircean preter, and then end up with a penta- sign consists of expression (repre- gram (cf. Dines Johansen 1993). In- sentamen), content for the initiator of deed, some unpublished passages in the sign (object) and content for the Peirce’s manuscripts (for instance, target of the sign (interpretant). The MS 318, quoted in Jappy 2000) seem signs ”tends” towards reality. This is to suggest that the object is simply the why the “dynamical object” is closer content as conceived by the address- to reality (and further from the origi- er, and the interpretant is the same nal sign situation) than the “immedi-

40 ate object”; similarly, the “dynamical viously hinges on what kind of rela- interpretant” is closer to reality (and tionship there is between these two further from the original sign situa- or three parts. This is no doubt im- tion) than the “immediate interpre- plicit in terms such as “expression” tant”; but even further from the sign and “content”. But if the concept of situation is the “final interpretant” sign should be of any use, that which which is only virtually present. Per- is implicit has to be spelled out. haps it would be more correct to say that the object is that which influenc- From pebbles to feathers. The es the creator of the sign so as to cre- notion of differentiation ate it, while the interpretant is that Let us start out from what might be which influences the receiver so as to called the Saussure-Piaget tradition. interpret it. Then the different kinds I am not sure whether anybody has of objects and interpretants would be ever stood in that tradition, except, phases of this process (cf. Fig 12.). of course, Piaget, who took all his There are no doubt some real semiotic vocabulary (opposing the differences between Saussure and sign to the symbol) from Saussure.26 Peirce, however. Saussure is really What Piaget added to Saussure was only interested in the linguistic sign most obviously a developmental per- whereas Peirce wants to characterize spective, in particular on the level of all possible signs. Peirce sometimes ontogeny. But, just as importantly, seems to extend the sign so far that it though it is less commonly observed covers everything. Peirce’s concepts (in fact never, except for Sones- can only with difficulity be separat- son 1992b, etc.), he realised that all ed from a specific philosophical con- meanings are not signs, and he even ception of reality. Peirce’s model began groping for a definition of that seems to be more involved with the which accounts for the specificity contact between the sign and reali- of the sign. More decisively, apply- ty, while Sassure is concerned with ing the developmental perspective to their difference. the sign, he made it into a particu- But they have one thing in com- lar stage of development (although, mon: none of them really tells us what unlike Vygotsky, he never allowed a sign is. It often seems as if anything semiosis to define that stage). which has three (or two) parts would 26 Sonesson (1989a; 1992a) certainly thereby be a sign. It is true that this stands in that tradition, and, as I discove- is a problem less with the Saussurean red very recently, so does Krampen (1991), than with the Peircean conception, who appears to be the only semioticians, since Saussure is adamant about pos- apart form the present author (and to some extent, from Bentele 1984), who has taken ing verbal signs as the best instance an interest in Piaget’s notion of semiotic of the category. But everything ob- function.

41 When Peirceans and Saussure- function (there is a differentiation be- ans quarrel over the presence of two tween the latter two instance, in the or three entities in the sign, they never double sense, I take it, that they do pause to ask themselves what kind of not go over into each other in time objects, defined by what type of fea- and/or space, and that they are per- tures, are involved: but, clearly, be- ceived to be of different nature). fore we know what we are counting, According to Piaget the semiot- it makes no sense to start counting ic function (which, in the early writ- at all. The whole question becomes ings, was less adequately termed moot, if there is no reason to analyse the symbolic function) is a capaci- meaning into two parts, as suggest- ty acquired by the child at an age of ed by both contemporary cognitive around 18 to 24 months, which ena- scientists and old-time existentialists bles him or her to imitate something and Lebensphilosophen. What, then, or somebody outside the direct pres- is it that permits us to determine that ence of the model, to use language, an object endowed with meaning is make drawings, play “symbolically”, made up an expression, or “repre- and have access to mental image- sentamen”, and a content, or “object” ry and memory. The common fac- (analysable into “immediate” and tor underlying all these phenomena, “dynamic”)? Peirceans and Saus- according to Piaget, is the ability to sureans alike would no doubt agree represent reality by means of a sig- that signs has something to do with nifier which is distinct from the sig- the classical formula, often quoted nified. Indeed, Piaget argues that the by Roman Jakobson (1975), aliquid child’s experience of meaning ante- stat pro aliquo, or, as, Jakobson also dates the semiotic function, but that puts it, more simply, with “renvoi”, is does not then suppose a differenti- or reference. What this means, how- ation of signifier and signified in the ever, is not at all clear. sign (see Piaget 1945; 1967a; 1970).27 Before we can separate signs The notion of differentiation, which from other meanings, we have to spell out those criteria for something 27 Not all of Piaget’s examples of the semiotic function may really be of that being a sign which are simply taken kind, even applying his own criteria. Cf. for granted, both in the Peircean and Sonesson 1992b. It should be kept in mind in the Saussurean tradition. This can that Piaget is here talking about the capa- be done by combining what Husserl city for producing language, pictures, and so on, not the ability to interpret them. As says about appresentation (something in the case of language, the capacity to which is directly present but not the- understand pictures must precede any abi- matic refers to something which is lity to produce them. However, if under- indirectly present but thematic) and standing really arrives as late as Deloache claims, as we saw in the last section, there what Piaget says about the semiotic is still a conflict with Piaget’s view. 42 is normally overlooked, is funda- The signifier of the index, Pi- mental in my view.28 In several of the aget says, is “an objective aspect of passages in which he makes use of the signified”; thus, for instance, the this notion of semiotic function, Pi- visible extremity of an object which aget goes on to point out that “indi- is almost entirely hidden from view ces” and “signals” are possible long is the signifier of the entire object before the age of 18 months, but only for the baby, just as the tracks in the because they do not suppose any dif- snow stand for the prey to the hunter. ferentiation between expression and But when the child uses a pebble to content.29 In this way, Piaget really signify candy, he is well aware of the anticipates the critique formulated difference between them, which im- by Colwyn Trevarthen (see Trevar- plies, as Piaget tells us, “a differen- then & Logotheti 1989), according to tiation, from the subject’s own point which the child is attuned to mean- of view, between the signifier and the ing, not only from birth, but in fact signified”. Between “indices and sig- already at the end of the fetal stage: nals”, on the one hand, and full signs, co-operation, and the capacity to on the other, moreover, Piaget places pick-up others meanings, is some- “symbols”, understood more or less how built into the organism. Clear- along the lines of Saussure. These ly, meaning is here used in a more “symbols” are already differentiat- general sense than that characterised, Piagets claims, but their parts are tic of the semiotic function, that is, still somewhat “adherent”. In addi- the sign, as I have tried to develop tion, this adherence seems to apply this notion taking my hints from Pi- as least as much to the relation be- aget and Husserl: it includes percep- tween the subject and the semiotic tion, particularly of an inter-person- resources he or she makes use of as al kind.30 to the relation between the signifier and the signified.31 28 Vygotsky (1999) also observes the difference between differentiated signs Piaget is quite right in distin- and other meanings, but he has no termi- guishing the manifestation of the nology for capturing the distinction... semiotic function from other ways of 29 Krampen (1991: 14ff) fails to see the problem here, perhaps because he more or less automatic imitation in the quotes Piaget in an English translation, infant, different from the explicit capacity which renders the French term “indice” for imitation, which matures much later. (that is, “index”) by the locution “signs or Cf. Gallagher (2005) and Mandler (2004). pointers”. 31 Piaget also insists a lot on the 30 This also brings Trevarthen to individual character of the symbol and the challenge the inclusion of imitation among social one of the sign. Therefore, Kram- the aspects of the semiotic function. As pen (1991: 18f) is clearly wrong in identi- we now know, mainly due to the work fying Piaget’s “symbol” with Peirce’s icon of Meltzoff, there is a very early stage of and Piaget’s “sign” with Peirce’s symbol.

43 Firstness Secondness Thirdness

Principle Iconicity — —

Ground Iconic ground Indexicality = — indexical ground

Sign Iconic sign Indexical sign Symbolicity = (icon) (index) symbolic ground = symbolic sign (symbol)

Fig. 13. The relationship between principles, grounds, and signs, from the point of view of Peirce. “connecting significations”, to em- employ a pebble to stand for a rock, ploy his own terms. Nevertheless, it without therefore confusing the part is important to note that, while the and the whole: then the child would signifier of the index is said to be an be employing a feature, which is ob- objective aspect of the signified, we jectively a part of the bird, or the rock, are told that in the sign and the “sym- while differentiating the former form bol” (i.e. in Piaget’s terminology, the the latter from his point of view. Only conventional and the motivated vari- then would he be using an index, in ant of the semiotic function, respec- the sense in which this term is em- tively) expression and content are ployed in semiotics, that is, as true differentiated from the point of view sign. In terms of socially more well- of the subject. Curiously, this dis- established signs, a similar example tinction between the subjective and would be the bull’s head used to indi- objective points of view is something cate, above a market stand, that beef Piaget seems to forget about in the is sold there. Although in France, for following. We can, however, imagine exampel, sculpted heads of bulls or this same child that in Piaget’s exam- horses are employed outside the rel- ple uses a pebble to stand for a piece evant shops, it is still possible to find of candy having recourse instead to a real heads used in traditional mar- feather in order to represent a bird, or kets in some countries.

44 The hunter, on the other hand, tify them, at least until they change who identifies the animal by means their perspective on the object by ap- of the tracks, and then employs them proaching it from another vantage to find out which direction the ani- point. And at least the adult will con- mal has taken, and who does this in sider a branch jutting out behind a order to catch the animal, does not, wall as something which is non-dif- in his construal of the sign, confuse ferentiated from the tree, to use Pi- the tracks with the animal itself, in aget’s example, in the rather different which case he would be satisfied sense of being a proper part of it.33 In with the former. Indeed, if the tracks the Peircean sense an index is a sign, are not differentied from the animals the relata of which are connected, in- having produced them, they can not dependently of the sign function, by be read as signs, but only as a part contiguity or by that kind of relation of the complex situation of which the which obtains between a part and the animal is a part. Both the child in our whole (henceforth termed factorali- example and the hunter are using in- ty). But of course contiguity and fac- dices, or indexical signs, where the torality are present everywhere in “real” connection is transformed into the perceptual world without as yet a differentiation in the sign.32 forming signs: we will say, in that On the other hand, the child case, that they are mere indexicali- and the adult will fail to differentiate ties. Perception is perfused with in- the perceptual adumbration in which dexicality.34 he has access to the object from the An index, then, must be un- object itself; indeed, they will iden- derstood as indexicality (an indexical relation or ground, to use an old 32 According to some current con- Peircean term) plus the sign function. ceptions, this would not necessarily be Analogously, the perception of simi- true in prehistory: chimpanzees and early humans appear to be unable to make use larities (which is an iconic ground) of tracks in their hunting behaviour, if cognitive archaeology is to be trusted 33 About proper parts, perceptual (Mithen 1996: 73ff). Actually, Mithen’s perspectives, and attributes as different examples suggest that apes are able to ways of dividing an object and thus diffe- interpret auditive signs of the hunted rent indexicalities, cf. Sonesson 1989a,I.2; animals, but will not even recognise the as well as Lecture 4 and 7. animal itself if presented with it visually, 34 I am using “indexicality” here which suggests indexicality is not invol- (just as “iconicity”) in the sense of so- ved at all in this distinction. Indeed, many mething which is necessary for a sign animals “lower” on the evolutionary scale being an index (or an icon), but which, are obviously able to interpret traces. analogously to the quotation from Peirce According to this conception, the develop- below, cannot function “as a sign until it ment of “art”, i.e. picture signs, is an even is embodied”. See, in particular, Sonesson later accomplishment of human prehistory 1992a, 1993a, c, 1994a, b, 1998a, 2000, (Mithen 1996: 150ff). 2001a, b, 2003a – and the next lecture!

45 will only give rise to an icon when it ing of the term, is not even a rela- is combined with the sign function. tionship; but once two iconicities are I therefore cannot agree with Dea- experienced together, they form an con (1997: 76ff) when he claims that iconic ground, which is an relation, camouflage in the animal world such but still not a sign. It is the experi- as the moth’s wings being seen by ence of bark on one place being sim- the bird as “just more tree” are es- ilar to bark higher up or lower down; sentially of the same kind as those or of the tree being similar to anoth- “typical cases” of iconicity we are er tree. A picture of a tree, however accustomed to call pictures. As al- (or even a tree on a theatre scene) is ways, there are passages in Peirce’s an iconical sign (cf. Sonesson 2003a work which may be taken in differ- and Fig. 13). ent ways, but it makes more system- While the introduction of the atic and evolutionary sense to look notion of differentiation is a substan- upon iconicity and indexicality as tial accomplishment on the part of being only potentials for something Piaget, he unfortunately never spells being a sign which still have to be out its import. As I have mentioned “embodied”, as Peirce (1998: 291) above, he defines it in terms of the suggests regarding another division subject’s point of view, but then uses of signs: examples in which the disconnec- A Qualisign /---/ cannot actually act as a tion already exists objectively. The sign until it is embodied; but its embodiment sense of objectivity and subjectivity has nothing to do with its character as a sign. A Sinsign /---/ involves a qualisign, or rath- employed here should of course be er, several qualisigns. But these qualisigns related to the common sense world are of a peculiar kind and only form a sign through being actually embodied. (that is, the Lifeworld) in which human beings stake out their life. In- An indexicality, then, is not a deed, what Piaget is concerned with sign; it is simply the perception of is precisely the “construction”, in his two things being connected. It will terms, by the child of the common be a sign only once these items are sense world. Once this edifice is fin- experienced as being detached from ished, the common sense world dis- each other. The foot touching the joins that which is subjective (which earth is an indexicality; the traces does not mean particular to one in- left on the soil is an indexical sign. dividual, but may very well be the The branch of the tree which is still “world view” of a particular lan- part of the tree is an indexicality; in guage, the way of segmenting reality the theatre, however, where it is cut opposing pictures to language gener- off from the tree, it may well be an ally, etc.) from that which is objective indexical sign for it. Strictly speak- (which is, strictly speaking, the sub- ing, iconicity, in Peirce’s understand- jectivity common to human beings).

46 But, in his later reasoning and ex- ferent general category of the world amples, Piaget seems to identify dif- than the signified. But these are per- ferentiation form the subject’s point haps less criterial attributes than fea- of view with conventional, or arbi- tures helping us to pick our exam- trary, signs, in the Saussurean sense. ples out. The basic idea, again, is no This will not do, for already “sym- doubt in the opposition between the bols”, in the Saussurean (and indeed two itemes being subjectively, rather Piagetean) sense, are differentiat- than objectively, separate from each ed in this way. Indeed, Piaget claims other. It is here that, probably with- that “symbols”, in his sense, are dif- out knowing it, Piaget is the most ferentiated, but still “adherent”, but Saussurean. I am thinking about the it is not clear what this means, and passage in which Saussure said that he never uses examples of this type semiotic resources were points of to illustrate differentiation. More view taken on material things (and, importantly, perhaps, he fails to see we could add, on the world general- that some indexical functions are not ly). It is in becoming a standpoint on mere “pointers”, but real, differenti- the world than the sign separates out ated signs, such as is the case with from the world.36 the pointing finger and the tracks as Nor should differentiation be interpreted by the hunter.35 identified with displacement as de- Indeed, the basic problem may fined by Hockett (1977), which well be that, in Piaget’s work, dif- (rightly, no doubt) appears as one of ferentiation is never defined. I have the “design features” of language in suggeted above that differentiation most introductory textbooks.37 As may be a result of the object which in the case of the tracks left by the serves as signifier not being continu- 36 This is the origin of what Deely ous in space and/or time with the ob- (2001), following the Latins, calls “mind- ject serving as signified, as well as dependent” (ens rationis) versus “mind-in- of taking the signifier to be of a dif- dependent being” (ens reale). Interestingly, Searle (1995), who talks about “language- 35 Other pieces of valid criticism may dependant” and “language-independent be levelled against Piaget, as discussed in facts” in what appears to be a similar Sonesson 1992b: the point that meaning sense, sometimes slips into the alternative emerges ontogenetically well before the terms referring to the mind, although he attainment of the semiotic function (as ex- would certainly deny having read any pressed notably by Trevarthen) is essential philosophy earlier than Austin. In the next to the following argument. The observa- section, I will suggest that this division is tion, made experimentally by Gardner & incomplete. al., that the semiotic function is not at- 37 And it has nothing to do with tained in different media, and in different Hjelmslev’s criteria for something being a respects, at the same age, is important, sign, the possibility of separating expres- but has nothing to do with the functional sion and content into smaller parts inde- definition of this stage of development. pendently. See Sonesson 1992a.

47 hunted animal, displacement may ular perspective. For this is requires be a consequence of differentiation. independence: that is so say, a body But differentiation only comes on of its own. Thus, the notion of differ- its own when the sign is in presence entiation itself needs to be clarified. of its referent, for then it allows us to construe reality in different ways Different ways of “connecting (“subjectively”, as Piaget would have significations” said), picking out that which is rele- The notion of differentiation has vant, and ignoring, or downplaying certainly not been satisfactorily de- other features. fined in these pages: expression and We must be careful not to con- content, I have suggested, do not go fuse different relationships involving over into each other in time and/or the sign. Differentiation, in Piaget’s space, and they are perceived to be sense, must pertain to the signifier of different nature. To get any fur- and the signified, which are always ther, both phenomenological and ex- equally present in the here and now perimental investigations are in or- of the sign user, since they are men- der. Some clarification of this issue tal (or, in most cases, intersubjective) when be given when we attend to the entities. To the hunter, both the sig- Augustinean-Husserlean tradition nifier and the signified of the tracks for the definition of the sign. All we are present here on the soil (or, to be can do at present is pointing out the precise, in his perception of the soil). contrast obtaining between signs and But the signified contains the infor- other kinds of meaning. mation that it is itself only part of Each time two objects are per- a larger whole (or rather something ceived together in space, there is once contiguous to a larger whole) contiguity; and each time something which was present here at an earli- is seen to be a part of something else, er time, but which is now elsewhere, or to be a whole made up of many more precisely further on in the di- parts, there is factorality (as defined rection indicated by the tracks. And in Sonesson 1989a). According to the displacement, in Hockett’s sense, Husserl, two or more items may en- has taken place between that signi- ter into different kinds of “pairings”, fied whole and the real animal which from the “paired association” of two is now present somewhere else. co-present items (which we will call When the sign, whether it is a perceptual context), over the “appre- stretch of discourse, a picture, or an sentative pairing” in which one item animal track, is present along with is present and the other indirectly the referent, however, the signified given through the first, to the real allows us to refocus the referent, in sign relation, where again one item other words, to present it in a partic- is directly present and the other only 48 indirectly so, but where the indirect- one phase of his action to know what ly presented member of the pair is has gone before and what is to come: the theme, i.e. the centre of attention that he has just raised his tool from for consciousness (cf. Husserl 1939; some lower level, and that at the next Luckmann 1980). moment, he is going to hit the trunk Whereas the items forming the of the tree. If we take a snap-shot of sign are conceived to be clearly dif- one of the phases of the wood-cut- ferentiated entities, and indeed as ter’s work, we could use it, like the pertaining to different “realms” of re- well-known traffic sign meaning ality, the “mental” and the “physical” “roadworks ahead”, as a part for the in terms of naive consciousness, the whole or, more oddly perhaps, as a items of the perceptual context con- phase signifying contiguous phases. tinuously flow into each other, and There has been a radical change from are not felt to be different in nature. the flow of indexicalities occurring In fact, both content and expression in reality, for not only is there now of the sign are actually “mental” or, a separation of expression and con- perhaps better, “intersubjective”, as tent “from the point of view of the classical Saussurean linguists would subject”, but this separation has been insist; but we are interested in the objectified in the picture. The picture respect in which the sign user con- is a sign, in the sense of it having a ceives them to be different. Piaget’s signifier which is doubly differenti- notion of differentiation is vague, ated from its signified, and which and in fact multiply ambiguous, but, is non-thematic and directly given, on the basis of his examples, two in- while the signified is thematic and terpretations can be introduced: first, only indirectly present. the sign user’s idea of the items per- The perceptual continuum may taining to different basic categories be reconstituted in a film, but not in a of the common sense Lifeworld; and, series of pictures. However, when we in the second place, the impossibili- ask the wood-cutter to stand still for ty of one of them going over into the a moment (like in a “tableau vivant”), other, following the flow of time or his position as such, before it is trans- an extension in space. formed into the motif of a picture, is Suppose that, turning around a already a sign for the whole of the corner of the forest path, we sudden- action, although the directly presently catch a glimpse of the wood-cutter ed position does not seem to be non- lifting his axe over his shoulder and thematic, continuity is only provi- head. This experience perfectly illus- sionally interrupted, and expression trates the flow of indexicalities which and content are felt to be of the same do not stop to become signs: it is suf- nature. If, at this very moment, Ve- ficient to observe the wood-cutter in suvius erupts, and our wood-cutter

49 is buried in many meters of volcanic continuous phase of a complex event ash, he will have been transformed, sequence, in which one phase fore- when he is rediscovered many cen- shadows another, not a sign, in the turies later, into a sign of the person sense of a signifier being differenti- he was, and of the particular phase of ated from a signified. More precise- his earlier action, as well as of many ly, in Husserlean terms, it is a proten- other things, and as such he will be tion occurring in the here and now of doubly differentiated, non-thematic the woodcutter, pointing forwards to and directly given, while the person the next immediately following mo- he was and the act he accomplished ment, and through that the moments is now thematic and indirectly given. to follow. To the archaeologist, on the His packed lunch, however, bread be- contrary, the carbonized body of the come carbonised, is less clearly dif- woodcutter is a true sign, not only a ferentiated. logical implication. It is to some ex- As Manetti (1993) has shown, tent outside of time and space.39 divination, together with medical Something like Husserl’s crite- symptoms, were the first semioti- ria are required, but perhaps not suf- cal phenomena studied; and they all ficient, in order to separate the sign have the form, as later formalized by function from other dyadic relations the Stoics, that if something is the between (more or less) differentiated case (p), then something else is also members. It is possible, no doubt, to the case (q).38 Indeed, this was that conceive of the sign as some kind of which to Antiquity, before Augus- mapping between “mental spaces”, tine, was known as a sign (semeîon), as suggested by Fauconnier (1994: which what we would call linguistic & Sweetser 1996), but this is not of signs were not (Cf. also Deely 2001). much use as long as we have no cri- Indeed, a linguistic signifier (or a pic- teria for separating the sign from all torial one) is not readily conceived as other instances of such mappings an effect permitting as to conclude to listed by Fauconnier, such as coun- the cause, identified with the signi- terfactuals, analogy, metaphors, me- fied. Our wood-cutter, surprised by tonymy, propositional attitudes, mo- the ash falling down (p), may well dalities, pragmatic terms, frames, conclude that Vesuvius in erupting models, and so on. This is of course (q); but at this very moment, this is a not to deny that some valuable gen- eralisations may be stated at this lev- 38 It could be said, as I have pointed out elsewhere (Sonesson 2000b) at least about divination, that but these signs beca- me interesting not as signs of Nature, but 39 See the next section about the time because they were conceived as messages characters of different kinds of signs, as from some kind of Super-Subject; but this well as Lecture 9 concerning the photo- is not the essential point at present. graphic sign. 50 el.40 sign is not even material, considered Another case in point is one of as a form (in Saussurean terms).41 the arguments employed by Fodor to Eco (1984:216f) has repeatedly posit the existence of a “language of denied that the mirror is a sign: in- thought”: that in order for us to be stead of standing for something it able to redescribe common sense psy- stands before something: the mir- chology in terms of brain function- ror image is not present in the ab- ing, there must be something mate- sence of its referent, is causally pro- rial, parallel to the expression of lan- duced by its object, and is not inde- guage that in the brain corresponds pendent of the medium or channel by to the neutral pathways, which is re- which it is conveyed.42 Indeed, in his lated to something mental, parallel most recent work, Eco (1998: 22ff; to the content of language. Indeed, 1999: 371ff) extends this description Fodor’s argument relies on expres- to some phenomena, notably television and content of the “language of sion, which most people would nat- though” being isomorphic, that is, urally consider to be pictorial signs. highly iconical, so that whatever is With reference to our more precise said to happen to the expression also concept of sign, I see really no rea- can be said to happen to the content, son to deny the sign character of the but I am not concerned with this spe- mirror: something which is com- cific claim here. Whatever the merit paratively more direct and less the- of this argument, the comparison of matic, the mirror image, stands for the relationship between brain anat- something which is less direct and omy and consciousness in terms of more thematic, the object in front of expression and content is fallacious. the mirror; and the person or thing in The neural pathways are not that front of the mirror is clearly differen- which is immediately given but not tiated from the image in the mirror. in focus, and consciousness is not in- The fact that the person repre- directly given but in focus. Between sented by the mirror sign is present neural pathways and thinking there contiguously to the sign is in no way is no doubt some kind of causal rela- an embarrassment to this concep- tionship, no matter how we choose to tion: in principle, this case is equiv- construe it; but there is no semantic 41 I have taken this description of relation. Indeed, the expression of a Fodor’s aims from Bermúdez (2005) who 40 This is the confusion that has gives other arguments, but of course not permitted numerous structuralists to claim this one, against Fodor’s theory. the presence of “double” or even “triple 42 Eco (1984, 1998, 1999) gives seve- articulation” in many kinds of semiotic ral other arguments for this claim, which resources. Cf. Lecture 3 and 4 below. I have shown to be invalid in Sonesson Something more will be said about propo- 2003b and forthcoming. Cf. also Lecture sitional attitude later on in this section. 4.

51 alent to the label with the names verse way. and the pictures of the different spe- The mirror clearly has a “body “ cies habitually appearing on the bird of its own. The framed picture even cage. Of course, animals and small more obviously has one. What is at children may have difficulty making stake, however, is much more than the required differentiation, but that the distinction, often made in cog- is exactly what happens in the case nitive science, between internal and of signs, as Piaget has indicated. The external representations. To see that, kind of differentiation which does we must take a step back to the world not obtain for animals and children before the emergence of the sign. is apparently not the one involving a discontinuity in time and/or space Signs and mediations : The (i.e., they do not think the mirror im- Fonseca-Peirce connection age is part of themselves) but rath- My concept of sign or representa- er that concerned with the different tion (like that of Piaget) does not in- nature of the two correlates (i.e. the volve ordinary perception being an cat takes its own image to be anoth- instance of it : our way of being in er cat). the world is not to be likened to the The mirror and the picture, just presence at some kind of private the- like verbal language, have in com- atre. Latter-day cognitive scientists mon being founded on a differenti- are therefore quite right in rejecting ation between two units which are the notion of representation of their asymmetrical in a double sense, first forebears. They are wrong, I submit, because on of the units is more im- to reject all kinds of representation mediately accessible to conscious- (to the extent that it corresponds to ness than the other, and second be- the semiotic function). More in par- cause the second units is more in fo- ticular, they commit a serious error cus than the first. This is not true of by not defining representation be- all kinds of conjunctions of “men- fore deciding that is has to be thrown tal spaces”, nor does it apply to Fo- out. dor’s “language of thought”. The Curiously, John Locke, who is kind of asymmetry involved here is on some accounts the father of sem- of course not at all opposed to the iotics (or at least of the term), sim- symmetry permitting the listener to ilarly seems to treat signs as being recover the same signified from the on a par with ideas, where an idea is signifier which prompted the speak- to be understood as any kind of tak- er to choose it in the first place, or ing account of the facts of the out- the possibility to look up the French side world.43 Thus, the experience of equivalent of an English word in a 43 Deely (2001: 590ff) argues that dictionary, as well as going the in- Locke’s last chapter, in which Locke 52 redness, or of a red book, is in some Moreover, it accords with some no- ways parallel to the word ”red” or tions of the scholastic philosophy the syntagm “red book”. This is not current in the Middle Ages. While I only strangely reminiscent of what do not think there is any direct link we find in “classical” cognitive sci- between cognitive science and scho- ence, but it also seems to correspond lasticism, this connection is quite ap- to at least some usages of the term parent in the case of Peirce (and per- “sign” found in the work of Peirce.44 haps Locke). A few notions of history may help us to disengage our- proposes to see all of philosophy, apart selves from the present-day concep- from physics and ethics, as a doctrine of sign, would require the reworking of tual muddle. the whole book, substituting “signs” for As was noted above, the church- “ideas”. It might be argued, however, that father Augustine seems to have been if you take the domain described by the words “signs” and “ideas” and put them responsible, certainly not for invent- together, it does not matter much whether ing, but for making explicit the com- you call all of it “signs” or “ideas” (just mon sense notion of sign on which as it does not matter much, to reverse a later thinkers, such as Saussure and classical Saussurean example, whether the French use “lamb” or “mutton” for both Husserl (and, at least in his defini- the domains covered be these terms in tions, Peirce) are tacitly building : it English – semantically, of course, because is, he tells us (in the convenient par- phonetically, this would be another mat- aphrase of Deely 1994 : 58) “some- ter). A reasonable retort would be that it is different to project the model of the sign thing which, on being perceived, to the domain of ideas, and the model of brings into awareness another besides the idea to the domain of signs. It is not itself“. Thomas Aquinas already had clear, however, what exactly is the dif- some misgivings about this defini- ference between these models. However, the followers of Locke in France, who tion, without ever daring to reject it certainly took their name for the concept outright. The followers of Aquinas of ideas, the “ideological school” (Pinca- in Paris may have been somewhat vet 1891; Gusdorf 1966), ended up talking bolder. In a written form which has very much about signs (Degérando 1800). come down to us, however, we first 44 However, although Deely (1982, know this criticism from the works 1994) seems to taken a different view of the matter, I believe most of Peirce’s defi- of Pedro da Fonseca, who was active nitions of the sign are more appropriately in Coimbra on the Imberian peninsu- construed as corresponding to (potentials la in the 16th century. To Fonseca and for) the semiotic function, as suggested his followers in Coimbra, the defini- above. On the other hand, many of his examples do not seem to confirm to it. See tion of the sign must be considera- examples in Peirce 1998. As any reader bly broader: a sign is anything which of Peirce must have noted, he quite often serves to bring into awareness some- quotes scholastic writings, and, as Deely thing different from itself, whether points out, particularly those of the followers of Pedro da Fonseca, on which more will be said below.

53 the sign (in the sense of the signifier) In fact, when closely considered, Fon- itself becomes subject to awareness seca’s observations really go against in the process or not. the grain of the by now familiar If the sign itself does not have to identification of our awareness of the be perceived in order for us to come world with the sign. It echoes Hus- to an awareness of that which is sig- serl’s as well as Gibson’s description nified, Fonseca described it as being of the perceptual act as something formal; but if the sign cannot lead to which points beyond itself without the awareness of anything at all un- itself being present to consciousness less it is itself perceived, he called it (cf. Sonesson 1989a,III.3.2). Indeed, instrumental (cf. Deely 1982: 52ff; Deely (2001:411ff) argues that to 1994: 58ff; 2001: 414ff). Put in more Fonseca, formal signs are not prop- convenient terms, a sign may either erly speaking signs: consist of a signifier (expression) Hence may be gathered the most striking which has to be perceived as such difference between instrumental and formal signs: since indeed formal signs do not in order to usher into the perception have to be preceived by us in order for us to of the corresponding signified (con- come to an awareness of the thing signified by the perception they structure; but unless tent); or it may consist in a signifi- instrumental signs are perceived, they lead er which is not ordinarily perceived no one to an awareness of anything (Fonseca quoted by Deely 2001: 413). as such but still somehow serves to mediate the perception of a signified. More exactly, in what in here Thus, Fonseca pointed to an analo- called a formal sign, the “sign” can- gy, but also to a distinction, of which not be perceived, if we are go gain a at least the latter seems to have been proper awareness of the thing signi- lost even on latter-day semioticians fied; for such an awareness in only and cognitive scientists. If so, this possible in what James Gibson calls would belie the origin of the distinc- the “pictorial”, and Edmund Husserl tion in the nominalist ambience (cf. calls the “phenomenological”, atti- Deely 2001: 390ff). What is here called an instru- scholastic philosophy. Deely (2001: 404) tells us it was introduced into scholastic mental sign clearly is that which philosophy in the 12th century as short- Husserl, following Brentano, has de- hand for indicating the essential relatio- scribed as the fundamental trait of nality of psychological phenomena. If consciousness, intentionality, that is intentionality has anything to do with “intensions” in the scholastic sense, however, the property of being directed to that it apparently pertains to “second inten- which is outside of consciousness.45 tions”, i.e. the things as known, while the first intentions would rather correspond 45 Brentano (1885), whose concept to what we nowadays consider to be the of intentionality was taken over by Hus- extension (but the actual distinctions are serl and more recently by Edelman (1992), really more complex, as Deely 2001: 470ff himself tells us he took the idea from points out). 54 tude, in which the content of con- mentary pieces we must construct sciousness, and not the thing cog- our perceptions (cf. Küng 1973), so nized, becomes the theme of the men- Gibson’s strawmen are the follow- tal act. This is exactly what does not ers of Helmholtz, the so-called “con- happens in the familiar Lifeworld, as structionists” (who have recently re- but Gibson and Husserl have pointed emerged within cognitive science, out. Indeed, the “pictorial attitude”, e.g. Hoffman 1998), who claim that similar to a picture, is “indirect per- hypotheses are needed to build up ception”, whereas ordinary percep- perceptions from the scattered piec- tion is “direct”. es offered us by sensation (cf. Son- When Gibson (1978:228) ob- esson 1989a;III.3.3).46 At least su- serves that, when we are confront- perficially, however, there is an im- ed with the-cat-from-one-side, the- portant difference; for whereas Hus- cat-from-above, the-cat-from-the- serl rejects the picture metaphor of front, etc., what we really see is all consciousness, by showing Bren- the time the same invariant cat, he tano and Helmholtz to be in error actually recovers the central theme in their very conception of pictures of Husserlian phenomenology, ac- and other signs because of ignoring cording to which the object is entire- the transparency of the expression to ly, and directly, given in each of its the content (cf. Küng 1973), Gibson perspectives or noemata (see Hus- (1978) instead emphasises the dis- serl 1939; 1962 a, b; 1973; and Sones- similarity of the picture from a real- son 1989a,I.2.2). In a similar fashion, world scene, thus showing the nu- Husserl’s favourite example is the merous experiments using pictorial cube (or perhaps the die) which can stimuli to study normal perception be observed from different sides. In to be seriously misguided. And yet, Gibsonean terms, these are “the sur- to both Husserl and Gibson, normal faces of the world that can be seen perception gives direct access to re- now from here” (Gibson 1978:233). ality, while Gibson thinks pictures Husserl’s cube and Gibson’s cat in- represent a kind of indirect percep- stantiate the same phenomenal fact – tion, and Husserl (1980) tells us (cf. for it is a phenomenal fact, and not Sonesson 1989a,III.3.6) that they are an experimental one, also in Gib- son’s work. 46 Reed (1996) notes some parallels Just as Husserl called into ques- between Gibson and the American prag- tion the conception of his contempo- matists (without, however, referring to rary Helmholtz, according to which Peirce!). “Constructionism” should be un- consciousness is like a box, with- derstood here as in perceptual psychology, in opposition to Gestalt psychology and in which the world is represented by ecological psychology, not in the sense of signs and images, from whose frag- Piaget or Vygotsky.

55 “perceptually imagined”.47 es, but the markings on surfaces may To perceive surfaces is a very possibly do so. The pattern of a sur- different thing from perceiving face and the pattern on a surface are marks on surfaces, Gibson (1980) different, and can usually be distin- maintains. Depth is not added to guished by an adult. The surface on shape, but is immediately experi- which a “graph” has been executed enced. In fact, the perception of can be seen underneath the “graph”. surfaces, of their layout, and of the However, a surface may be decorat- transformations to which the latter ed, regularised, textured, painted, or are subjected, is essential to the life embellished in other ways without of all animal species, but the mark- acquiring a referential meaning; and ings on these surfaces have only deposits of dirt or blots of pigment gained importance to man, notably may be left on the surface without in the form of pictures. The marks, the surface being made to stand for produced by what Gibson calls the something. The two cases, intuitive- graphic act, can be deposits, traces, ly describable in terms of the opposi- lines, or shadows projected on the tion between order and disorder, are surface. They may be produced by not distinguished by children. finger-tracing, drawing, painting, or To Gibson, then, the picture is a engraving, with a tool such as a sty- surface among other surfaces before lus, brush or pen; or otherwise a sim- becoming a sign. Gibson (1978:231) ple device, like the ruler or the com- observes that, besides conveying the pass, may be used, or a complex one, invariants for the layout of the pic- such as the printing press, the gadg- tured surfaces, the picture must also ets of photography, or the projector contain the invariants of the surface of lantern slides (Gibson 1980:xii; which is doing the picturing: those of 1978:229). Surfaces have the kind the sheet of paper, the canvas, etc., as of meaning which Gibson elsewhere well as those of the frame, the glass, calls “affordances”; the markings on and so on. Although Gibson does not surfaces, however, have “referential use the term, he clearly describes the meaning”. Without discussing the ex- picture as a sign, in the strict, Au- act import which should be given to gustinean sense of the word: as a the term “affordance”, we may safely surface which, on being perceived, conclude that “referential meaning” brings into awareness something be- is a property of what we have called sides itself. Gibson never specifies the semiotic function. That is, sur- what he means when he claims that faces do not stand for other surfac- surfaces are only seen to stand for something else by human beings, in 47 The precise import of this latter contradistinction to animals and chil- term is part of what will be discussed in the third lecture. dren. If he meant to suggest that sur- 56 faces can never be taken to be some- by von Uexküll.48 In the following, thing else than surfaces by animals we will use the term mediation for and children he was clearly wrong: this general sense of meaning which we know that even doves may re- Fonseca called sign and to which act the same way to a picture as to Peirce sometimes also may be hint- that which is depicted (cf. Sonesson ing.49 Mediation, in this sense, has a 1989a,III.3.1). The difficulty, clearly, least a double aspect, even if we ex- consists in seeing, at the same time, clude signs: it corresponds to impli- both the surface and the thing de- cational relationships such as those picted. In other words, in consist in called signs by the Stoics, and it also making a differentiation: in telling involves intentionality in the sense of the “body” of the sign apart from the Brentano and Husserl. In the former “body” of the object to which it al- respect, it seems to have something ludes. to do with Gibson’s “affordances”, We should grant Fonseca the in- and with Piaget’s notion of “connect- sight that there is some kind of anal- ing significations”. Once we have of- ogy between signs and intentional fering a wider context for inserting acts. However, to use the term sign these meanings of meaning in the in both cases dangerously suggests next section, however, it will be ea- that there is no important distinction seir to take a closer look at the notion to be made. The difference as well of intentionality, as related as related as the similarity can be spelled out: to what is known, in other traditions, intentionality (formerelly known as a the psychology of propositional at- formal sign) is the kind of realtion- titudes. ship in which the first item is not thematic and not in focus, and where Summary the second item is thematic and in Neither Saussure nor Peirce offers focus. any real definition of what the signs is. Nor is the notion of representation In his late life, Peirce realised that all his notions were too narrow: 48 This concept, as well as contem- instead of “sign”, he reflected, he re- porary biosemiotics, will be thoroughly ally ought to talk about “medium” or discussed in the next section. “mediation” (manuscript quotations 49 Whether it also has something to given in Parmentier 1985). Also Ernst do with the Vygotskyan concept of mediation is something that cannot be discus- Cassirer (1942; 1945) sometimes used sed here. May it just be noted in passing the term “mediation” (that is, “Ver- that the Vygotskyan concept of mediation mittlung”) in a more general sense seems to be reduced to language-depen- of meaning than “sign” (which he dence and, perhaps in a few instances, dependence on other semiotic resources called “symbol”), notably compris- which are signs in our sense, such as pic- ing the Umwelt ascribed to animals tures.

57 in cognitive psychology defined. The ences, sometimes perhaps suggested discussion whether the sign has two, independently by different scholars. three or more parts has no meaning Husserl posit the Lifeworld so as to before we have determined the do- explain the foundation on which the main that we are analysing and what models of the natural sciences are con- criteria we will apply to its segmen- structed, both serving as the primary tation. Those who have rejected the objects studied and transformed by notion of sign or representation, such the model, and as the common sense as Greimas in semiotics, and con- world in which the scientists are ac- temporary cognitive scientists such complishing their work: indeed, you as Lakoff and Johnson, have never cannot treat the accelerator permit- defined that which they reject. In- ting you to study the electrons as be- stead of rejection the notion of sign, ing at the same time a bundle of elec- we have to clarify it, so at to separate trons itself. Students of Husserl such it form other notions of meaning, as Aron Gurwitsch, Alfred Schütz, which we will call mediations. The Maurice Merleau-Ponty and Herbert sign, in our sense, supposes the con- Marcuse considerably extended, not comitant awareness of at least two the meaning, but the function of the items, which are subjectively differ- concept of Lifeworld, using it to ex- entiated from each other, while one plain social reality itself. We owe to of them is directly given but not the- Schütz, in particular, the description matic and the other indirectly given of the Lifeworld as “the world taken and thematic. The signs is thus dif- for granted”. The “commens” char- ferent from other mediations, such acterised by Peirce (1998: 478) would as intentionality, in which one item seem to be a similar domain of shared is both directly given and thematic, assumptions. When the psychologist while the other is neither, and impli- James Gibson postulated the world cational relationships, which are not of “ecological physics”, so as to ex- differentiated. In this sense, pictures plain the possibility of immediate are signs, but they refer to intentional perception, where the older school of relations, and they contain implica- constructionists had to suppose com- tions. plex calculations, his does not refer to Husserl explicitly anywhere in his 2.3. Meaning before the writings, but he often uses the same sign: the commens of phrases and examples. Greimas cer- perception tainly took the idea of a semiotics of the natural world from Husserl The idea of a common sense world via Merleau-Ponty. Common sense has reappeared numerous times in has always been the basis of Anglo- philosophy as well as in the social sci- Saxon philosophy, from the British

58 Empiricists to the Oxford school. At mally be considered the subject mat- long last, however, even this tradi- ter of physics and chemistry. How- tion has come to appreciate the gap, ever, if it is reduced to the meaning diagnosed by Husserl, between the it has for us, then, depending on the contemporary natural sciences and particular culture and context in- the world of our experience, postu- volved it may stand for the ancestral lating both a “naive physics”, and a gesture thought to mark the begin- “common sense psychology”, which nings of civilisation, for the operat- together would seem to make up the ing force of steel furnaces, for one of Lifeworld. In a more general sense, the four elements, the universal con- what Searle (1995:127ff) calls the verter of the alchemists, the confla- “background” would also seem to gration of the neighbour’s house, the correspond to the Lifeworld, as does, infernal flames, the cosy fire place in if Searle is right about his parallel, a the country house, the log-fire of the lot of things written by Wittgenstein barbecue party, the cowboy’s watch- and Bourdieu. Coming from a very fire, and so on (cf. Sonesson 1989a: different tradition, Jakob von Uexküll 26-29). When fire appears in a par- introduced the notion of Umwelt to ticular culture, in a ritual, a film, or a serve as some kind of world taken picture, its presence its probably mo- for granted of the animals –although, tivated rather by one of the aforemen- of course, in a deeper sense, the tick tioned meanings or similar ones than and his kin do not have choice of tak- by the chemical formula. In some of ing anything for granted at all. these cases, fire is a sign, in the others it is a functional object. The ecology taken for granted: Historically, meanings of this the Lifeworld kind have constituted “epistemolog- The celebrated semiotician A. J. Gre- ical obstacles”, as Bachelard (1949) imas (1970:49) suggested that there put it, for the quantitative reduc- could be a cultural science of nature, tion, which is a prerequisite of all re- a semiotics of the natural world – search in the natural sciences. The which was concerned, then, with the result of Bachelard’s “psychoanalyse world which is natural to us, just as du feu”, which is really a social psy- a particular language is our “natural chology of early attempts at explain- language” (Swedish, English, Span- ing fire, strangely echoes Arnheim’s ish, German, etc.). This amounts to (1966:63) observation, that it takes a an attempt to consider the traditional very peculiar attitude to see in fire a domain of the natural sciences from collection of shapes and colours rath- a human point of view. One of the er than “the exciting violence of the cases Greimas mentions but does not flames”, though of course the chem- dwell on is fire, which would nor- ists have to go beyond the shapes and

59 colours too. There seems to be room both Greimas and Gibson took there for a study of the meaning of fire, cue from Husserl.50 quite apart from what natural science Greimas, Gibson, and Husserl tells us about it. In this sense, fire is all felt the need for such a science a category, like the phoneme, which because they realised that the “natu- introduces discontinuities in the per- ral world”, as we experience it, is not ceived world, and which subsumes identical to the one known to phys- many, somewhat differing instances. ics but is relative to human beings. Quite independently of the presumed Husserl’s Lifeworld as well as Gib- identity of the chemical formula, the son’s ecological physics, but not Gre- fire of Hell and of the cosy fireplace imas’ natural world, takes this lev- may or may not have semantic fea- el to be a privileged version of the tures in common. world, “the world taken for granted”, However central fire may be to in Schütz’s phrase, from the stand- hominisation, the human body no point of which other worlds, such as doubt precedes it. This is the body those of the natural sciences, may be as seen from the horizon of human invented and observed (cf. Sonesson consciousness, and it is a formidable 1989a: 26-29, 30-34, and passim).51 epistemological obstacle to the nat- Moreover, while Greimas’ semiotics ural sciences in general, and biolo- of the natural world largely seems to gy in particular. But it is also found be a kind of lexicon of the meaning at the start of the construction of the of things, Husserl and Gibson tried human world. to formulate a set of general princi- But Greimas was not the first ples, which underlay all our doings to conceive of a cultural science of in the everyday world. nature. His semiotics of the natural 50 Another “science of normality” world, together with Husserl’s sci- is the time geography of Torsten Häger- ence of the Lifeworld, and “ecolog- strand (1983), which is concerned with general invariants of space and time, which ical physics” as invented by the per- tend to be trivial, rather than exceptional ceptual psychologist James Gibson in kind, and which impose restrictions on are all sciences of normality, of that the actions of individuals. So is of course which is so much taken for granted “naive physics” as conceived in cognitive science. that it is ordinarily not considered worthy of study (cf. Sonesson 1989a, 51 Indeed, since he tells us language and the natural world are the two main 1994b,c; 1996a; 1997a). It may seem divisions of semiotic systems, Greimas strange to put together ideas and ob- probably thought of them as equally being servations made by a philosopher, a representations, not in the wide sense of psychologist, and a semioticians; yet Fonseca or Peirce, but in that of French structuralism, constructivism in percep- these proposals are largely the same; tual psychology and classical cognitive indeed, there are indications that science. 60 It is a basic property of the Life- tered in the Lifeworld is referred to world that everything in it is given a general type. According to Schütz, in a subjective-relative manner. This other people, apart from family means, for example, that a thing of members and close friends, are al- any kind will always be perceived most exclusively defined by the type from a certain point of view, in a to which they are ascribed, and we perspective that lets a part of the ob- expect them to behave accordingly.53 ject form the centre of attention. As Closely related to the typifications we noted above, Gibson observes are the regularities, which obtain in that when we are confronted with the Lifeworld, or, as Husserl’s says, the-cat-from-one-side, the-cat-from- “the typical ways in which things above, the-cat-from-the-front, etc., tend to behave”. This is the kind of what we see is all the time the same principles tentatively set up which invariant cat. To Husserl, this seeing are at the foundation of Peircean ab- of the whole in one of its parts is re- ductions. Many of the “laws of eco- lated to the etc. principle, our knowl- logical physics”, formulated by Gib- edge of being able, at any one point, son (1982:217ff), and which are de- to turn the dice over, or go round the fied by magic, are also such “regu- house, to look at the other sides. This larities /that/ are implicitly known”: principle applies to the temporal and that substantial objects tend to per- the spatial organisation of the world sist, that major surfaces are near- alike. In time, it accounts for our ex- ly permanent with respect to lay- pectancy, at every moment, that life out, but that animate objects change will go on, or that something will as they grow or move; that some ob- change, or something more definite, jects, like the bud and the pupa trans- such as that the dice will turn out form, but that no object is converted to have a certain number of eyes on into an object that we would call en- the hidden sides (the protensions), as tirely different, such as a frog into a well as our knowledge that we exist- prince; that no substantial object can ed in the moment immediately pre- come into existence except from an- ceding the present one, that the dice other substance; that a substantial de- did so to, and perhaps also our mem- tached object must come to rest on a ory of the sides of the dice we have horizontal surface of support; that a seen before, and the context in which solid object cannot penetrate another they dice appeared (the retentions).52 solid surface without breaking it, etc. Every particular thing encoun- Clearly, many of these regularities

52 This model of time consciousness 53 A similar point is made by Gur- was used in theatre semiotics, and in lite- witsch (1979: 104ff) in terms of roles. rary semiotics, by members of the Prague More will be said about typification in school, notably by Mukarovsky. Lecture 4.

61 do no longer obtain in present-day and plant cells. The observer himself physics, but they are necessary for underlies the consequences of the ri- the human environment to hold to- gidity of the environment and of his gether. Some of the presuppositions own relationship to gravity. of these “laws”, such as the distinc- The Husserlean description of tion between “objects that we would regularities also fits in with the no- call entirely different”, are also at tion of abduction, which Peirce puts the basis of what we have called the alongside the more familiar proce- Lifeworld hierarchy, and the defini- dures of deduction and induction, tion of the sign function (cf. Sones- and which reasons from one particu- son 1992a; 2000a; 2001a).54 lar instance to another, not, however, More than Husserl, Gibson at- exclusively on the level of individual tends to the general background of facts, for the facts, Peirce tells us, are the world taken for granted. The mediated by certain “regularities”, “terrestrial environment” of all an- principles that are tentatively set up imals has continued to possess cer- or taken for granted. Some of “typi- tain simple invariants during the mil- cal ways in which things tend to be- lions of years of evolutionary history, have”, of which most may be of more such as the earth being “below”, the regional import than those formulat- air “above”, and the “waters under ed by Gibson, would seem to be at the the earth” (Gibson 1966: 8ff). The origin of “signs”, in the Stoic sense ground is level and rigid, a surface of of the term, that is, inferences or im- support, whereas the air is unresist- plications. In discussing the Meso- ing, a space for locomotion, and also potamian art of divination, Manetti a medium for breathing, an occasion- (1993: 6ff) distinguishes three kinds al bearer of odours and sounds, and of relationships between the protasis transparent to the visual shapes of (p, that is, the if-clause) and the apo- things by day. As a whole, the solid dosis (q, that is, the then-clause): di- terrestrial environment is wrinkled, vinatory empiricism, when p and q being structured, at different levels, have occurred together in the past; by mounts and hills, trees and other chains of associations, when there is vegetation, stones and sticks, as well a similarity between the signifiers, as textured by such things as crystals or a rhetorical figure linking the signifieds; and coded relationship be- 54 When latter-day constructionists tween a finite number of identifiable such as Hoffman (1998) start formulating 55 general laws, they do not seem to be so cases. far from Gibson as the imagine. On the whole, however, Hoffman’s laws seem to 55 One may recognize, in the first apply to pictures, rather than the percep- two cases, Frazer’s (1922: 11) two prin- tual world, more like those of Kennedy ciples of magic, according to the laws of (1974a). contact and similarity. Even more ob- 62 The first type if of course closest pearance which it also had last time to purely perceptual reasoning, and a rebellion occurred. Both connec- could be formulated in terms of pro- tions, however, at first may be based tentions (what can be expected next) on the experience of how things tend and retentions (what can be taken to to behave in the Lifeworld. Only at have happened before). It could also later stages will they be separated.56 be said to depend on an indexical re- More complex abductions may lationship. That which is described be necessary, not only in the case of in the protatis-clause may have ap- “coded” relationships, but also those peared in the neighbourhood of that based on similarity, since some prin- which is in the apodosis-clause, in ciple for picking out the relevant space and/or in time. All experience properties will always be needed. taking place in time is of this kind, Still, as long as all this takes place as for instance our expectancy, when a matter of course, we are at the level seeing the wood-cutter with the axe of inferences (or Stoic signs), not that raised over his head, that in the fol- of real signs. lowing moment, he is going to strike the piece of wood (contiguity proten- The affordances of a game of tion), as well as our knowledge that, chess in the moment just preceding, he lift- But let us get back to “the things ed the axe to its present position (con- themselves”, and in particular to Hus- tiguity retention). Perhaps the regu- serl’s favourite example: the cube, or larity which is here taken for granted the dice — “Würfel” may mean the would be an abduction, as Peirce un- one or the other. But we will begin derstands the term, if only in a very with the cube. Like any other ob- trivial sense: it does not take much perspicacity to posit the general rule 56 In formulating his laws of ecolo- which connects the two individu- gical physics, Gibson (1982:218) claimed that, contrary to what is often thought, al cases. There is certainly a differ- children do not spontaneously believe in ence between seeing the woodcutter magic. At least some kinds of divination lift his axe over his head, and wait would clearly be contrary to these prin- for him to split the log, because one ciples of ecological physics. According to Piaget, of course, children do go through event has followed the other in earlier a magic stage, and anthropologists appa- circumstances, and to predict that a rently have found many adults believing in rebellion will take place, because the magic, too, even though the cases quoted liver of a certain animal which has in the first section of the lecture concerning magical interpretations of pictures do been inspected has a particular ap- not seem to be authentic. Still, the Life- world of everyday praxis, in which instru- viously, the three cases are reminiscent of mental and other goal-related actions take indexicality, iconicity, and symbolicity (in place, may have to be distinguished from that order). the ideological Lifeworld.

63 ject, the cube is necessarily given in be specified or rejected when the ear- perception from a particular point lier unseen sides come into view (Cf. of view. Husserl calls what is seen Husserl 1939; 1962a, b).57 the object (“Gegenstand”), and the Gurwitsch (1957; 1974a), who aspect through which it is seen is compared this Husserlean concep- termed “noema”. In our normal life tion to the “spontaneous phenome- in the Lifeworld, we do not attend nologies” of the Gestalt school, has to the particular acts and the corre- pointed to the “Gestalt-coherence” sponding aspects through which the with which the mutually confirm- object is given. While the particular ing noemata form the object of per- noema by means of which I presently ception. Criticizing Husserl because see the cube only contains three of he seems to consider the object itself its sides in different perspectival de- as a separate instance, an “X” which formations, I immediately see it as is the bearer of the noemata, Gur- a cube, complete with its six sides, witsch (1974a; 254) tells us that the not as some strange object I hypoth- perceived thing is “nothing else than esize to be a cube. Through an act the internoematic system itself, i.e. that Husserl calls reflexion, the phe- the system of multiple adumbration- nomenologist, the psychologist, and al presentations and of the proper- the aesthetically-minded contempla- ties and qualities exhibited in those tor may choose to attend to the acts presentations”. Similarly, the predi- of consciousness and their corre- cation (“X is red”, and so on) which sponding noemata instead, thereby Husserl conceived to be a “synthe- transforming them into new objects sis”, an adjunction of new properties, with their own noemata. In normal is really an “analysis”, an explicita- consciousness however, the act will tion of what is already contained in only give a particular modification the horizons of the perceptual thing. to the perception of the object, a While phenomenology does tinge of meaning: some parts of the not have any historical connection to object appear more specified, others contemporary psychologies of per- only roughly outlined. What is just ception, as it has to Gestalt psycholo- sketched out in one noema may be gy, Gibson (1971; 1978) tells us, just filled in in a number of others, and like Husserl, that the object is di- the knowledge that we can always go rectly seen, complete with its hid- further in the exploration of the object is part and parcel of our percep- 57 Thus, like Ames’s famous chair tion of the object, as expressed in the seen from a peephole, one possible noema etc. principle. Whereas retentions of of the cube may be simulated, without already seen sides are the basis for there being an object which gives rise to further exploration, protentions may further, coherent noemata of the same object. 64 den sides, without any inferences be- ly sketched in may be fully known. ing necessary: even the child will see Thus we meet indexicality in another “the invariant cat”. What assures the sense, as the continuity of one view identity of the object through all the to another – and certainly not as a differing views we may take on it, is, “sign”, though Gurwitsch, like Ja- according to Gibson, “the formless kobson, uses the word “renvoi”.59 and timeless invariants”, reminis- There is a problem with this de- cent of the “common core” in Gur- scription of the Lifeworld that should witsch’s “noematic matrix”, which be as critical to Gibson as to Hus- defines perceptual coherence.58 Still serl: suppose that what I am looking closer to the noematic matrix sug- at is not just a cube but more partic- gested by Gurwitsch is Gibson’s dis- ularly a dice. Then the argument ad- ciple Hagen (1979; 1980), who main- duced by Husserl and Gibson contains that the existence of pictori- tinues to be valid: I will see the ob- al perspective requires the mind to ject as directly to be a dice as a cube. take account of “the entire family of But this information is certainly not possible perspective views of an ob- there simply to be picked up: Hus- ject” (1980a: 29), quite apart from serl’s (1962b; 1973) Bantu negro who the Gibsonian invariants. Accord- is supposed to operate the reduction ing to Gurwitsch’s (1957:152) pro- to the common Lifeworld would be found analysis of the notion of per- at a loss to see the dice, at least if he ceptual noema, each point of view is otherwise as naive as Husserl sup- is really “l’appréhension d’un sys- poses. And yet, to a grown-up mem- tème d’apparences dans la perspec- ber of Western culture, the dice is at tive et du point de vue d’un de ses least as directly seen as the cube. membres”. This means that each While both Gibson and Hus- noema contains the whole object, but serl exclude the cultural layer of in- in such a way that some parts will terpretation from the Lifeworld, Gib- be at the centre of attention, given in son at least take care to single out all their details, while other parts are what he calls “affordances” as a kind perceived marginally and vaguely, of meaning distinct from referential only in their general outlines. There meaning, and thus from the kind of are references (“renvois”; Gurwitsch meaning conveyed by signs. There is 1957:191) from each noema to all the no proper definition of the notion of others, in which what is here mere-

58 To Gibson, however, these in- 59 In this sense, the picture can never variants are mathematical, though not be a noema: whereas one noema will expressible in present-day mathematical imperceptibly fade into another, the pic- language. Pending the invention of this torial surface has clearly fixed limits. The mathematical system, however, it is dif- frame, however, may interrupt lines that ficult to make sense of this claim. are easily continued in imagination.

65 affordance in Gibson’s work, but he culture. Similarly, for most people gives some suggestive examples: it is in contemporary Western culture, a the graspability, or the edibility, of computer keyboard has an immedi- a thing. Graspability can be under- ate property of writability (not nec- stood as the aptness to be grasped. essary less immediately present than Edibility must be interpreted as the the depressibility of the keys). susceptibility of being eaten. These Of course, the meaning of the are inferences which might be said dice is not exhausted by its throwa- using a phenomenological term, to bility: it means different things, ac- be “sedimented” onto a object of the cording as different faces with a dif- Lifeworld: accordingly, an apple, ferent number of eyes turn up, and once it is seen to be an apple, is also in account of which kind of game it perceived as something which may is thrown. This is perhaps even tru- be grasped and then eaten, because er of the different items used to play these are events being known to have chess. Saussure, it will be remem- taken place (and “properly” so) with bered, used chess as a ready analo- other apples at other times. There- gy to language, arguing that any odd fore, the apple is apt to be grasped set of buttons may be used to play and eaten, both in the sense of nor- chess, as long as the rules specify- mality and normativity.60 While it is ing the possible movements of each possible for graspability to be a prop- buttons were known, just as, in prin- erty of things in some respect inde- ciple, any sound may stand for any pendent of culture, this could hardly meaning in a language. Anything is be the case with edibility. Anthropo- a king, as long as it is permitted to logical studies are full of examples move in the ways a king moves, just of things being eaten in some plac- as anything (with some exaggeration, es and considered entirely inedible in no doubt) may be an /a/, as long as it other places. And it is easy to think functions as an /a/ in the vowel sys- of other meanings that are clearly of tem. This may be true, but to some- the same kind as those mentioned one knowing how to play chess, only and which are yet culturally specific. a chessman looking like the king im- We just have to think about the dice. mediately affords the kinds of move- Suppose there is some human culture ment that are allowed to the king in where die have not been invented: it the game of chess. might yet seem as if the throwability Deacon (1997: 41, 59ff) goes of the dice may be perceived direct- even further, comparing “rule-gov- ly by those coming from the proper erned games”, of which chess must be an instance, together with eti- 60 This is the double sense of the notion of norm, to which we will attend quette rules and music, to language, further in Lecture 7. while excluding “portraits”, claim- 66 ing that the former, but not the lat- The case of chess, however, is ter, have “symbolic reference”.61 In more difficult to deal with. What fact, if we suppose “symbolic refer- makes some pieces of wood or oth- ence” to convey the general idea of er material and a board into a game something being “about” something of chess are the restrictions imposed else, or, equivalently, to stand for on the permitted movements of the something else, then it makes much chessmen and the consequences of more sense attributing it to at least certain chessmen taking up partic- some instances of animal communi- ular positions. In fact, as Searle has cation, and certainly to pictures as observed, the rules of chess are not used by human beings, than to such like traffic regulations, applying to things as etiquette, games, and mu- movements on a board which were sic. Etiquette rules and the rules de- hitherto unregulated: the restrictions fining games are not “about” any- on movement create chess, but traf- thing at all: they impose restrictions fic regulations do not create traffic. on the behaviour allowed. As Dea- In other terms, the rules of chess are con (1997: 61) claims about laugh- constitutive, but the rules of traffic or ter, it is certainly odd to say that eti- only regulatory.62 Clearly, it could be quette has a meaning, at least in the argued that the queen means “able to sense of reference. To shake hands move in any straight direction as far (in a given context) means that you desired”, in a sense in which /a/ does greet somebody; to move a particu- not mean “low, frontal, sonorous”. lar chessman means that the queen More to the point, perhaps, chess takes up a new position causing per- is really comparable to language at haps a checkmate. As I understand the level of syntax and hyletics (as the term “etiquette rules” (but Dea- we will say in Lecture 4), that is, as con gives us no clue) is does not in- something which may occupy certain volve something like shaking hands. positions and not others, as well as I would describe this as an interac- something which has some invariant tive gesture carrying a meaning just traits, and others which may be ex- as any other sign. Etiquette rules, changed freely. The chessman does however, are those that tell us un- not carry a meaning differentiated der which circumstances it is appro- from its expression, as is the case priate to shake hands, and when it with language and pictures. Again, is not. In this sense, they impose re- the chessman affords certain move- strictions on the behaviour allowed. ments – but only in a given culture Indeed, they determine the cultural 62 In the Pufendorf lectures, given at affordances of handshakes. Lund University, May 30 to June 2, 2006. The distinction between constitutive and 61 For the details of my critique, cf. regulative rules is made already in Searle Sonesson 2003a. (1969). Also cf. Searle 1995.

67 for which chess is a cultural fact. Searle (1995: 43ff) describes Saussure’s comparison involve the constitutional rules giving rise the chessmen and the elements of to games (and to institutional re- languages, such as phonemes and ality generally) using the formu- words. It does not pertain to sentenc- la “X counts as Y in C”. His exam- es, let alone utterances. But if the af- ples are such things as paper mon- fordance carried by a chessman con- ey and chess. To my mind, we may tains not only the sequences of acts very well say that a chessman (or a having been accomplished with it be- button having been substituted for it forehand, and sedimented onto it, but on the board) counts as an item apt also the disposition to carry out those to move in certain specified ways on same acts in the future, then perhaps the board. To say that an expression each single act, once realised, could (of a word, a gesture, a picture, and be comparable in some sense to an so on) counts as its content, however, utterance, or, more, exactly, the act is fairly misleading. Signs may really of uttering, the enunciation. Indeed, be surrogates for things, in a way, but Clark (1996: 40ff) suggests that each they fulfil different functions than move in chess could be seen as an the things themselves. They permit act of communication, modifying us to take a stand on things, so as the state of the common knowledge to chess, for the purpose of the life- of the two players. If so, each move- world, the meaning of these things. ment of the queen would be a kind of No chessman, nor even a move by a “chess act”, comparable to a speech chessman, really counts as a state- act, in case of which chess would be ment modifying the meaning of the a highly repetitive type of discourse. game of chess, let alone that which is Considered as a sign system, chess outside of the world of chess. would therefore possess a very limited domain of validity, or, as we will Von Uexküll on how it feels to say later on, very restricted content be a tick resourses.63 It has been suggested (notably by Smith & Varzi 1999) that the Life- 63 Cf. in Lecture 4, the discussion of world, understood as above, is simply Lessing’s analysis of pictures as opposed to language. Clark’s (1996:48f) observa- the niche, in the sense of (non-Gibso- tion that, in addition to the commonly ac- nean) ecology, in which the animal cepted description of the series of moves known as the human being stakes out made so far, there is also an “annotated his life (cf. Sonesson 2001a: 99). The record” in which one move may be characterized from the point of view of one niche, then, in this sense, is the en- player as “a blunder” or “a bold move”, vironment as defined by and for the refers to what we, in Lecture 4 and 5, will specific animal inhabiting it. In Hus- term different intensional levels of de- sserlean language, the niche is sub- scription. 68 Fig. 14. The model of the Umwelt according to Jakob von Uexküll

jective-relative – relative to the par- sends a message to its legs to let go ticular species. The precursor of the (Wirkzeichen), so that it drops onto niche, understood in this way, is the the mammal’s body. This starts a notion of Umwelt introduced by von new cycle, because the tactile cue of Uexküll, which is today the defining hitting the mammal’s hair incites the concept of the speciality known as tick to move around in order to find biosemiotics.64 its host’s skin. Finally, a third cir- Uexküll’s notion of mean- cle is initiated when the heat of the ing centres on the environment, the mammal’s skin triggers the boring Umwelt, which is differently de- response allowing the tick to drink fined for each organism (cf. Fig. 14.). the blood of its host. Together, these As opposed to an objectively de- different circles consisting of per- scribed ambient world, the Umwelt ceptual and operational cue bearers is characterised for a given subject, make up the interdependent wholes in terms of the features which it per- of the subject, corresponding to the ceives (Merkwelt) and the features organism, and the Umwelt, which is which it impresses on it (Wirkwelt), the world as it is defined for the sub- which together form a functional cir- ject in question. cle (Funktionskreis). According to a Scholars involved with biosem- by now classical example, the tick iotics tend to take this model, im- hangs motionless on a bush branch mensely enlightening as it is in itself, until it perceives the smell of butyr- and simply project onto it the sign ic acid emitted by the skin glands conception suggested by Peirce. The of a mammal (Merkzeichen), which first difficulty with this approach of course resides in finding out the real 64 I have presented the Lebenswelt import of the Peircean sign concep- as a particular kind of Umwelt in earlier papers of mine, before realising that Deely tion. Since this is in itself an infinite (2001) had also made this comparison, task, any scrutiny of the parallel risk without however entering a discussion of getting bogged down very early on. the import of the Husserlean notion.

69 If we confront the sign conception content would in turn correspond to defined here with the world of the the pieces of behaviour which tend to tick, however, it will be easy to see make up the reaction of the animal that there is no room for it there. Not to the features in question. There is only is there no distinction between no point getting lost here in Peircean expression and content to the tick; exegesis: if anything, we are faced there is no separation of sign and re- with a “formal sign”, as conceived in ality. At least in part, this is also an the Fonseca tradition. As we are us- opposition between the Umwelt and ing the terms, we would have some the Peircean sign. kind of mediation (Cassirer’s ‘Ver- Before the invention of biosem- mittlung’), but not a sign.65 Howev- iotics, Ernst Cassirer (1942: 29ff; er, there are, as I will explain in the 1945: 23ff) was no doubt the first following, two differences between thinker outside of biology to take what is happening in the Funktion- von Uexküll’s ideas seriously. After skreis and what we have here defined pointing out that, to human beings, as a sign. all experience is mediated (a case As Ziemke & Sharkey (2001:709) of ‘Vermittlung’), he goes on to ob- point out, it is hard to find the object serve that this is also true of animals, of the sign, in the ordinary sense of as described by von Uexküll. But he its referent in the “outside world”. makes no mention of the fact that, to What is for us, as observers, three von Uexküll, the Funktionskreis is a cues to the presence of a mammal, “theory of meaning” (‘Bedeutungsle- the smell of butyric acid, the feel of hre’). In fact, he opposes “animal re- skin, and the warmth of the blood, do actions” to “human responses”. Cas- not have to be conceived, in the case sirer may be wrong in not seeing the of the tick, as one single entity hav- similarity between signs and oth- ing an existence of its own (a “sub- er meanings (though he suggests it stance”, in Gibson’s terms), but may in passing using the term ‘Vermit- more probably constitute three sepa- tlung’), but he is quite right, I sub- rate episodes producing each its own mit, in insisting on the difference. sequence of behaviour. In fact, Ziem- Very tentatively, let us suppose ke & Sharkey go on to quote an early that, in the biosemiotic conception, text by von Uexküll, in which he says the features of the world observed by that “in the nervous system the stim- the animal correspond to the sign- ulus itself does not really appear but vehicle or expression (Peirce’s “rep- 65 It will be observed that we are here resentamen”); the object or referent simply equating the triadic, or Peircean, would then be that which causes the- conception of the sign with the so-called dyadic, or Saussurean, one, in accordance ses features to be present to the ani- with the interpretation suggested in the mal; and the Peircean interpretant or second section of this lecture. 70 its place is taken by an entirely dif- is not the “dynamical object”. At the ferent process” (my italics). Uexküll very most, it is the “immediate ob- calls this a “sign”, but it should be ject”. It will be remembered that, in clear that is does not in any way ful- Peirce’s conception, while the “im- fil the requirements of the semiot- mediate object” is that which direct- ic function. Indeed, expression and ly induces the sign process, the “dy- content are not differentiated, al- namical object” is something much ready because they do not appear to more comprehensive, which includes the same consciousness. The butyr- all those things which may be known ic acid is there to the tick; the mam- about the same object, although they mal is present only to us. In addition, are not present in the act of induc- it does not make sense to say that ei- ing. Indeed, the “dynamical object” ther the butyric acid or the mammal is that which corresponds to the po- is in focus or not. Nor is there any tentially infinite series of different sense in determining whether the bu- interpretants resulting from the same tyric acid or the mammal is directly original immediate object. It should given. be clear that, for the tick and simi- What is lacking is real Third- lar beings, there could be no distinc- ness: the reaction to the primary re- tion between direct and dynamical action, that is, the reaction that does object, because there is no room for not respond to a simple fact (First- any further development of the chain ness), but to something which is al- of interpretants. In this sense, Dea- ready a reaction, and thus a relation con’s (1997: 63), idiosyncratic read- (Secondness). Without having to en- ing of Peirce, according to which ter into the earlier discussion of dif- only signs such as those found in hu- ferentiation, we see that, even from a man language (his “symbols”) give strictly Peircean point of view, there rise to chains of interpretants seem is no Thirdness for the tick: it does to have some justification – in reali- not respond to any relationship, since ty, if not in Peircean theory (cf. Son- it is not aware (even in the most lib- esson 2003a). This is true, however, eral sense of the term) of any second only if one does not separate indexi- term (the mammal) to which the first cality and indexical signs, or iconic- term (the butyric acid) stands in a re- ity and iconic signs. lation. From Umwelt to Lebenswelt : In fact, things are even more the thematic field complicated. In a true sign relation, the mammal is not really the As I have often pointed out, to ac- object, in the Peircean sense, for count for the distinction between the which the butyric acid is the repre- “immediate object” and the dynami- sentamen. Or, to be more precise, it cal object, we need the concept of

71 ground.66 In one of his well-known of two black things (1.293). It there- definitions of the sign, a term which fore seems that the term ground must he here, as so often, uses to mean the stand for those properties of the two sign-vehicle, Peirce (2:228) describes things entering into the sign func- it as something which “stands for tion by means of which they get con- that object not in all respects, but in nected, i.e. both some properties of reference to a sort of idea, which I the thing serving as expression and have sometimes called the ground some properties of the thing serving of the representation” (my italics). as content. In case of the weather- Some commentators have claimed cock, for instance, which serves to that Peirce is here talking about some indicate the direction of the wind, properties of the expression, whereas the content ground merely consists others favour the content.67 In fact, in this direction, to the exclusion of however, the ground must concern all other properties of the wind, and the relation between them. Such an its expression ground is only those interpretation seems to be born out properties which makes it turn in by Peirce’s claim that the concept of the direction of the wind, not, for in- “ground” is indispensable, “because stance, the fact of its being made of we cannot comprehend an agreement iron and resembling a cock (the lat- of two things, except as an agree- ter is a property by means of which ment in some respect” (I.551). In an- it enters an iconic ground, different other passage, Peirce himself iden- from the indexical ground making it tifies “ground” with “abstraction” signify the wind). If so, the ground is exemplifying it with the blackness really a principle of relevance, or, as a Saussurean would say, the “form” 66 This was independently noted by connecting expression and con- Søren Brier 2001. tent: that which must necessarily be 67 Indeed, Peirce talks about “the present in the expression for it to be ground of the representamen”, and even claims that the representamen is con- related to a particular content rather nected to three things, “the ground, the than another, and vice-versa (cf. Son- object, and the interpretant”. This cor- esson 1989a:III.1).68 responds to the interpretation given by Savan (1976), but is opposed to that of 68 Some passages in the work of Greenlee (1975). Indeed, a quotation Peirce (quoted and discussed in Eco 1999) from Peirce (I.551-3) given by Deely suggest that the ground should be an (2001:642f), but not commented upon in instance of Firstness. Deely (2001:343, this sense, seems to suggest that Peirce 641) clearly condones this interpretation. would reserve the term “ground” for the It would be true of the respective lists of portion of the expression singled out and properties of the thing serving as expres- use the term “correlate” for the correspon- sion and the thing serving as content, but ding part of the content. This would ho- not about the principle establishing the re- wever seem to do away with the relational lation between them. However, as I noted character of the notion involved. in the preceding footnote, Peirce would 72 It has been suggested by Deely formal objects considerably. If for- (2001: 343ff; 641ff) that the notion of mal objects are indeed “ten formally ground is equivalent to what is known distinct cognitive channels” and may in scholastic philosophy as the “for- be defined as “whatever is directly mal object”. It may at first seem that and essentially attained by a power this would support my interpretation. and by reason of what whatever else In fact, however, the formal object is attained is attained” (Deely 2001; turns out to be that which describes 344), then it may perhaps have some- the domain to which particular sense thing to do with what I have suggest- organs are receptive: the eyes to dif- ed here, but it remains considerably ferentiated lights, the ears to sound, less specific. smell to odours, touch to textures, The butyric acid, the hairiness, etc. This is of course a kind of prin- and the warmth form the immedi- ciple of relevance, but a very broad ate objects of the tick, the mammal one indeed. Such a notion could per- as such is the dynamical object. The haps account for “the blackness of difference, however, is that there is two black things” as an instance of no way that the tick, unlike human differentiated lights (where the ab- beings, may learn more about the straction would separate blackness “dynamical object” than that which out from other properties of the hue is given in the immediate one. Mean- and of the things to which they ap- ing here appears as a kind of “filter”: ply). But it seems that the “respect” it lets through certain aspects of the in which there is “agreement be- “real world” which, in is entirety, in tween two things” would often have unknowable, though less so for hu- to be much more precise to charac- man beings than for ticks. The Kan- terise a sign relationship. However, tian inspiration of von Uexküll is of apart form the five external senses, course unmistakable. Indeed, the fil- medieval philosophy distinguished ter model can best be expressed in a synthetic sense (called “common terms of another Kantian philoso- sense”), memory, imagination, and pher, Karl Bühler, who talked about estimation. This would seem to open the principles of “abstractive rele- up the application of the concept of vance” and “apperceptive supplementation”, where the first accounts apparently call this first list “ground” and for the neglect of such physical the second list “correlate”, but I would properties which are not endowed prefer to use the term “ground” for the whole phenomenon, distinguishing, when with meaning, while the second ex- appropriate, the expression ground from plains the projection of properties the content ground. – For the discussion of not physically present in percep- biosemiotics which follows, see, in parti- tion to the meaningful experience. cular, Sonesson 2003a. In fact, Bühler tried to explain such

73 linguistic phenomena as Saussure cisely, it is “an active organisation of and Hjelmslev described in terms of past reactions, or of past experienc- “form” as opposed to “substance”: es, which must always be supposed that certain properties of the physi- to be operating in any well-adapted cal sound may vary a lot without the organism’s response”, with the result units of meaning (the phoneme, the that responses do not occur in isola- word, etc.) being changed; and that tion, but “as a unitary mass” (p.201). other properties which are not phys- The last definition (in spite of intro- ically present may yet be perceived, ducing a socio-historical dimension) because they are expected in the con- is reminiscent of Uexküll’s notion of text. It can now be seen that Bühler’s Umwelt. principles of abstractive relevance This notion of schemes was and apperceptive supplementation used before Bartlett by Janet and go much further than the sign. They Halbwachs, and it has been taken up have been found in the studies of the later by Piaget, as well as by the phe- systems of cooking and clothing re- nomenologist Alfred Schütz. It has alised by Lévi-Strauss, Barthes, and of course also become a fundamen- others (as demonstrated by Sonesson tal concept in cognitive psycholo- 1989a). gy, linguistics, and artificial intelli- The same general idea is found gence, but perhaps sometimes with a in the work of the cognitive psychol- lower intentional depth. Elsewhere, I ogist Fredrick Bartlett (1932: 32, have summarised the results of these 44), who introduced the concept of studies in the following way (Sones- scheme to account for our “effort after son 1988): a scheme is an overarch- meaning”. Bartlett used the notion of ing structure endowed with meaning, scheme in his studies of memory, in which, with the aid of a relation of order to explain the successive mod- order, in the form of syntagms and/ ifications which a story stemming or paradigms, joins together a set of from an alien culture was subjected in other respects independent units to, as the experimental subjects were of meaning. Among its further prop- asked to recount it from increasing erties, two, in particular, are to be temporal distances; but also in or- noted here: a) schemes contain prin- der to explain how one and the same ciples of relevance which extricate drawing was transformed in later re- from each ineffable object such fea- productions from memory, in differ- tures as are of importance relative ent ways according as it had been la- to a particular point of view (this is belled the first time as a pair of glass- Piaget’s assimilation, and the princi- es or as a dumbbell. The scheme is ple of abstractive relevancy, accord- to Bartlett “the setting which makes ing to Bühler 1934); b) schemes also perceiving possible”, and, more pre- supply properties missing from the

74 Fig. 15. The Organon model according to Karl Bühler, with abstractive

relevance and apperceptive supplementation ineffable objects, or modify the ob- criticising the concept of informa- jects so as to adapt them to the ex- tion in information theory, that they pectancies embodied in the schemes are interested in “a difference that (this is another aspect of Piaget’s no- makes a difference to somebody”. tion of assimilation, and what Büh- They go on to say that living be- ler terms apperceptive supplementa- ings “respond to selected differenc- tion; also, it is involved in what Hal- es in their surroundings” (their ital- bwachs and Bartlett call reconstruc- ics in both cases). The formulation tion).69 clearly invokes relevance, and even Returning to modern day bi- some kind of filtering device. Later osemiotics, it can be easily shown on in the paper, however, when the that what these authors are involved Peircean sign concept is introduced, with has nothing to do with meaning the DNA-sequence of the gene is said as sign function, but very much con- to be the representamen, the protein cerns meaning as relevance, organi- its object, and the interpretant the sation, configuration and/or filtering. cellular-biochemical network. It is In their early joint paper, Emmeche difficult to detect any sign function & Hoffmeyer (1991: 4), point out, in here, in the sense in which we hare defined it. According to our authors, 69 Some schemes incorporate (some of) the results of their own use on ineffa- the contribution of Peircean semi- ble objects, and are themselves changed in otics is to show us that “the field of the process, which is what Piaget calls ac- genetic structures, or a single gene, commodation, and perhaps what Lotman cannot be seen in isolation from the calls “internal recoding”. Cf. Sonesson 1988:II.1.3.3. larger system interpreted”(1991: 34).

75 Fig. 16. Peirce’s notion of sign as a kind of “formal sign” – that is, meaning in the sense of relevance This certainly suggests meaning as This implies that the mean- a whole or a configuration. In a lating of the enzyme “is structural” in er paper, Emmeche (2002) sets out to the sense that “the cell’s molecules show that in the living being function form a system of dissimilarities (like and meaning are the same. This can the elements of language in Sau- also be demonstrated, because Em- ssure” (Emmeche 2002: 20). This is meche understands meaning in the of course true to the extent that there sense of function: the relation of the are relevancies in cells, in particular part to the whole. But even in this ar- if these relevancies result from a sys- ticle, there are traces of the filtering tem of oppositions, like those of Sau- concept of meaning: we learn that ssurean language. From this point of “the whole operates as a constraint”. view, everything that is in the cells Indeed, is also in language. But the opposite

Saying that cytochorme c means something cannot be true. There is, of course, to the cell is the same as saying that is has no semiotic function as we have de- a function. It is not just any molecule. We could well synthesise small proteins and ar- fined it. tificially introduce them into the cell. They It may be useful to distinguish would be without importance or they would be dysfunctional or, with certain fortuitous two elements which always go to- strokes of luck, they would actually fulfil gether, both in Uexküll’s notion of some function in the cell” (Emmeche 2002: 19). Umwelt and in the concept of scheme

76 (as discussed in Sonesson 1988; and Lecture 5). Indeed, relevance lets 2003a): organisation, which may de- the difference between “immediate rive from structure or configuration, object” and “dynamical object” sub- and relevance, which may or may not sist, in the vague sense which they be a result of organisation. It is clear retain in the “scholastic” interpreta- that in language, as Saussure under- tion of Peirce (see above): that which stands it, relevance is a result of or- is directly given, in contrast with that ganisation, and more exactly of struc- which is potentially given for further ture. In Uexküll’s notion of Umwelt, exploration. Thus, the principles of it rather seems to be a product of the “abstractive relevance” and “apper- configuration.70 Lacking the compe- ceptive supplementation” still apply. tence, I prefer not to pronounce my- In contrast, filtering simply crosses self on the case of genes. out that which is not let through the It is useful also to distinguish filtering device. relevance from filtering, although The difference between rel- they have something in common: the evance and filtering no doubt has picking up a limited set of features something to do with the capacity from the totality of the environment. to be aware of the borders of one’s However, relevance, strictly speak- Umwelt. It requires some kind of ing, does not exclude anything: it “metacognition”, or, as cognitive merely places some portions of the en- scientists are want to say, “a theory vironment in the background, ready of mind”. To the tick, to paraphrase to serve for other purposes. Thus, in Wittgenstein, the limits of its lan- the case of language, properties that guage are the limits of its world, but are not relevant for determining the not so (in spite of Wittgenstein) to meaning of the words and the sen- human beings. Or rather, the limits tence, still may serve to inform about of our Umwelt are not the limits of the dialect, or even identify the per- your Lebenswelt. son speaking (Hjelmslev’s “connota- According to the phenomenolo- tional language”; cf. Sonesson 1989a; gist Aron Gurwitsch (1974), we may 70 My reason for saying so is that talk about different sociocultural Uexküll insists that the three properties to lifeworlds, apart from the common which the tick reacts form a whole, or an structures of the Lifeworld, which we experiential world, to the animal. This is all share as human beings. Such a so- the sense in which the Umwelt is a subjective concept. Cf. Brier 2001. In deny- cio-cultural lifeworld would then cor- ing the robot an Umwelt, Emmeche 2001 respond to a culture, in the sense of also puts his emphasis on the experiential cultural semiotics. However, the phe- whole. Not being a biologist, I have some nomenologist Alfred Schütz (1967) difficulty seeing why we have to suppose any connectedness between the features to suggested there are really “multi- which the tick reacts. ple provinces of meaning”, such as

77 dreaming, religious experience, the self forces us to go through a series art world, the play world of the child, of “finite provinces of meaning”, in and that esoteric practise we know as the sense of Schütz. A temporal di- science. The peculiarity of the Life- mension is thus added. world, in this context, is that is of- It might therefore be said that fers access to the other worlds, and what most perspicuously differenti- is accessible to all of them. In this ates the tick from the human being sense, the human Lebenswelt is dif- (without prejudging for the moment ferent from the Umwelt of other an- on the question where the exact bor- imals. Or at least is has the capacity der is to be placed) is the structure for being different. of the field of consciousness: in Gur- In Peircean terms, human be- witsch’s (1957; 1964; 1985) terms, ings may reach for the dynamical human consciousness is made up of objects beyond the immediate ones. a theme which is the centre of atten- They may try to transform Nature tion, a thematic field around it con- into Culture. However, as Wittgen- sisting of items which are connected stein observed, even if we had a com- to the present theme by means of in- mon language game, we would per- trinsic links permitting it to be trans- haps not have so much to talk about formed into a theme in its own right, with a lion. The lion, presumably, as well as other items present “at the does not try to go beyond his own margin” at the same time, without Umwelt to grasp the properties of the having any other than temporal rela- objects that lie behind it. There is, so tions to the theme and its field.71 The to speak, no “dynamical object” be- tick of course has access neither to yond the immediate one to the lion. the thematic field nor to the margin. And this is why there may not be In a way, this is simply another way much hope for us ever being able to of saying that the tick cannot reach discuss semiotics with a chimpan- beyond the immediate object. But zee. Gurwitsch’s analysis breaks up that If the Umwelt is a organised of Peirce: it implies that, not only is network of filters and/or relevancies, there no way for the tick to “go on as I suggested in the last section, it seems that maturing in the child con- 71 Gurwitsch is right, I believe, in sists in breaking out of one Umwelt suggesting that this thematic structure translates to language (and no doubt also and going on to another, broader one, to other semiotic resources), as most until reaching the human Lifeworld. clearly illustrated in the transposition of Between each Umwelt and the next, the functioning of pronouns from the per- which encompasses it, there is al- ceptual world to discourse (cf. Gurwitsch 1985); it is unfortunate, however, that he ways a “zone of proximal develop- fails to attend to the difference in structur- ment”. In this sense, ontogenesis it- ing occasioned by the semiotic function. 78 from here” (the Husserlean etcetera Common sense psychology and principle), its experience of the here intentionality and now is also very limited. In oth- Taking into account the Fonseca tra- er words, there is no real “immediate dition, we earlier noted that one kind object” to the tick, not only because of mediation (for which I prefer to it is not opposed to a future more ex- reserve the term “sign”) consists of a tensive dynamical object, but also signifier (expression) which has to be because even in the here and know, perceived as such in order to usher what is immediately experienced into the perception of the correspond- does not appear as a thematic struc- ing signified (content); and another turing, or perspective, on such a dy- one (which following the Brentano- namical object. Husserl tradition, I prefer to call in- I have suggested, then, that an tentionality) which may consist in a important difference between hu- “signifier” which is not ordinarily man beings and (some) other animals perceived as such but still somehow consists in the thematic structure of serves to mediate the perception of a consciousness, or, in other words, “signified” (where it may be better to the function of attention.72 As noted avoid terms as these, ordinarily as- above, there really are two differenc- sociated with the sign function, and es between the way in which ticks simply talk about an item ushering and other lower animals have access into another item). It will be remem- to meaning and the human way. The bered that, according to von Uexküll, first of these is the thematic struc- “in the nervous system the stimulus ture: there is no immediate object, itself does not really appear but its because there is no dynamical object place is taken by an entirely differ- in relation to which it may be seen ent process” (my italics). As human as an adumbration. But there is more beings, as Husserl and Gibson have to it: there is no representamen, ei- insisted, we are alternatively con- ther, if we identify this term with ex- fronted with the-cat-from-one-side, pression, because no distinction can the-cat-from-above, the-cat-from- be made between such a representa- the-front, etc., but what we really see men and the object, either immediate is all the time the same invariant cat. or dynamic. The tick smells the same invariant butyric acid, period. In the world of 72 Differences in the structure of at- the tick, there are no signs, as distinct tention have been discussed in very diffe- from the world itself. Differentiation rent quarters already, although at a much has not even started. But there is higher level separating human beings and no noematic matrix either, properly apes, as well as children of different ages (cf. Tomasello 1999; & Carpenter, Call, speaking. The noematic matrix in- Behne, & Moll 2005; Zlatev 2002, 2003). volves seeing the whole of the thing,

79 but from a particular point of view. ogy, together with naive physics, To the tick, the thing and the point of correspond to the Lifeworld, or the view cannot be separated. In this re- commens, which we have presented spect, even intentionality is beyond in this section. However, it figures the ability of the tick. there mainly as a problem, concern- Intentionality as it is under- ing how (if at all) it might be mapped stood in the Brentano-Husserl tradi- onto scientific psychology. For this tion simply involves the directedness purpose, common sense psychol- of consciousness. Every act of con- ogy is often formulated in terms of sciousness is about something “in the propositional attitudes. In linguistic world”, in a more immediate sense terms, propositional attitudes are ex- than which this is true about signs. pressions beginning with “I think, The relation between consciousness believe, imagine, etc. that p”.74 More and the thing that is the object of generally, if someone is said to have consciousness may be called an in- a belief that p, then he may be said tention. An intention in this sense is to have a propositional attitude with not a purpose, although a purpose is reference to the content p (cf. Ber- a kind of (very complex) intention. múdez 2005: 244ff). Since the verbs Nor should an intention, in this par- used in the formulation of proposi- ticular sense, be confused with an tional attitudes are by definition men- intension, in the sense in which this tal descriptions, it would seem that term is opposed to extension.73 And they should correspond to intentions. yet, as a linguistic phenomenon (but However, if an intention is the fact we will see in Lecture 4 that they are of consciousness being directed to more than that), intensional contexts, something in the world, it seems that also known as propositional atti- the object of an intention is a thing tudes, seem to have something to do (a “substance” or something compa- with intentions. rable to a substance such as a nomi- In Anglo-Saxon philosophy and nalised property), but the object of a contemporary cognitive science, the propositional attitude is a state of af- notion of common sense psychol- fairs (corresponding to a clause).

73 Searle (1995) makes a distinction, 74 In this sense, propositional at- which appears to be similar, between titudes are intensional. If I think about, “intension-with-an-s” and “intention-with- or even perceive, the Evening Star, this is a-t”. The very same distinction was made not the same thing as thinking about, or in Sonesson 1978. In Lecture 4, I will talk perceiving, the Morning Star, although the more about intensions and extensions, in Evening Star and the Morning Star hap- the guise of intensional and extensional pen to be the same celestial body, Venus. hierarchies, and in Lecture 5, I will dis- Although there is thus referential (exten- cuss their relation to different uses of the sional) identity, the two terms cannot be terms connotation and denotation. exchanged with meaning being preserved. 80 It might be argued, however, tent, however, it may stop at the noe- that although that which is the ob- ma of content (also known as the in- ject of the intention is a thing, that tension), or go on to the centre of the by means of which it is intended, the noematic matrix (the extension).75 noema, is a state of affairs. Thus, There is something curious, to intend the dice, one has to enter- however, in identifying common tain the proposition that there is a sense psychology, if it comes in the dice seen from above right, the cen- guise of a set of propositional atti- tral face of which shows four eyes, tudes, with what, following Husserl, against the background of the table- Gibson, Peirce, and others, I have top, etc. Yet one must not forget that, characterised as the Lifeworld, the while this may well be the thematic world taken for granted. If anything, noema within the complete noemat- the Lifeworld is implicit, sediment- ic matrix, in comparison to states of ed knowledge. In Husserlean terms, affairs such as the dice seen from be- a propositional attitude is a “judg- low left, the central face of which ment”, which stands in direct oppo- shows three eyes, against the back- sition to the so-called ante-predica- ground of the floor, and so on, it is tive experience, which is at the or- only relatively thematic, when com- igin of the structures of the Life- pared with the entire noematic ma- world (cf. Husserl 1939). In contem- trix which is identical to the dice it- porary cognitive science, is has been self, according to the phenomenolo- argued that the domain claimed by gy favoured by Gibson and Husserl common sense psychology is really alike. Indeed, in the intentional rela- made up of such tings as frames and tionship, the dice is that which is the- routines (cf. Bermúdez 2005: 172ff). matic and directly given, the inten- This does not seem to be very dif- tion going right through the noema. ferent from my old argument against We are here at such a subtle lev- Searle (Sonesson 1978): you do not el of phenomenology that it is all too see the marks on the desert sand as easy to go wrong. If the sign consists writing because you think someone of two objects, the expression and the has had the purpose of you to see it content, then it seems that the inten- as writing, but, on the contrary, be- tionality of the sign will be directed cause you see it as being a typical in- most immediately to the expression, stance of writing, you take for grant- not as a noema, but as the X that is ed that there must be somebody (if it at the centre of the noematic matrix. But the intention does not come to a 75 It may also in some ways return close there. It goes on to the indirect- to the expression, or to the form of the ly given object, which is the theme of content, which is what Jakobson calls the poetic function and Mukarovsky terms the the sign, the content. Within the con- aesthetic function. Cf. Lecture 6.

81 can only by God, a ghost, and some ent” (Searle 1995: 33, 61, etc.). This other spirit, so be it) that has had the is certainly very different from both purpose for you to see it as writing Husserl and Gibson, both of whom (or more, simply, who has brought tend to reduce the Lifeworld to that about that these marks have the sem- of everyday physics, although none blance of writing, an act which is of them would probably describe the normally made on purpose).76 This latter as independent of mind.77 is a scheme of interpretation, sedi- There is however something mented from earlier instances of ex- different in Searle’s new conception, perience. In this sense, it goes back which in some ways is more similar to earlier judgmental acts, but it nor- to the Lifeworld, that is, the “back- mally operates as a matter of course. ground”, defined as “the set of non- In his more recent work, Searle intentional or preintentional capaci- (1995: 24) similarly rejects the idea ties that enable intentional states of of mutual knowledge (of the type function” (Searle 1995: 129). It is “I believe that you believe that I be- important to note that, in this defi- lieve…”), instead arguing for what he nition, Searle takes “enable” to de- calls “we intentionality” or “collec- scribe a causal, not a logical relation- tive intentionality” being a biological ship, as would be the case in proposi- primitive, not reducible to a combi- tional attitudes, which Searle seems nation of individual intentions. This to identify with intentions. Sear- is an excellent point, but Searle only le also claims that intentional states applies his insight to what he calls are at least potentially conscious, “institutional facts”, identified with which is not true of the background. social reality. Clearly, the Lifeworld The “functions” of the background, in its entirely reposes on collective however, are reminiscent of those of intentionality in this sense. Yet, this the Lifeworld: the background ena- is apparently not what Searle want- bles linguistic and perceptual inter- ed us to understand: according to pretation, such as adapting a word his idea of “our contemporary world with a single meaning to different view”, the physical world is not to circumstances, or finding the duck be understood in terms of “naïve” or or the rabbit in the Wittgensteinean “ecological physics”, but as “natural concepts” which are “language-independent” and even “mind-independ- 77 I am certainly not out to deny the existence of a real world, which is a thesis 76 Formulations like these are nor- Searle (1995) rebukes in the second part of mally made using the expression “having his book. I am simply not convinced that the intention”, but I will avoid this expres- the descriptions stemming from physics, sion and similar ones here, in order not to considered as a natural science, are closer confuse intention in the sense of purpose to this real world than are those of ecolo- with intentionality. gical physics. 82 figure;78 it structures consciousness, Summary so that even in the Mexican jungle, In this section, I have taken pains to we can find the sky and the earth; it distinguish two kinds of mediation, organizes sequences of experience one, for which I prefer to reserve the into dramatic categories; it struc- term “sign” consisting of a media- tures our preparedness in relation tor (signifier/expression) which has to the activity to which we devote to be perceived as such in order to ourselves, as for instance the read- usher into the perception of the cor- iness for other skiers becoming po- responding mediated item (signified/ tential dangers when we are skiing; content); and another one (which fol- etc. These are obviously things taken lowing the Brentano-Husserl tradi- for granted, which we have meet be- tion, I prefer to call intentionality) fore, in the form of the typifications which consists in a mediator which is of the Lifeworld, its temporal hori- not ordinarily perceived as such but zons, the laws of ecological physics, still somehow serves to bring along the affordances, the structure of the the perception of some kind of medi- field of consciousness, and so on. It ated item. Meaning is much broader is not clear, however, why these phe- than sign: it is given already in per- nomena are said not to be intention- ception, notably in the form of index- al. Clearly, in the sense in which in- icalities or neighbourhood relations, tentionality means directedness to or in the form of iconic grounds, an object of the world, they remain or identity relations. In this general intentional, whether we are active- sense, meaning may be understood ly entertaining them or not. This is as a way of picking up selected in- why Husserl would count them as information from the real world, either stances of passive intentionality. As by means of filtering out everything all sedimented acts, they must be ca- else, or by organising the environ- pable of attaining consciousness, at ment into a thematic hierarchy. The least in a phenomenological analy- first case is well known from the sis. And while they may be, in some work of Uexküll and his followers sense, causal (which to Searle means in biosemiotics. The second case is “neurophysiological”), it is not at that more typical of the human Lifeworld. level that they form the background The sign, however, is a peculiar crea- of consciousness, that it to say, the ture of the Lifeworld: it supposes the Lifeworld. concomitant awareness of at least two items, which are subjectively differentiated from each other, while one of them is directly given but not 78 As so often, we find picture inter- thematic and the other indirectly pretation to be taken as the prototypical case of perception. given and thematic. It typically also

83 supposes an (potential) awareness of phenomenon — in the sense, at least, the difference between the sign and of ecological physics. But there are the world, between (to paraphrase, reasons to think that the picture sign partially, Peirce) the immediate and constitutes a decisive step in the crea- the distal content. Among meanings tion of the cultural world – the world other than signs, we may distinguish that is independent of any particular those that are, in a manner of speak- human subject but is only accessible ing, horizontal to the Lifeworld, such through human subjectivity. Signs as inferences or abductions, and are often thought of as being objects those which are vertical to this same the business of which it is to circulate world, that is, the intentional relation- through the world from a sender to a ships connecting subject to their ex- receiver, but it is important to realize periences. Intentionality is much like that signs also have the function to propositional attitudes, but while the conserve meaning, in time as well as former description the direction of a in space. In this sense, signs are mem- consciousness to an object, the latter ory devices. It even seems that those is a description of the state of affairs who talked about signs during the arising from this connection. Signs early “Modern Age” (contemporary are, from this point of view, double with Deely’s late “Latin Age), such intentional relationships. While the as Hobbes and Leibniz, conceived noema in which an object is given is of signs mainly as markers (“notae”) thematic in relation to other noema- for permitting us to remember ear- ta, it is non-thematic when compared lier thoughts, that is, mainly as mes- with the noematic centre; moreover, sages to ourselves (cf. Dascal 1978; in a sign, the noematic centre of the 1983; 1998). But even a culture may expression is non-thematic in relation be said to take notes for its own use, to the content. However, proposition- in which case we are confronted with al attitudes or collective intentional- what Lotman (1979) called “culture ity do not seem to be able to account as collective intelligence”, or, per- for the passive nature of Lifeworld haps better, in an earlier terminolo- meaning. Nor can this meaning, as gy, as “collective memory” (in these meaning, be properly explained in sense of Halbwachs and Batlett). terms of neurophysiological causal- If indeed mental images and ity. (personal) memories are signs, as Pi- aget suggests, then they are certain- 2.4. Steps to the ly less useful for both the purpose of cultural world circulation and accumulation than So far we have been talking about language, pictures, and even ges- the picture sign as an almost natural ture. Indeed, it may seem that it is because meaning may be conserved,

84 in space and in time, that human cul- ings on the surface, they might at ture, with all its variety of socio-cul- first have been created on surfaces tural lifeworlds, becomes possible. (such as sand) which only preserve In some ways, signs may be persist- them for a short time. And it is not ent enough once they are known by easy to establish any clear-cut rela- more than one individual, and may tion between language capacity and be accessed both by the one creating the sophistication of the depictions them and one other person, as hap- (whatever that is). There are, how- pens with gesture and spoken lan- ever, more fundamental reasons for guage. But the sign character, in the supposing pictures to be later in de- sense of the capacity for circulation velopment than language: they sup- and accumulation, becomes even pose a record which is independent more pronounced, once the sign has of the human body; and they require acquired a more enduring materi- us to see a similarity within an over- al embodiment, as is the case with arching dissimilarity. Here we will drawing and written language. It be concerned with the first property. has been suggested by Merlin Don- 79 ald (1991; 2001) that there are several Semiotics is often styled as a phylogenetical discontinuities (which science of communication. However, can be extended ontogenetically, as if, unlike rhetoric and hermeneutics, suggested by Zlatev 2002, 2003, in it is concerned with the resources by press a, b; with Persson & Gärden- means of which meaning is conveyed fors 2005) in the development which from the sender to the receiver (as ar- leads from non-human animals to gued in Lecture 1), the properties of human beings, all involving the ac- these resources become as important quirement of a distinct kind of mem- as the way they may be transferred. ory, considered as a strategy for rep- Within semiotics proper, the Tar- resenting facts. In this story, the pic- tu school has observed that the ac- ture represents a decisive, final step. cumulation of information as well as of merchandise precede their inter- Signs as portable memory change and is a more elementary and Students of prehistoric pictures (such more fundamental characteristic of a as White 2000) often suggest that culture. According to Lotman (1976), creators of such works must have material objects and information are been capable of language. In fact, similar to each other, and differ from not much can be concluded on the other phenomena, in two ways: they basis of the depictions having come can be accumulated, whereas for ex- down to us: even though pictures, by ample, sleep and breathing cannot their nature, must have been made of material which conserves the mark- 79 It is the task of Lecture 3 to investigate the second property.

85 be accumulated, and they are not ab- es in the mind. In both senses they sorbed completely into the organ- maintain a certain distance with re- ism, unlike food, instead remaining spect to the body. Thus far the par- separate objects after the reception. allel holds. Yet Lotman’s parallel is It is interesting to note, that in this arguable in the opposite sense, too: respect, Lotman would not seem to food which he opposes to merchan- agree with von Uexküll and his fol- dise and information may be a kind lowers in biosemiotics, because the of merchandise, too, and it is just as kind of “information” which is tak- accumulable qua merchandise as all en in by the animals within their other kinds; and breathing is an ac- Umwelt (and certainly by the cells) tivity or perhaps rather a process, and appears to be entirely absorbed at the processes can never be accumulated, end of each cycle. not even the processes of transfering In another way, however, Lot- or accumulating (although they can man’s claim is problematic, for it obviously be converted into tapes does not take into account the mate- and records), if not as processes rial resources necessary for making themselves (which may, contrary to up (most) signs. Although Lotman what Lotman maintains, suppose an pinpoints the parallels between mer- incorporation of sorts into the organ- chandise (and therefore, by exten- ism, such as in the case of gesture). In sion, at least as Lotman seems to un- saying that both merchandise (and by derstand the term, material objects), implication material objects in gen- he treats the sign as pure information eral) and information may be circu- (perhaps because he thinks mainly lated and accumulated, it seems that about verbal texts, notably in their Lotman does not say very much. The oral form, where the material base is real question is perhaps in which way extremely mutable), without which and to what degree information and the parallel would have been point- material objects may be accumulated less. Clearly, however, signs are also (and circulated). material objects, and therefore sub- Some of the characteristics that ject to the kind of circulation and ac- Lotman attributes to information are cumulation attributed by Lotman to remiscent of those which are men- merchandise. More obviously than tioned by Masuda (1980), one of the language, a picture is as much a ma- first propagandists of information terial object as a piece of informa- society, but in some respects Masuda tion, as much an artefact as an object appears to say something very differ- of perception. This is why we can ac- ent : in his view, information is not cumulate pictures in a double sense: consumable, no matter how much it as material things, in the safe-depos- is used, and it can be transferred to a it box of a bank, or like experienc- new place without disappearing from

86 the point of origin; it is not accumu- (and of language systems which dis- lated if it is not used as is the case of appear when the last speaker dies – material goods but, on the contrary, or, rather, when the last two speak- by being used increasingly and be- ers do), only the second case applies ing integrated with other informa- to pictures. Pictures must really be tion. Where Lotman pinpoints paral- conserved in a material form inde- lels between merchandise and infor- pendent of the human body.80 Today, mation, Masuda insists on their dif- that material form may very well be ferences, observing that information, a computer record. But also compu- contrary to material objects, may be terised information is dependant on transferred to new places without the wear and tear of the units of stor- disappearing from their point of de- age such as compact discs and hard parture, as well as being used with- discs. out being dissipated and spent; and In this sense all information where Lotman argues that informa- goods are temporarily limited – even tion stays separate from the organ- though some limitations can be of ism, Masuda claims it is integrated relatively long duration. Roland Pos- with other information, which could ner (1989) distinguishes two types of be taken to refer to a process taking artefacts: the transitory ones (as the place in brain structures, but also, sound of a woman’s high heeled shoes more reasonably, could be expressed against the pavement) and enduring in terms of semantic, or more broad- ones (as the prints that the woman’s ly, semiotic, structures. shoes may leave in clay, in particular Against Masuda as much as if the latter is later dried). The transi- against Lotman it is possible to ob- tory artefacts, in this sense, also have ject that even the most elusive kind a material aspect, just as the lasting of information must be incarnated ones; they only have the particularity in some type of material substance, of developing in time, which is why quite apart from the fact that all ac- they cannot be accumulated with- cess to the information in question out first being converted. Normal- depends on some material appara- ly, it is Posner’s transitory artefacts tuses called computers, hard discs whose development in time causes and compact disc player. In the world them to seem somehow “less” ma- of ideas the content of a book exists terial (which is of course nonsense indefinitely; but in reality, it evapo- but must be taken seriously in the rates with the last paper copy which moulders away or the last person that 80 They can, however, be preserved dies or forgets the content. It could be as the capacity for reproducing them, that argued, however, that while the first is, as the sequences of repeatable actions, which is an instance of Donald’s mimetic case is feasible in the case of books memory (for which see below).

87 Lifeworld). It is easy to understand man agent (that is, “anthropogenic” that thinkers of the Enlightenment movements). This suggests that the like Diderot and Lessing could con- first organism-independent records ceive of language (which they tended are indexical, rather than iconic, in to imagine in its spoken form) as a character. However, even if objects “more subtle material” than the pic- like these were independent objects ture that endures in time (at least un- already in prehistory, there is noth- til air is let into the prehistoric cav- ing to prove that they were perceived erns or car exhaust is allowed to dev- as signs, that is, as expressions dif- astate the frescoes of a later time). ferentiated from contents, before pic- Strictly speaking, the sound tures were so perceived. sequence produced by high heels Harold Innes (1950) differen- against the pavement, and other tiates all cultures according as they transitory artefacts, can of course favour more lasting storage media be accumulated (as opposed to be- which are difficult to transport, such ing converted into an enduring arte- as stone tablets, or media which are fact, which is the case of the sound less enduring, but easier to transport tape), in the form of the (typical) leg like the papyrus. In other words, it movements producing this sound, could be said that some media are that is, as a mimetic record, accu- better for conserving information in mulated in the body, but still distinct time, while other do a better job of from it, since the movements can be sustaining it in space – which could learnt and imitated, and even inten- also be expressed in Lotman’s terms tionally produced as signs of (tradi- by pointing out that some media pro- tional) femininity. Posner’s exam- vide mainly for accumulation and ple of an enduring artefact is inter- others for circulation.81 But, again, it esting in another way: the cast of may be better to ask what degree and prints left by the woman’s high heels kind of accumulation and circulation is of course an organism-independ- pertain to different storage media. ent record, just as the marks of a Ro- An even more fundamental man soldier’s sandals found in prehistoric caves, and the hand-prints 81 In a similar way, Metz (1990) has on cave walls. Another case in point claimed that a photograph, but not a film, may very well be the so-called Be- could become a fetishist object, in the rekhat Ram figure (Fig. 7), which, if Freudian sense, precisely because the former has more of a material character. All it is not the likeness of a woman, as cases considered by Innes are of course has been claimed with very little jus- enduring artefacts, as is the photograph tification, could be the result of abra- (while the case of the film is more com- sion produced by regular movements plex); it is only that their capacities for accumulation and communication respec- indicating the intervention of a hu- tively are more or less emphasised. 88 Fig. 17. Semiotic stages of development, according to Donald, in relation to the semiotic function and the different principles question, however, may be what this neutics), as rumor, but also as collec- phenomen called information is. We tive memory. Another name for tra- have supposed so far (as Lotman cer- dition (and rumor) is history – which tainly did) that it can be identified may also comprise prehistory. with what we have called meaning, According to Donald’s (1991: and perhaps even more specifical- 2001) conception of evolution, many ly with signs. Unlike Masuda, most mammals, who for the rest live propagandists of the socieity of in- in the immediate present, are al- formation, also called the knowledge ready capable of episodic memory, economy, have not tried to explicate which amounts to the representation their terms. Clearly, however, the of events in terms of their moment term “accumulation”, used by both and place of occurrence (cf. Fig. 17). Masuda and Lotman, as well as the The first transition, which antedates term “storage media”, employed by language and remains intact at its Innes, suggests that we are somehow loss (and which Donald identifies concerned here with what can be pre- with homo erectus and wants to re- served, not, as material objects, in a serve for human beings alone) brings storehouse, but in memory. Individ- about mimetic memory, which corre- ual memories, however, may well be sponds to such abilities as the con- accumulated (and integrated), but not struction of tools, miming, imita- transferred. In order to be both accu- tion, co-oordinated hunting, a com- mulated and transferable, it seems, plex social structure and simple rit- memory must be social: we know it uals. This stage thus in parts seems as tradition (in the sense of herme- to correspond to what we have called

89 the attainment of the semiotic func- time, knowledge may be stored eter- tion (though Donald only notes this nally to the organism. The bias hav- obliquely, in talking about the use of ing been shifted to visual perception, intentional systems of communica- language is next transferred to writ- tion and the distinction of the refer- ing. It is this possibility of conserv- ent). Yet, it should be noted already ing information externally to the or- at this point that while all abilities ganism that later gives rise to sci- subsumed in this stage seem to de- ence. pend on iconic relations (perceptions of similarity), only some of them are The schemes of perception and signs, because they do not involve memory any asymmetric relation between an In recent time, the notion of scheme expression and the content for which has met with a rare popularity among it stands. writers associated within artificial in- Only the second transition telligence, cognitive science, and lin- brings about language (which, Don- guistics, but the history of the term, ald muses, may at first have been ges- and, to some extent, the notion goes tual) with its semantic memory, that much further back in the scholarly is, a repertory of units which can be literature. The notion of scheme has combined. This kind of memory per- been applied to memory, perception, mits the creation of narratives, that, and action, as well as to the ways in is mythologies, and thus a complete- which perception is anticipated in ly new way of representing reality. memory and built up from action. Interestingly, however, Donald does According to Rumelhart & Nor- not think development stops there, man (1978:41), schemes are “active, although there are no more biologi- interrelated knowledge structures, cal differences between human be- actively engaged in the comprehen- ings and other animals to take ac- sion of arriving information, guid- count of. However, the third transi- ing the execution of processing op- tion obviously would not have been erations”. Examples given by these possible without the attainment of authors, as well as by others within the three earlier stages. What Don- AI, are stories, typical behaviour se- ald calls theoretical culture suppos- quences such as visits to restaurants, es the existence of external mem- menues, etc. Also the cognitive psy- ory, that is, devices permitting the chologist Neisser (1976:51ff) employs conservation and communication of the term, with reference to the work knowledge independently of human of Minsky and Goffman, who, how- beings. The first apparition of the- ever, in the discipline of artificial in- oretical culture coincides with the telligence and sociology, respective- invention of drawing. For the first ly, use the term “frame” to designate

90 the same or similar phenomena; but nents of the AI approach also refer it seems clear from the context, that to the work of the social psycholo- the term “scheme”, as employed by gist Bartlett (1932), who used the no- Neisser, is also akin to “hypothesis- tion of scheme in his studies of mem- testing” as discussed in earlier per- ory, notably in order to explain the ceptual psychology, and to the notion successive modifications which a of “set” in social psychology: story stemming from an alien cul-

A scheme is that portion of the entire percep- ture were subjected to, as the ex- tual cycle which is internal to the perceiver, perimental subjects were ask to re- modifiable by experience, and somehow specific to what is being perceived. The count it within increasing temporal scheme accepts information as it becomes distances. The scheme is to Bartlett available at sensory surfaces and is changed by that information; it directs movements “the setting which makes perceiving and exploratory activities that makes more possible”, but also, more dynamical- information available, by which it is further modified (p.54). ly, an “effort after meaning” (p.32, 44); more precisely, it is “an active Here, then, is first of all prepar- organization of past reactions, or of atory to perception. This definition past experiences, which must always should remind us of the double facet be supposed to be operating in any of the scheme, as it is conceived by well-adapted organism’s response”, Piaget (1967b: 20ff, 25): that is, as- with the result that responses do not similation and accommodation. At occur in isolation, but “as a unitary first, the organism perforce assimi- mass” (p.201). lates stimuli to a pre-given scheme, but at the same time the scheme is Bartlett himself claims his em- modified, as it accommodates to the ployment of the term was inspired by outer environment. In Piaget’s view, the usage of the physiologist Head, to grasp an object with both hands who applied it to body conscious- constitutes, to the 5-6 months old ness (cf. Bartlett 1958:146), but in child, essentially a scheme of assim- the original work, he also alludes to ilation, an incorporation of the outer the psychologist Janet, as well as to world into the self, but in this same the sociologist Halbwachs, and these scheme, there are also factors, such references seems more directly to the as the distance of operation, which point, both because the latter authors must be accommodated to the size of evoke the notion of scheme in the the object, which means adapting the context of a discussion of memory, inner representation to the world.82 and because they do so, like Bartlett (in particular in Bartlett 1923), to em- Both Neisser and the expo- phasize the part of social construction in memory. Janet’s (1928:284ff) 82 Piaget sometimes makes a distinc- indications on this matter are, to be tion between the scheme and the schema, which we will ignore here. sure, very brief: he notes that many

91 people are in the habit of using im- événement d’autrefois” (1925:38f). aginary spatial arrangements, i.e., a This is already the scheme as “schéma tiré de l’espace”, where they conceived by Bartlett; and it already place information they would like to serves as a lattice of pegs on which remember, in the same manner as we individual facts may be affixed. enter an important date in the calen- The tradition from Bartlett dary grid furnished by our diary83. has been taken up again recently, The example he gives is an ancient not only inside AI, but in cognitive Nahua map (i.e. the exodus of Tot- psychology and linguistics. Kintsch omihuaca, Puebla, Mexico), which (1974;1977) has resumed the mem- he shows to be a history book, where ory experiments along the same the imaginary paths form a scheme lines, and has, together with van on which to append the events wor- Dijk (1978), demonstrated, with the thy of notice. Here, then, we are al- aid of summarizing tasks, that “sto- ready concerned with an organism ry grammars” are particular cases of independent artifact, which however schemes. Also taking his point of de- also serves as a series of interrelated parture from Bartlett, Chafe (1977) hints for reconstructing the memory shows how, for instance, the chunk of in the mind (going, notably, from de- experience labeled “my childhood” ployment in space to action in time). is verbalized through a number of As a loyal follower of Durkhe- steps, after being broken down into im, Halbwachs (1925;1951) also in- “subchunks”. In a less precise way, sists on the projection of memory the term “scheme” is also employed onto tangible space but he is even by the art historian Gombrich (1960), more emphatic when it comes to the when considering the historical de- social character of the act of recol- velopment of styles, and by the phi- lection: losopher Goodman (1968), in a dis- En réalité, c’est parce que d’autres sou- cussion of the origin of metaphors. venirs en rapport avec celui-ci subsistent autour de nous, dans les objects, dans les A quite different tradition is, as êtres au milieu desquels nous vivons, ou en it appears, represented by the phe- nous-mêmes: points de repère dans l’espace et le temps, notions historiques, géo- nomenologist Schütz (1932), whose graphiques, biographiques, politiques, don- only indication of sources is a nega- nés d’expérience courant et façon de voir familières, que nous sommes en mesure de tive one, insisting that he is not con- detérminer avec une précision croissante ce cerned with the schema concept fa- qui n’était d’abord que le schéma vide d’un miliar from the writings of Kant. A 83 However, he seems unaware of the scheme of our experience (“ein Sche- fact that a long tradition concerned with ma unser Erfahrung”), as Schütz’s such an ”art of memory” was prominent all through the Middle Ages and the Re- wording more precisely goes, is naissance; cf. Yates 1966; Gomez de Liano ein Sinnzusammenhang unserer erfahrenden 1982. 92 Erlebnisse, welcher zwar die in den er- however, with the exception of Pi- fahrenden Erlebnissen fertig konstituierten Erfahrungsgegenständlichkeiten erfaßt, aget, schemes are not the results of nicht aber das Wie des Konstitutionsvor- individual experiences, but of expe- ganges, in welchem sich die erfahrenden Er- lebnisse zu Erfahrungsgegenständlichkeiten riences inscribed into a social con- konstituierten” (p.109). text. In other words, a series of ear- In an earlier work (Sonesson lier “polythetic acts” are now recon- 1988), relying on the work of Bar- ceived “monthetically”. Once consti- tlett, Piaget, Halbwachs, Janet, and tuted in this way, these schemes are, Schütz, I determined that the scheme as it is later explained (p.111), applied might be understood as an overarch- to the interpretation of other experi- ing structure endowed with meaning, ences. We are reminded of the charac- which, with the aid of a relation of terization of Halbwachs and Bartlett, order, in the form of syntagms and/ according to which the schemes stem or paradigms, joins together a set of from earlier actions and are applied in other respects independent units of to later experiences. This is clear- meaning; and it is constituted out of ly the same procedure which Hus- earlier experiences, i.e. they are sed- serl and Gurwitsch called formaliza- iments of lapsed sequences of behav- tion, and which the second compared iour (although at much higher levels to what Piaget describes as “abstrac- of abstraction for Piaget than for Bar- tion from the action” (as opposed to tlett and Schütz,); and, more specifi- “abstraction from the object); and cally, they are socially constituted, it obviously related to the notion of i.e. the actions from which they de- sedimentation, which I have already rive, and/or their results, arise in in- adapted from Schütz.84 In the theo- teraction with other members of the ries of Bartlett, Piaget, Halbwachs, socium, and thus possess a least some and Schütz, as well as in recent AI, amount of intersubjective validity, the scheme thus seems to be a (more inside the limits of a particular soci- or less) static result of earlier actions, ety. Each scheme contains principles which in turn is applied to present ac- of relevance which serve to extricate tions in order to interpret them. In so from each ineffable object such fea- doing, they connect present actions tures as are of importance relative to and/or objects (and perhaps also ear- a particular point of view (this is Pi- lier and later instances) into a coher- aget’s assimilation, and the princi- ent whole. For all of these thinkers, ple of abstractive relevancy, accord- 84 In later works, Schütz ing to Bühler 1934); and it also sup- (1967:299,327f), describes the sign as plies properties missing from the in- made up of four different schemes, thus effable objects, or modify the objects containing the sediments of experiences so as to adapt them to the expectan- deriving from different spheres of existence. cies embodied in the schemes (this is

93 another aspect of Piaget’s notion of er, etc.). The spatial terms, however, assimilation, and what Bühler terms are said to be generalizations of “a apperceptive supplementation; also, recurrent pattern, shape, and regu- it is involved in what Halbwachs and larity in, or of, […] ongoing ordering Bartlett call reconstruction). Final- activities” as actions, perceptions, ly, the scheme schemes incorporates and conceptions (Johnson 1987: 29, (some of) the results of their own use original italics). This seems to recast on ineffable objects, and are them- the schemes, much like those of Pi- selves changed in the process (this is aget and Schütz, as sedimentations Piaget’s accommodation; and it also of earlier actions, primarily perhaps seems to correspond to what Lotman of our own body in relation to the en- calls the internal recoding of “texts”, vironment. In terms of von Uexküll’s and to the Bachtinean intertext con- Umwelt, such schemes could be conceived as a matrix for engendering ceived as a kind of resegmentation, other “texts). however solitary, of the environment Although it is a much vaguer no- from the point of view of the body. tion, the so-called “image schemas” In turns out, however, that this is not invoked by George Lakoff and Mark at all what is meant by Lakoff and Johnson also seem to correspond Johnson, who postualte some kind to some kind of overarching struc- of innate neurophysicological structure connecting items into coherent tures (cf. Zlatev 2005). It is however wholes. Johnson (1987, 2005; & Rohr- the former notion that we are going er in press), at least, describes image to explore in the following. schemas as being abstractions from the interaction of organism and envi- Collective memory and the ronment. As we have seen, the idea “tragedy of culture” of a spatial, if not specifically bodi- The notion of collective memory, if ly, projection, is important to the no- not that of scheme, has recently been tions of scheme in the psychology and taken up again by James Wertsch, in sociology of Janet, Halbwachs and relation, in particular, to the work of Bartlett. However, while this spatial Bartlett. Wertsch, however, conceives projection seems to take place in real an opposition between the static con- space, much like that realised by the ception of memory attributed to Hal- orator of Antiquity and the Renais- bwachs and a more dynamic idea of sance, the projection with which La- “remembering” for which he makes koff and Johnson are concerned rath- Bartlett responsible. But if schemes er seems to go from the vocabulari- are the result of actions and are ap- ty used to speak about space to non- plied to actions, this opposition does spatial vocabularities (conceiving life not make sense. Not only is the dy- as a voyage, the body as a contain- namic aspect present in Halbwachs’

94 work, as Wertsch (2002:21ff) him- ogies between persons and cultures, self remarks, but the static aspect is Peirce originally put them in place of incorporated into that of Bartlett, by what was later to become the three means of the notion of scheme. This, fundamental categories of First- however, leads to Wertsch having ness, Secondness, and Thirdness. qualms about collective memory be- But Peirce did not identify the sec- ing some kind of super-mind sepa- ond person, as one may at first na- rate from that of individuals. Instead ively expect, with Secondness, but he favours something which he calls with Thirdness. In his view, the sec- a “distributed version of collective ond person was the most important, memory”. not the first: “all thought is addressed To understand the human Life- to a second person, or to one’s fu- world, however, it is necessary to posture self as a second person” (quot- it at least two kinds of social memo- ed from Singer, 1984: 83f). In terms ry, on of them being similar to the that Peirce took over from Schiller, Saussuren language system, which is the first person stood for the infinite a Durkheimean notion, and the other impulse (Firstness), the third person comparable to the Saussurean parole, for sensuousness (Secondness), and which is said to derive form Gabriel the second person for the harmo- Tarde’s idea of conversation. There nising principle (Thirdness). Peirce is nothing mystical about the former: called his own doctrine “Tuism” as Husserl (1962a: 365–386) pointed (from “Tu”, as opposed to “Ego” and out in the case of geometry, abstract “It”), and he prophesised about a “tu- systems are dependant for their ex- istic age”, in which peace and har- istence on some kind of material in- mony would prevail. So the Peircean carnation, but cannot be entirely re- other is a friend and collaborator; he duced to the latter. From the Bakthin is not the spirit that always says no, cercle to pragmatics, there has been the devil in a Biblical sense. an unfortunately tendency to reduce It is striking that not only Peirce, sociality to dialogue, or more gener- but also the late Cassirer and Popper ally, joint action. But there is more to came up with threefold divisions of society than interaction. If we start “what there is”. If one of these in- out form the Ego, there clearly are stances can be identified by subjec- different kinds of alterity: that of the tivity, then all three thinkers would other person (alter), that of the envi- seem to agree that there are two kinds ronment (alius), and that of the sign of alterity. Even though both Peirce system itself (aliquid). and Cassirer, at times, identified the Having recourse to the meta- triads with the personal pronouns, it phor of the three common types of does not seem that they were think- personal pronouns to describe anal- ing about exactly the same thing; nor

95 was Popper. sonal pronouns, it is the third person, The most general sense of alter- and not the second person, which ity seems, at last according to some corresponds to the later notion of definitions, to be contained in Peirce’s Secondness. But this only becomes notion of Secondness: like Berkeley, self-explanatory, when we remem- Destutt de Tracy and Maine de Bi- ber that, to Peirce, the other is nev- ran before him and Sartre after him, er someone who stands opposed to Peirce identifies our sense of reality the Ego, certainly not as in the He- with resistance, that is, “this sense gel-Sartre tradition, but not even in of being acted upon, which is our the more general sense of the Bak- sense of the reality of things” (Peirce thinean conception. Indeed, the sec- 1998:4) ond person is a harmonizing influ- ence. A door is slightly ajar. You try to open it. Something prevents. You put your shoulder The basic problem, however, is against it, and experiences a sense of effort and a sense of resistance. These are not two that Alter is thus given the function forms of consciousness; they are two as- later assigned to Thirdness. But this pects of one two-sided consciousness. It is inconceivable that there should be any effort means the sign as such, which later without resistance, or any without a contrary becomes the incarnation of Third- effort. This double-sided consciousness is Secondness (Peirce 1998:268). ness, has no part to play in the earlier conception. Like the pragmatic This explains that in Peirce’s models I have criticized elsewhere, it early trichotomy, using the three per-

Fig. 18. Peirce’s first trichotomoy, expressed in terms of the three personal pronouns, as related to the categories of Cultural semiotics (Cf. Lecture 8).

96 Fig. 18. Cassirer’s three basic categories, as related to the categories of Peirce, Bakhtin and Cultural semiotics (Cf. Lecture 8). thus presents a situation of commu- a kind of Secondness, just as Ego nication in which speaker, hearer and would be to Alter. In this sense, just referent encounter each other without as the outside world, the sphere of any mediation. Indeed, like pragmat- reference, Alter is something which ics, as well as the Bakthin circle, this resists us, and which we resist. But model tends to reduce the sign sys- even the sign, which is of the nature tem to the interaction with the oth- of law, and thus Thirdness, must par- er (cf. Sonesson 1999b). There is thus take of Secondness, because all sem- no other alterity than the second per- iotic structures impose constraints son (which is not really an other, be- on our possibilities of dialogue, and, cause he is in harmony with the Ego) in the end, of being. – and that of the exterior world. In this interpretation, the tri- As far as I know, Peirce never chotomy is roughly similar to Pop- put his later trichotomy in relation per’s (1972) more generally well- to the three pronouns, but if he had known conception of the ”three done so, I think he should have ar- worlds”, with a different number- rived at a quite different conception. ing: the first world corresponds the If Firstness remains akin to “the infi- third person, the sphere of reference, nite impulse”, then both the Ego and and both the first and the second per- the Alter would basically be of this son pertain to the second world. The kind. But as an Alter, as partner in third world, however, is of the same a dialogue, Alter would already be general kind as Peircean Thirdness:

97 it involves the kind of generality that thing “out there”, but to the to the is the result of organism-independ- world of our objectifications, epito- ent representations. In the sociology mized by “Werke”. of the early 20th century, as well as The latter terms seem to be in latter-day Marxist writings, this is equivalent to the notion of opus that known as objectification or reifica- plays an important part in the theo- tion: the transformation of relations ry of Augusto Ponzio (1993; where between people into artefacts stand- it seems to derive both from Rossi- ing on their own. In a late book, Cas- Landi and Levinas): it is a kind of ex- sirer (1942: 113ff ) argued, against teriorisation of the self (and perhaps Simmel more than against the Marx- also its relations to the other). Indeed, ists, that such processes of objectifi- Ponzio talks about the other as being cation were not only negative phe- only an instance of “relative alteri- nomena, not only a “tragedy of cul- ty”. “Absolute alterity”, on the other ture”: rather, they represented the or- hand, seems at times to involve the igin of culture. material world, at times the world of When later on, in his Nach- signs or opus. Both descriptions are, lass, Cassirer defines the three Basi- in my view, correct. Both the materi- sphänomene in terms of the three pro- al world and the world of objectifica- nouns, objectification is mentioned tions impose much more severe con- only in passing, but it seems essen- straints on our personal being than tial to the whole conception (Fig. the other person as such; they are, so 18.). The first person, the “Monas”, to speak, much less negotiable in the also characterized as “Leben”, is no form of dialogue. doubt close to the “infinite impulse” The suggestions made by Peirce of Peirce (which is not so strange, be- as well as the late Cassirer concern- cause, while Peirce starts out from ing the basic categories (of the situ- Schiller, Cassirer refers to Goethe). ation of communication if not of be- More explicitly than in Peirce’s dis- ing) are fragmentary and difficult to cussion, the second person is not analyse. Nevertheless, even our su- characterized in itself, but precisely perficial considerations may offer as being second to a first: it involves some insights of value to semiotics “Wirken” and “Zusammenleben”, all generally and cultural semiotics in of which is can only be in relation particular. From the point of view of to a first person. However, it is also cultural semiotics, three categories “Wirkung und Gegenwirkung”, just of understanding seem to be insuffi- as the Peircean Secondness, which, cient. It may be necessary to distin- as we have seen, does not concern guish the relationship between per- the second person. The third person, sons (Peirce’s tuism, the Bakhtinean finally, does not correspond to some- dialogue, etc.) from the thing charac- 98 ter of signs (“Werk”/opus/reification). no strategy of representation, and And the latter must be kept separate thus of memory, at all, but at the very from the resistance offered by the most the protentions and retentions material world. Starting out from an of consciousness. However, Don- egocentric definition, however, eve- ald goes on to quote Tulving’s con- rything else turns out to involve dif- cept of episodic memory, referring to ferentiations within the sphere of al- its insertion in space and time, and terity. he observes that, while procedural memory is common to all animals, From mimesis to theory in episodic memory is shared only by evolution and development some mammals, notably apes and Donald’s theory of human evolution birds. Episodic memory therefore al- really posits four different kinds of ready is a quite sophisticated prop- “cultures”, which he also character- erty of mind.85 While memory of izes as different “representational this kind would seem to give rise to strategies”. When introducing the the use of signs in the form of notae first “culture”, epitomized by a as conceived by Leibniz and find its strategy of episodic representation, apotheosis in the calendar, a memo- Donald (1991: 148ff) evokes Tulv- ry device discussed by Halbwachs, it ing’s well-known notion of episodic clearly is not dependent of such or- memory, which corresponds to a rec- ganism-independent representations ollection of events, often in a narra- for its existence. tive form, and involving the time and Mimetic culture starts out with place of the event as well as associ- the emergence of “conscious, self-inated emotions. Episodic memory, itiated, representational acts, which in this sense, is a kind of declara- are intentional [i.e. in the sense of de- tive memory, of “knowing-that”, as liberate, not object-directed] but not opposed to procedural memory, the linguistic” (1991: 168). The examples “knowing-how”. given by Donald are such things as According to Donald, humans gesture, dance, ritual, mime, play- and apes and probably many oth- acting, and (precise) imitation, but er mammals share the capacities for both procedural memory and epi- 85 Indeed, it was only recently that sodic memory. Since Donald (1991: Clayton & Dickinson (1998) showed that 149) characterizes the behaviour of western scrub-jays remember where they cached different food types and discrimi- animals living in episodic culture nately recovered them, depending on the as being “unreflective, concrete, and perishability of the item and the amount situation-bound”, and as a mode of of time that elapsed since caching, which “living entirely in the present”, one seems to suggest they are able to remember the ‘what-where-and-when’ of specific would expect this term to describe caching events in the past.

99 also tool use (or perhaps rather the out the tongue to one who does just social generalization of tool use) and that, and other instances of “neonatal skill. Curiously, Donald (1991: 170) mirroring”), and, on the other hand, claims to have derived his idea of mi- a more explicit capacity for imitation mesis from the literary theorist Eric which matures much later (Cf. Gal- Auerbach, who wrote a history of re- lagher 2005 and Mandler 2004; also alist literature with this very title – see Donald 2001: 264ff). Interesting- although Donald observes what Au- ly, imitation, in this advanced sense, erbach discusses is not pure mime- is claimed by Piaget (1945) to be the sis in his terms. It is not clear, how- origin of the semiotic function. Yet, ever, that this is Donaldean mimesis it would seem that imitation, even in in any sense. Rather, it would have the latter sense, is not necessary what been more fitting to refer to the sense we have describe above as a sign. of the term mimesis in Antiquity, not Or perhaps the different under- perhaps as used by Plato to describe standing of the place of imitation in the relationship between perceptu- Donald’s two works could be referred al reality and the world of ideas, but to the distinction made by Tomasello rather to one of the usages to which (1999) between imitation of the goal, the term in put, mainly by Aristotle’s, of which he believes apes to be capa- as the representation of action by ac- ble, and imitation of the means, which tion, different from (verbal) narra- is a capacity Tomasello would like to tion or diegesis. restrict to human beings, although he In fact, in his early book, Don- later on (in Tomasello et al 2005) rec- ald (1991: 168f) opposes mimesis to ognizes its presence in at least some mimicry and imitation, both of which apes. Indeed, Donald (1991: 168f) are said to be quite common in ani- claims imitation “is found especially mals but lacking “a representation- in monkeys and apes”. At first it may al dimension”. Though the import seem strange that imitating the goal of this claim is not clear, it could be is presented as being easier than imi- taken to mean that mimicry and im- tating the means by which the goal is itation, in this sense, lack differenti- achieved. But no doubt it is less de- ation. In Donald’s (2001: 260f) later manding to recognize the interest of book, however, “(precise) imitation” the aim (getting the banana) than the is an instance of mimesis. Perhaps the interest of the steps requisite for re- difference between imitation as realising the goal. At another level, it is ferred to in these two passages could like attending to the content, not the be taken to involve, on the one hand, expression, of a sign. Indeed, it is an the very early stage of more or less instance of quite ordinary Lifeworld automatic imitation in the infant dis- behaviour. covered by Meltzoff (such as sticking Not only is the means by which 100 a goal is realised not identical, though municative” (or at least, as we shall in some sense parallel, to a sign rela- see “public”), having reference (“in tion, but the imitation of such an act mimesis the referential act must be is not properly speaking a sign. As distinguished from its referent”, that Searle (1995: 40f) points out, while is, in our terms, there must be differ- anthropology texts routinely at- entiation), standing for an unlimit- tributes fundamental importance to ed number of object, and being auto- the emergence of tool use in human cued (produced without an exter- society, they tend to ignore the first nal stimulus and therefore being the imposition of meaning by means of earliest form of “thinking”). As we collective intentionality, which, on have seen, generativity is a property the face of it, seems a much more of many kinds of meaning, which are important dividing line.86 Why, one not signs. However, it is not clear in may wonder, would tool use by part what sense tool use and many other of mimetic culture, and why would kinds of skill are “communicative”, skill is general by such a part? One and therefore, in which way they may wonder whether these types of have reference and stand for un un- behaviour are not simply “routine lo- limited number of objects. comotor acts” or “procedural mem- After introducing “communi- ory” which Donald (1991: 168) else- cativity” as a criterion of mimesis, where takes pains to separate from however, Donald (1991: 172) goes on mimesis. No doubt Donald (1991: to say: 171ff) would answer that they are Although mimesis may not have originated different because they comply with as a means of communication, and might the criteria for mimetic acts in behave originated in a different means of re- productive memory, such as toolmaking, mi- ing “intentional” (that is, voluntary) metic acts by their nature are usually public “generative” (that is, analysable into and inherently possess the potential to com- municate. A mimetic act can be interpreted components which may be recom- by others who possess a sufficient capacity bined into new wholes)87, and “com- for event perception. Given the pre-established primate capacity for event perception, the presence of mimetic skills would inevi- 86 Searle actually talks about the tably lead to some form of social commu- “imposition of functions” in a sense that nication. seems considerably wider than our sign function. Prieto suggested signs were spe- In view of this, I would say that cial instances of tool use, and Eco reduced tool use and other kinds of skill as tool use to the general case of meaning such are not mimesis, because they relationships. As I have argued elsewhere are not communicative, but they are (Sonesson 1989:133ff), I think both these theories are unfounded, though signs and “public”, and they lend themselves to tools have in common being defined by cussed in Lecture 1, and which in earlier something outside of themselves, that is, sections of this lecture were said to relate they are allo-functional. Cf. Lecture 5. to a more general concept of meaning that 87 Cf. syntagm and paradigm, dis- that of sign.

101 imitation – which leads to generali- which would produce a failed act of zation of tool use and skill in socie- imitation. Or, I may merely simu- ty. This is where they become differ- late the outer actions of cracking the ent from routine acts and procedur- shell open, without letting them have al memory. They are socially shared. a sufficient impact on the physical But this is only possible if the act environment, in which case I may can be separated from the unique either be engaged in symbolic play, tool user and transferred to another playing the theatre, or simply prac- user. That is, the act as token must ticing the movements. be abstracted to a type in order to be Imitation may thus be said to be realised in another token. What is differentiated, in the sense of sepa- shared is the type, in other words the rating the mediator and that which scheme of interpretation, which de- is mediated, but it is not asymmet- fines the principles of relevance. In ric, neither in the sense of focus, nor this sense (not in the sense of refer- in that of directness. Indeed, it is re- ence), a single mimetic act may cor- ally the type that is mediated by the respond to various events. token. This also means that the pur- It is therefore by means of impose of the act of imitation is not itation that the “extension of con- to present the original act to anoth- scious control into the domain of ac- er subject (or even to oneself). Ben- tion” (2001: 261) may be obtained. tele (1984) in fact argued against Pi- But the act of imitation is in no way aget that imitation does not manifest a sign. If I see somebody use a stone the semiotic function, but is a pre- as a tool to crack open the shell of requisite for it: indeed, it will func- a nut, I may do the same thing, not tion as a sign only to the extent that to bring into mind the act of the oth- it is taken to refer back to the imitat- er person I have observed, but to ob- ed act, instead of just being another tain the same effect. I attempt to re- instance of the same kind. The same alise the same act as he did, that is, observation should apply to “sym- to open the shell up, so that I can take bolic” play, and is in fact made by out the nut and eat it. Instead of pro- Bentele in another context: the toy is ducing an expression that is non-the- a sign, he claims, only to the extent matic but directly given which refers that the child takes it to represent the to a content that is thematic but indi- real thing, which cannot be true, for rectly given, I am realising a new in- instance, in the case of a toy lion if stance of the category of acts consist- the child has no experience of the ing in cracking open a nutshell. Like real animal. In fact, the extent of the Tomasello’s apes, I may of course knowledge of the child may not be try to obtain the same effect with- the relevant factor, but rather the atti- out attending to the adequate means, tude taken by the child: according to

102 the degree of fictionality of the play opment, talks about “bodily mime- world, i.e. its separateness from the sis” as being based on a cross-modal real world, which is grasped by the mapping between “exteroception (i.e. child (cf. Winner 1982; Gardner & perception of the environment, nor- Wolf 1983) the lion may be made to mally dominated by vision) and pro- instantiate a real lion act or to present prioception (perception of one’s own into to the other children. body, normally through kinaesthet- Acts of imitation in this sense ic sense)” (Zlatev, in press b).89 This have two interesting properties: they supposes a principle of relevance for are “public”, in the very broad sense realising the mapping and it would characterized by Donald, i.e. they also seem to require a record of this may be perceptually, often visual- mapping in the body. However, since ly, inspected; and they can be cop- this is also a property of what Zlat- ied by means of the observer’s own ev calls proto-mimesis (which would body, with or without some addition- include, for instance, “neonatal mir- al implement such as a stone.88 In roring”), such a principle of rele- both these ways, imitation is differ- vance must be capable of being in- ent from episodic memory; and it is nate and/or resulting from a direct different from procedural memory in stimulus instead of auto-cuing. being a public record. Like in proce- Real mimesis, according to dural memory, the record is located Zlatev, would in addition require a in the own body, but it can only func- number of properties which I have tion as memory to the extent that it is already introduced in the definition somehow separable from the body as of the sign: the signifier and the sig- such. While being in the body, it is nified should be differentiated (with not of the body. In fact, this can only reference to my discussion of this be so, to the extent that some memory concept); the subject of the act has traces are instantiated in other bod- the intention (in the sense of pur- ies as well as in the own body. This pose) “for the act to stand for some supposes a distinction between token action, object or event for an ad- and type (that is, relevance) preced- dressee (and for the addressee to rec- ing that of the semiotic function. ognize this intention)”;90 and the act

Jordan Zlatev (in press a, b; with 89 Zlatev (in press a) defines “mime- Persson & Gärdenfors 2005) who tic schemas” as “categories of acts of overt has adapted Donald’s concept of mi- or covert bodily mimesis”. This seems to mesis and extended it to child devel- be compatible with my characterization of schemes, in particular as the mimetic 88 I will suggest later on, in Lecture schemas are said to be not necessary cons- 10, that something more is required for an cious but accessible to consciousness. act having a public character, that is, what 90 If schemes of interpretation are I will call the spectacular function. normally applied as a matter of course,

103 Type of memory Type of accumulation Type of embodiment

Attention span (event in — Episodic time/space)

Action sequence co-owned Own body Mimetic by Ego and Alter

Transient artefact co- In the interaction between Ego Mythic produced by Ego and Alter and Alter

Enduring artefact co- External in relation to Ego and externalised by Ego and Theoretic Alter Alter

Fig. 19. Different types of memory, an interpretation of Donald’s hierarchy is not conventional-normative, nor metic. Here Zlatev also places ordi- does it have system character. Only nary spoken language. cross-modal mapping and differenti- Dyadic mimetic acts are thus ation is necessary for dyadic mime- still not signs. The differentiation sis, such as action imitation, shared they suppose is that between Ego attention, and mirror self-recogni- and Alter, not necessarily, it seems, tion. Triadic mimesis also requires between expression and content. If declarative pointing, iconic gestures however the own body is made to and full joint attention. Mimetic acts imitation the action first perceived that are conventional and/or system- on the body of the other, differenti- ic such as sign language are post-mi- ation of expression and content here coincides with differentiation of self although they may in principle be made conscious, then, as I have already hinted and other. It is, however, important above, and as we shall see in further detail to note that these are two different in later lectures, it is better to define the kinds of differentiation, for, first, sign from the point of view of the addres- this explain why the emergence of see: the addressee takes the addresser to use the expression with the purpose of the sign function can only take place representing the content to the addressee within mimesis, and, second, it rais- and he takes the addresser to have the pur- es the question how this double dif- pose that this purpose shall be recognized. ferentiation is then narrowed down Later on, relying on the Prague school heritage, I will try to give a more adequate to that between an expression sepa- formulation to this observation. I obvious- rate from the body and a correspond- ly take representation to be explicated ing content.91 with what I have called double asymmetry above 91 The notion of “symbol”, as the 104 In describing ordinary language normative existence, which goes be- as post-mimetic, Zlatev would seem yond individuals. In this sense, it is to reject the third stage posited by clearly a constraint imposed on the Donald, the mythic stage, which is individuals, as is Popper’s “objective dominated by language. Yet in terms world” (World 3), a structures which of memory, as Donald originally ex- puts up resistance to the individuals, pressed it, language is certainly dif- in the Peircean sense. Already in its ferent from mimesis. Language may oral form, as conversation and tra- reasonable be thought to have orig- dition, it is part of collective memo- inated as a kind of mimetic device, ry, as Bartlett recognized (but Wert- being different at first, perhaps, be- sch somehow ignored), initiating, as cause it does not rely any more to Donald (2001: 298ff) points out, the any appreciable extent on iconic and/ “collectivity of mind”. or indexical relationships. Once it Husserl’s (1962a: 365–386) de- evinces system character, however, scription of the origin of geometry at least of the magnitude present in may be taken as a case in point. Ge- human languages, it acquires an ex- ometry starts out from the acts ac- istence of its own. In a way, language complished by the land surveyors, only seems to require the presence of which is a kind of skill or even tool at least two human beings to exist, use, and therefore pertaining to mi- who somehow maintain it between metic culture, being subject to imita- them, and when these two speakers tion, though never becoming signs in day, the language also dies. And yet themselves. Indeed, it may be added a language, while it exists, seems to that, at first, the acts of land survey- be something more than its speaking may well have been inextrica- ers. The manifold relationships be- ble parts of more global acts involv- tween its terms must subsist some- ing the practice of agriculture. They where, in a place that cannot be iden- have to be imitated, and thus typi- tified with any individual mind. As fied, in order to become part of mi- Searle observed, language itself is metic culture. Acts of land survey- the foremost language-dependant ing may be sedimented in the form fact. Language is not accumulated in of surveyor’s maps. Husserl, how- the body like mimetic memory, nor ever, is more interested in the way as individual facts in the single his- the general quantitative relationships torically situated mind, as is episod- of space are abstracted out, giving ic memory. More than mimesis, it rise to the mathematical speciality has at the same time a systemic and a known as geometry. Geometry, like language, has an existence, beyond term is used by Piaget, also seems to con- all the fields it may be used to sur- fuse these both senses of differentiation, as we have noted above. vey, in the abstract system of quan-

105 titative relationships we call geom- but overlapping, view, also presented etry, as soon as it can be conveyed by Clark, the notion of scaffolding, at least from one addresser to an ad- as used by Vygotsky, may be taken dressee (who may be the same per- not exclusively in a social sense, but son at another point in time). It gains could be construed so as to comprise in independence becoming a coher- all kinds of extensions of the organ- ent system where everything works ism. together, as in the Saussurean con- Human beings are all “natural- cept of language, unknowingly tak- born cyborgs”, Clark (2003) claims, en up by Deacon (2003) in terms of not only because auditory prostheses “semiotic constraints”. Yet, like lan- may be merged with the organism guage, as Husserl, recognized, ge- substituting for the auditory nerve ometry retains only a precarious ex- making connection with ventral co- istence, a long as it cannot be materi- chlear nucleus at different depths, but alised outside the minds of its users. also, apparently more trivially, but Geometry, as it happens, can be ex- in fact much more importantly, be- ternalized, both as lines and figures, cause, in multiplying a big sum, we and as mathematical notation. This is can use pen and paper to write down the beginning of what Donald calls the intermediary sums, or in order to theoretic culture. remember what we have to do dur- The cognitive scientist Andy ing next week, we can write down a Clark (1997; 2003) has rediscov- list of things to do. Clark’s observa- ered the old (but still excellent) idea tions go a long way in our sense. Yet, (which I traced back to Husserl in there is still one important respect Sonesson, forthcoming b) accord- in which I must disagree with him: ing to which embodiment it not only the social character of the extensions a question of the embeddedness of makes a big difference. the body into the world of our expe- Indeed, the auditory prosthe- rience, but also involves the redistri- sis in not only truer to our idea of a bution of mind onto structures inde- cyborg, as formed more by science pendent of the own body. As Clark fiction writers than by the scientists points out, the environment (in the who coined the term in 1960 (cf. sense of Uexküll’s Umwelt, for in- Clark 2003: 13ff), it is, on the other stance, to which Clark 1997 refers) is hand, not very relevant to the devel- not only that which is in the neigh- opmental history of the human spe- bourhood of our physical bodies, cies. The auditory prosthesis, if any- but also all kinds of artefacts which thing, is an extension of the individ- function in a symbiosis with our or- ual mind. Mathematics and writing ganisms, as bodies and/or as minds. are extensions of the social mind. It According to a somewhat different, is essential for mathematics to exist 106 equally for all of us: not only should ry, are destined for reproduction. In- it be possible for one person to put deed, they permit repeated acts of down an intermediate sum on pa- perception, as do no earlier memory per, and then go back to the calcula- records. The development of the cation later, but it should also be possi- pacity for reproducing the record it- ble for another person to use the in- self has a long history recently giv- termediate sum in his own calcula- ing rise to xylography, photography, tion. Therefore, while there certainly and the computer picture. Writing, are non-social types of scaffolding, like pictures, is a result of the graph- social scaffolding appears to be im- ic act, the creation of marks on a sur- mensely more important for under- face. Yet these marks, considered as standing human evolution. tokens, have a different relation to Visuographic markings first the corresponding types, because the appear, according to Donald (1991: types must precede all their tokens.92 276ff; 2001: 305ff), with Marshack’s Moreover, writing, unlike picture, engraved rib from Pech de l’Azé in but similarly to language, follows a France, which is however an isolat- linear order, in Western writing sys- ed instance (if it is anything at all). tems from left to right. The anthro- It is followed up later by purposeful pologist Jack Goody (1977; 1982), arrangements of objects in ritualistic who, going beyond Vygotsky’s lin- settings, as well as by pictorial repre- guistic determinism, has argued for sentation epitomized by cave paint- the sway of writing, has however in- ings. The existence of pictures allows sisted in particular on such examples language to be given a visuo-graph- as lists, tables and recipes, which ic representation, which we know as precisely are not read in an exclu- writing (but which would also in- sively linear fashion, but are over-de- clude geometrical notation). Writ- termined by a visuo-graphic charac- ing and pictures together permit the ter approaching that of pictures. emergence of science, which is inde- However, it is important to re- pendent of individual minds not only alise that, even when marked out in as representation, but also, at least in the sand (as were Archimedes’ cir- its aspiration, as referent. cles), pictures, or more broadly vis- Ivins (1953) pointed out that it uo-graphic structures, are spatially, is the reproducibility of pictures (as though not temporally, organism-in- in Floras, for instance) that makes dependent artefacts. This also applies, them into scientific instruments. In of course, to the writing in the desert this sense, in their capacity of being sand imagined by Searle. Neverthe- permanent records, pictures are not, 92 On this difference, see the last part as art historians are wont to say, un- of the lecture, and also the discussion of avoidably unique, but, on the contra- Goodman in lecture 4.

107 less, no spatial record can be entirely referring to different tokens instan- outside of time. Drawings and writ- tiated in different bodies. Typifica- ings in the sand simply have a very tion, in this sense, does not require limited temporal life span. This still the semiotic function, but is probably allows them to be objects of repeated a prerequisite for it. Mythic memo- acts of perception. Indeed, according ry (which I would prefer to call lin- to the Antique story, Archimedes, on guistic memory or perhaps, as Don- being surrounded on the beach, told ald sometimes does, semantic mem- the soldiers not to disarrange his cir- ory) is different again: it has a sep- cles. To finish his accounts, he need- arate existence, but, like some kind ed to perceive them again. To have of real-world ectoplasm, is requires the status of theoretic records, there- the collaborative effort of a least two fore, pictures simply have to be spa- consciousnesses (which no doubt tially organism-independent; to be have to be embodied) for this exist- available to our archaeology, howev- ence to be sustained. Transitory arte- er, they also must be temporally or- facts, as verbal language or (as Pos- ganism-independent. ner would have it) the sound of high- The artifact, then, is that which heeled shoes on the pavement, ac- bridges the distance between the act quire a body only to the extent that of addressing and the act of receiv- a sender and a receiver agree rough- ing. But if the record has at least a ly on what they are. Only theoret- minimal temporal existence, there ic memory has a distinct body of are indeed several distinct possible its own: it subsists independently of acts of reception. the presence of any embodied con- Episodic memory is most clear- sciousness, because it itself embod- ly disembodied. It may refer to a bod- ied. It has acquired the ability to per- ily act, such as going in our out of a sist independently of human beings. container-type object (such as, for in- Of course, without anybody around stance, making love), but it is unable to perceive it, organism-independ- to generalise this movement beyond ent records are not of any use. With- a particular moment and place, and out any human beings present, they thus it does not give rise to any kind are really worse off than the famous of independent embodiment. Mimet- acorn falling from a tree without an- ic memory still accumulates in the ybody around to hear its sound. own body, but it only becomes such, The three world circulations: to the extent that what is recorded Mates, money and love in the body also exists elsewhere, in at least one other body, which sup- Even if accumulation may be as im- poses generalisation or, more exact- portant as circulation, as Lotman ly, typification, the creation of a type maintains, also the study of circula- 108 tion may need to be revised and am- sise that they all imply the presence plified. When studying circulation, of language or other signs, so that, in the comparison between signs and the end, in may seem that this more goods once again come to the fore. general science is semiotics itself.93 According to Lévi-Strauss (1958: In an early work, Dan Sperber 329), there are three vast circulations (1982) has taken exception to these going on in the world: the circula- parallels, arguing that, while circu- tion of words, of merchandises, and lation is a constitutive factor of the of women. They are studied, in turn, kinship system, it is only an acciden- by linguistics, economy, and social tal property of language, which is anthropology. Jakobson (1990: 19f, essentially a repertory of messages; 460f) took this idea up and extended and when information has circulated it: the three circulations concern mes- for a sufficient time, we will all be sages (not only verbal signs), com- in possession of it, but a woman or a modities (which comprise goods and horse which is exchanged is lost for services), and mates (men or woman the donor; and while language signi- as the case may be). The sciences that fies by means of a code, women only study these phenomena are semiot- acquire meaning by means of the at- ics, economy, and social anthropol- tention being directed to them.94 It is ogy in conjunction with sociology. easy to agree with the general drift The latter addition is perhaps not cir- of Sperber’s argument, but some- cumstantial: Lévi-Strauss is thinking times he is widely off the mark. To about the kind of societies studied by begin with, a language that does not anthropology, in which friendly rela- circulate (i.e. is not used in any acts tions are established between tribes of communication) is not much of a by one tribe giving wives to another, language; in fact, it is what we call which then may give wives to a third a dead language (like Latin, or He- one, until, in the end, the first tribe brew until it was reborn). On the oth- receives wives back from one or other hand, the circulation of women is er tribe in the chain of exchange. In certainly not constitutive of women. the societies studied by sociology, on In fact, I think that, in the kinship sys- the other hand, the circulation would tem, women do not signify at all; it is rather consist in one man and one woman given themselves up to each 93 In fact, Jakobson’s position as far as the different sciences goes is much less other (or so the rhetoric goes). Jakob- clear-cut than I suggest here; cf. the passa- son and Lévi-Strauss agree that these ges referred to above. Roosi-Landi (1983: sciences studying circulations are all 73) actually claims economics is a part of part of some more general science semiotics. which they call the study of commu- 94 This would correspond to the no- nication, but Jakobson also empha- tion of meaning as relevance discussed in the section above.

109 Position of Displacement S e ve r a l displacement Final stage: of receiver B A B B A a at B, nothing Transport a → a no — at A Exchange (of a → a, a at B and b women, goods, no — b ← b at A etc.) Language a → a a at A and B possible — Picture a at B (somtimes a → a no — postcard also at A) Television a1. ...bn at A1. film (and to a → a1. picture, film ...Bn yes some extent ...bn picture (usually at A) TV): b ← b a → a1. a1. ...bn at A1. Publicity poster yes b ← b ...bn ...Bn

Fresco painting — a at A no b ← b

Fig. 20. Some varities of circulation for different kinds artefacts the act of exchanging them that car- may even carry meaning as the in- ries meaning. And this is certainly a dividual person she is: even after re- difference to the exchange of signs, ducing the message to make transla- in which the latter carry at least the tion possible, as Lotman (1979: 91) so primary sense, which the exchange nicely puts it, the message may still serves to convey. In fact, it is easy contain indications for reconstruct- to imagine a way in which a wom- ing the personality of the other (cf. an, arriving from one tribe to anoth- Sonesson 1987; 1992a: 91ff). er, does carry meaning in herself: Suppose, however, that it is re- speaking another language, having ally the woman (or, more generally, different customs, etc., she may ap- the mate) as such which is the mes- pear as a “non-text” to the members sage. This would presumably make of the receiving culture.95 In fact, she her into a kind of “natural meaning”, in Grice’s sense, similarly to the way 95 In the sense of the semiotics of in which red spots mean measles, or culture (on which something will be said in lecture 8). This may be to suppose too much heterogeneity between tribes that properly to women or men marrying into exchange women; it applies much more another society at the present time. 110 clouds mean rain, as opposed to the the cloud only means rain to those “non-natural meaning”, epitomized who know about the relationship be- by language (and, I suppose, money). tween clouds and rain, and who for In this view, there is an identity be- reason of the Lifeworld choose to ig- tween cause and expression, on the nore other causes. Red spots of a cer- one hand, and effect and content, on tain type only mean measles to those the other hand, the cloud being both who know about the symptoms of the cause and the expression of the measles, and who do not care to take rain; or between cause and content, other causes or effects into account. on the one hand, and effect and ex- The woman means “effect of an expression, on the other hand, the read change with another tribe” only to spot being both the expression and those who are familiar with this kind the effect of measles. Non-natural of exchange pattern, and who think meaning, as in language, on the oth- this is the only (or most) relevant as- er hand, relies, in the conception of pect of the woman in question.96 the Griceans, on the recognition of If the woman of the mate ex- someone having the purpose to com- change is really a message, then her municate something, on this purpose circulation as a message in depend- being recognize, and so on. But what ant on her circulation as a material would the woman mean in this case? object. But signs do not have to cir- I suppose something like “effect of culate, in this material sense at least, an exchange having taken place”. in order to be signs. They certainly Interestingly, however, in his have to cover the space between the later existence as a Gricean, Sperber, addresser and the addressee, but this writing together with Deirdre Wil- does not have to be a space in the real son (Sperber & Wilson 1986: 53f), world, however small. And signs may denies the existence of two kinds of travel from very far (as many signs meaning: there is a continuum behave undoubtedly done so in time as tween that which Grice calls natu- well as space) without being able to ral and non-natural meaning. In do- function as signs, if there is no coming so, however, Sperber & Wilson mon system of interpretation. seems to reduce all meaning to “rele- Communication in the materi- vance”, without there being any prin- al sense (in the sense of the current ciple to the relevance, which amounts spatial metaphor) really implies that to some kind of “natural meaning” something which leaves one place which includes the manifestation of is not there any more when it ar- purpose. On the contrary, I think rives at a second place: this is true of there must be a principle determining the train, as well of the letter which what is relevant also in what Grice 96 More will be said about this kind (1989) would call natural meaning: of meaning in later lectures.

111 it may transport, and even of con- Lisa will remain at the point of ori- tent of the latter, but not of course gin while being sent of to some dis- of the units of which the message is tant place: as a type, if not as a to- made up. The circulation of wom- ken (cf. Sonesson 1992a: 91ff). Thus, en (and of mates generally) as well circulation, like accumulation, has as of commodities suppose a double more to do with the kind of temporal movement from one place to anoth- and spatial artefact in which the sign er: one tribe gives women to another is embodied then with the sign func- tribe and receives women back (or a tion as such. man and a woman “give themselves Indeed, circulation, in the case up” to each other); and when receiv- of signs, seems to be a misleading ing a horse, I give money or perhaps metaphor, not only because it sug- a donkey back. But the exchange of gests transport (spatial displacement) signs is not necessarily double; it and recoding (the translation from does not even necessarily imply any one code to another), but in partic- spatial movement in the Lifeworld. ular because it conceives sign crea- A television picture or a web page is tion as an unitary happening, bind- transferred from afar but they are not ing one addresser to one addressee. perceived to move in space. It seems In order to take into account the dif- rather absurd to speak of the mean- ferent kinds of “circulation” which I ing of a fresco painting being trans- have described above (and many oth- ferred by circulation – though there is ers not mentioned here), we should of course a movement of the photons rather have to conceive of one (or from the rocky surface to the eyes of perhaps several) act(s) of addressing, the observer. A fresco painting is an followed by what is normally many example of a sign that would certain- acts of reception, mediated by tem- ly not remain at its place of origin if porally and/or spatially enduring ar- it were transferred to a museum. In- tefacts. deed, it is an instance of a sign sys- Apart form Lévi-Strauss, the tem where it is the addressee that has author most responsible for the iden- to seek out the message, rather then tification of “two basic modes of hu- the opposite (Fig. 20). man behaviour, /…/ the production But neither would my draw- and circulation of goods (in the form ing of the street systems remain in of commodities) and the production my study if I send it to you by the and circulation of sentences (in the post. As we said, it is also true of form of messages)” is no doubt Ros- the accompanying letter, though not si-Landi (1983: 65), who calls these of its constitutive elements. Howev- two modes “non-verbal” and “verbal er, there is a sense in which a picture communication”, respectively. It is postcard or a reproduction of Mona interesting that, in addition to circu- 112 lation, Rossi-Landi attends to paral- “the corporeity of, for instance, roast lels between production, not accumu- chickens, lies in the fact that they can lation, as Lotman suggested. How- be eaten” (which I take to be his ex- ever, on both counts, the comparison tra-sign, which would correspond to seems flawed from the beginning: a Gibsonen affordance), but, in ad- the term “non-verbal communica- dition, chicken is also “upper class tion”, which is a misnomer already food in one country and everyday, if in its common usage to refer to ges- not actually cheap fare, in another” ture, facial displays, paralanguage, (which I suppose are instances of his and the like, is here extended so as to non-verbal signs but which I would include practically everything in the rather describe as cultural affordanc- world which is not verbal communi- es). But if it is true, as Rossi-Landi cation, such as politics, economics, says, that “we must distinguish be- law, fashion, cuisine, etc. Curiously, tween the production and consump- Rossi-Landi still opposes these “ver- tion of the body and the production bal and non-verbal signs” to “non- and consumption of the sign”, then it signs”. However, the only basis of does not seem that material produc- the comparison seems to be the fact tion, consumption, and circulation of exchange (which, as we have seen, have much to teach us about the paris not necessarily a fact as far as real allel functions (to the extent that they signs are concerned). It might indeed exist) in signs. be profitable, as Rossi-Landi claims, The comparison between mon- to analyze commodities in the termi- ey and signs was made already in nology of signs, and vice-versa, but Saussure’s Cours, where it was for- such a comparison would have to at- mulated in terms of “values”, prob- tend also to their difference. It is, in ably only to bring home the impor- fact, easy to agree with Rossi-Lan- tance of the interrelationships be- di (1983: 68) that “a commodity is a tween the items making up the sys- commodity, rather than a mere prod- tem. Basically, money is only a par- uct, because it is a message” (his ital- ticular instance of goods, conven- ics) – but this is so, exactly because tionally taken to be the equivalence something has to be added to the pro- of any other kind of goods. In this duction of a good, in order to make it sense, we should expect it to obliga- into a commodity. In the end, Ros- torily circulate in a spatial sense, as si-Landi (1983: 71ff) actually knows goods do, not only optionally, as is this, because he observes, with ref- the case with signs. This is of course erence to the Lévi-Straussean wom- no longer true, when a money trans- an, that, apart from being a message, action can be made by pressing some she is “extra-verbal and also extra- buttons on the Internet page of the signs”. He goes on to observe that, bank or the Internet store. Within

113 a very different tradition, money is seen as a continuing conversion of one of the instances of “institutional signifiers into signifieds, gold stand- facts” most thoroughly discussed by ing for exchange value (to which it Searle (1995: 32ff, 37ff). Money is in is indexically related), paper mon- Searle’s view a kind of “status func- ey standing for gold, and electron- tion” (“X counts as Y in C”), just ic money standing for paper money. as chess and language, that is, it is This description is true enough, but a “language dependant fact”, wheth- it raises the question what the next er it is commodity money, which may step may be. However, Hornborg’s constitute of gold or other things re- further discussion seems to indicate garded as valuable in themselves, that all money, at least in Western contract money, in which the value society, is fundamentally deprived is ascribed to the promise to pay the on meaning, which makes it into a bearer the equivalent amount in gold, very curious sign indeed. Accord- or fiat money, which are simply piec- ing to Hornborg (1999:151), money es of paper declared to be money by is “a code with only one sign” (his some official agency such as a cen- italics), which would be like “imag- tral bank. Commodity money is, of ining a language with one phoneme, course, as we noted above, simply a an alphabet with one letter, or a DNA privileged type of commodity. As for molecule with only one kind of nu- fiat money, as presented by Searle, it cleotide”. This is of course a strange still has some kind of embodiment, thing to say (quite apart from the fact in a Husserlean sense, but the mate- that the word, not the phoneme, is the riality of Internet transactions seems elementary sign of verbal language), to be considerably more subtle. because all kinds of currency appear In the posterity of Saussure, to be made up of different units (such the most recent instance of the mon- a “euro” and “cent”), to which fur- ey metaphor seems to have been of- ther denominations are added by the fered by Alf Hornborg (1999; 2001a, number system. Indeed, this is prob- b), who continues to consider money ably why Saussure chose to com- to be some kind of sign, although, in pare language to money in the first my view, he gives very good reason place.98 for abandoning this identification.97 98 A sign system having only one Hornborg suggests that what has hap- sign, as Prieto (1966: 43ff) argued, would pened to money historically could be be for instance be the white cane which signifies that its bearer is blind. This is 97 It must be noted, however, that, so only because the absence of the white although he refers to both Saussure and cane does not signify that the bearer is not Peirce, Hornborg (2001b) employs the blind, which is different from sign sys- term “sign” is a very wide sense, which tems having more signs, such as the flag includes what we would call meaning, of the admiral’s ship, where the presence specifically, perception (“sensory signs”). of the flag stands for the presence of the 114 It soon becomes clear, how- Type ever, that Hornborg is really thinking about something very different, which, with Benveniste’s (1969) term, may be called the “domain of validity” of the system, that is, the limit- Token 1 Token 2, 3, etc. ed content resources. He opposes the Western concept of money to that of Temporal axis pre-modern societies such as the Ni- gerian Tiv, where there are three dif- Fig. 21. The temporally bound type of ferent kinds of value, that is, three typicality different kinds of circulations of ob- but only for more love, friendship and jects, which do not connect with each honour. With such exceptions, how- other. Indeed, not only is it possible, ever, the whole domain of goods can to express it in more adequate terms, be exchanged for money in Western to have several different money sys- society. To this may be added a pecu- tems, each with its own domain of liar “mode of operation”, in Benven- validity, between which no exchange iste’s (1969) sense, that is, a limitation is possible (contrary to what happens of expression resources, because, as in the case of the currencies of dif- Hornborg (1999: 153) notes, quoting ferent countries), but, at least at this Polanyi, “’only quantifiable’ objects point in history, it is still true that may serve as money”.99 If love is “all societies recognize spheres of only to be exchanged for love, then, I human life which are not to be medi- take it, love would not be money, be- ated by money” (Hornborg 1999:157; cause it is not quantifiable. his italics). Although Hornborg does The correlate of money being not give any examples, I believe it is able to stand for everything it that taken for granted in our society that it is unable to stand for anything in such things as love, friendship, and honour are not to be had for money, 99 On these terms, cf. Lecture 4, where they are used to compare language admiral on board, and the absence of the and pictures, following the inspiration of flag for his absence. Lessing.

Type Fig. 22. The Ex. phoneme, temporally free type word, of typicality standardized Ex1 Ex 2 … object

Temporal axis

115 Fig. 23. Linde’s new tokens of Duchamp’s ready-mades

particular: as Hornborg (1999: 153) cause there is not even a remnant of observes, money does not corre- natural connection between signifier spond to any particular concept. It and signified. But Hornborg must be might be more correct so say, how- wrong about this: in fact, Saussure ever, that money only corresponds (1973: 115f) does not say that coins to the concept of monetary value, and words may be exchanged for un- which is really the same thing as say- like units, such as commodities and ing that it is limited to a very nar- concepts, respectively, as Hornborg row domain of validity.100 Still, this quotes him to say, but for work and means that it does not make sense to concepts, and he goes on to contrast say that money is somehow direct- the natural relationship in the first ly given but not thematic while that case with the arbitrary one in the sec- which it is exchanged for is indirect- ond case (which should not be sur- ly given and thematic. Hornborg also prising since Saussure always tends claims that money cannot be a “sym- to single out the arbitrariness of lan- bol” in the Saussurean sense, be- guage). Clearly, Saussure has an idea of the “true value” of things, meas- 100 If most things in our society may ured in the amount of work, as we be bought for money, then the domain of validity of the money system may not know it from Ricardo and Marx. As appear to be particularly limited. Here Marx recognised, however, this does we must separate the intensional and the not really describe the way money extensional domain. Money redescribes has been functioning in Western so- everything from the point of view of their monetary value. This only becomes a ciety over the last few hundred years. problem when the point of view of mone- Still, I think there is some truth to tary value is the only point of view that is Saussure’s observation: as a special sanctioned by society. 116 kind of exchange of commodities, ist. Considered as a totality, this arti- money is basically of the same kind cle is a single type, but it will appear as that for which it is exchanged. But in as many tokens as this lecture is the signifier is not really of the same downloaded. As a first approxima- kind as the signified. I may take some tion, it seems that it is a sign the money for my work instead of the type of which provides for more than food I really need, but I would hard- one token which may be universally ly accept the signifier “food” in place shared, in the sense of Sperber and of its signifier. Masuda, and which may be said to Like the woman of the mate ex- remain at the point of origin while change, money only signifies in a sec- being sent out to circulate. ondary way, because it stands for the Yet it is not evident that the react of exchange of which it is a part. lation between type and token al- The circulation of mates and the cir- ways is of the same kind. It seems culation of goods are really first of reasonable to say that a painting first all circulations, and then they may must be made in one exemplar be- be made to signify the fact of cir- fore existing as a type; the first ex- culation. Even though a sign which emplar serves to establish the type, does not circulate is not much of a from which then further exemplars sign, circulation is not constitutive of can be derived. In the same way, the sign-hood. first exemplar of an article must be written by the author, before a type is Two types of types, with established, which then plays the role iconicity and mimesis of a directive guideline for the dif- The difference between type and ex- ferent exemplars that are later creat- emplar is described by Peirce with ed. In the case of phonemes, words, the terms “type” and “token” (or musical notes and so on, the proce- “replica”). In the previous phrase, dure is different: there is not a first for example, the word “and” appears “l” which only then creates the type once, considered as a type, but twice which is then repeated. It may cer- considered as a token. The letter “t” tainly be possible to determine when is also one type only, at the same time a phoneme, or in any case a word, that only in the first sentence of the was used for the first time, but nor- paragraph there appears eight tokens mally this is not relevant for the na- of it. This reasoning is easily ex- tive speaker. And to the extent that it tended to other systems of meaning; becomes relevant, the typicality has a reproduction of Leonard’s “Mona changed its character. Lisa” is of the same type as another It will be convenient to distin- reproduction, but they constitute two guish between temporarily bound exemplars or tokens of those that ex- and temporarily free relations be-

117 tween type and token; in the first case tokens. Thus, also pictures may still (Fig. 21) but not in the second the be present at the place of origin while type (Fig. 22) is established in time reaching other coasts. by means of the creation of a first ex- The distinction between tem- emplar (cf. Sonesson 1998c). Tokens porally bound and temporally free of temporally free types may be sent types is not identical to the one that off in all directions, but the types Goodman (1968) makes between are still always available at the place autographic and allographical arts. of origin and elsewhere. The case Among the temporarily bound typ- of temporally bound types is more icalities previously mentioned, the complex. Written texts are temporal- verbal text is allographic, whereas the ly bound types, but they are entire- visual work of work is traditionally ly made up of temporally free types. autographic; in other words, the art The scribbles made by the famous work, but not the work of literature, author on the back of his bar bill may is defined as to its identity as well as remain the only tokens of their type, to its value within our inherited so- if they are not rescued by his editor. cial practice by means of its tempo- But once these notes make it onto the rary association to the first exemplar printing house, they are made availa- created by a certain individual. This ble everywhere, at the bar where they is why we do not have to queue up in were written as well as at any other front of the Stockholm National Li- place. Until recently, a picture was brary to read the only exemplar of almost always, from the very begin- “Röda Rummet” written by Strind- ning, a temporally bound typicality: berg, while a similar conduct is ex- whether it was a drawing or a pho- pected of us in the case of a work tograph, all its elements where tem- of visual art. The copy of the novel porally bound — although the pho- that we may buy at any bookshop is tograph is more easily made into a a token of the temporally bound typ- first exemplar engendering an indef- icality produced by Strindberg, and inite number of tokens. With the ad- so is the reproduction of Mona Lisa, vent of computer graphics, however, which we can buy at the Louvre. But a picture can be made from the com- our current social practice assigns bination of temporally free entities, different values to these two instanc- whether these are items of clip-art or es of multiple tokens from one type. scanned images, or the product of al- To understand the decisive gebraic formulas. This means that, events of the history of the art, we also in the case of pictorial commu- have to take into account the entire nication, both the temporally free hierarchy of values present in our and the temporally bound types may society: the radicalism of the first give rise to an indefinite number of ready-made that Duchamp exhibit- 118 ed did not consist in treating a tem- Summary porally free typicality as if it were a Accumulation, as Lotman said, is temporarily bound one. In fact, the- just as important to signs as com- bottle dryer and the urinal also de- munication. The picture, just as any pend on typicalities that are tempo- other sign, may be seen as a memory rarily bound. They are founded on device, a tool for accumulating in- some kind of prototype. The differ- formation. As such, it is at least more ence is to be found in the different complex to produce (though not nec- values attributed to the first exem- essarily to interpret) than verbal lan- plar that creates the type in the pro- guage, since, unlike oral language, duction of the object of use and the but similarly to writing, it supposes work of art. By signing the urinal the presence of organism-independ- (with a false name, to be sure), Du- ent vehicles of representation. Fol- champ did not only transform a tem- lowing Merlin Donald, pictures are porally bound type with an indefinite precursors of theory in phylogeny, number of tokens into a type having and thus perhaps, as others have sug- only one token, but he also transmut- gested, also in ontogeny. The model ed an allographic entity into an auto- of communication, which poses an graphic one. And this is where the analogy between the conveyance of adventure of Modernist art begins: information and transport in space, very soon (thanks to Linde), Duch- is problematical on any account, but amp’s urinal becomes a type engen- particularly so, in the case of pic- dering new tokens. But to exist as art tures. As a vehicle, the picture, like it still needs autography – the signa- any other material objects, would ture of the author (Fig. 23). seem to disappear from the place of If we know return to the act of origin, once it is conveyed to another imitation, as it pre-exists to the sem- place. But this is only true of paint- iotic function, it seems to have more ings (and not even applicable to fres- in common with pictures than with coes); other kinds of pictures, from language. Unlike truly mimetic re- picture post cards to web pictures, sources, and unlike language, the clearly exist as types, independently act of imitation, similarly to the pic- of their material manifestations. Also ture, first has to create the typicality, pictures have types, distinct from which is then used for reproduction. their tokens. But unlike the elements Between mimetic resources and pic- of language (and like compound lin- tures there seems to be more similar- guistic signs such as verbal texts), ity than simply the predominance of pictures are temporally dated types: iconicity. In the next lecture, we will they have an origin in time, to which look at a particular kind of iconicity, any token must refer. In this respect, which we will call pictorality. pictures seem to be more similar to

119 the act of imitation than language. Entwicklung. Vorüberlegungen zu einer Although decidedly post-mimetic genetischen Semiotik. Tübingen: Narr. Benveniste, Émile, (1969). Sémiologie de la and post-linguistic, they recuperate langue, in Semiotica I:1, pp. 1-12; and I.2, some of the properties of mimesis pp. 127-135. not only in the sense of iconicity. Bermúdez, José Luis (2005) Philosophy of psychology. New York & London: Routledge. Bierman, Arthur K. (1963) That there are References no iconic signs, in Philosophy and phenomenological research, XXIII:2: Anati, Emmanuel (1976). Evolution and 243-249. Boysen, Sarah, & Kuhlmeier, Valerie, style in Camunian rock art. Capo di (2002) Representional capacities Ponte, Edizioni del Centro. for pretense with scale modles and – (1977) Methods of recording and photographs in chimpanzees, in Animals analysing rock engravings. Capo di and children, Mitchell, Robert W, Ponte, Edizioni del Centro. (ed), 210-228. Cambridge: Cambridge – (1991) Archetypes, constants and universal paradigms in prehistoric art. University Press. Brentano, Franz (1885) Psychologie von Mimeographed text circulated at the empirischen Standpunkt. Leipzig: Yinchuan conference, October 1991. Mener. Arbib, M. (1979) Minds and millennia: Brier, Søren, (2001) Cybersemiotics and The psychology of interstellar Umweltslehre. On the cybersemiotic communication. In Cosmic Search, 1, 3. integration of ‘Umweltlehre’, enthology, In http://web.archive.org/web/radobs/ autopoiesis theory, second order vol1no3/minds.htm Arnheim, Rudolf, (1966) Towards a cybernetics and Peircean biosemiotics. psychology of art. London: Faber & In Semiotica. Special issue on Jakob Faber. von Uexküll, 134-1/4, 779-814. Bühler, Karl, (1934) Sprachtheorie. – (1969). Visual thinking. Berkeley & Fischer, Stuttgart. Pocket edition: Los Angeles, University of California Ullstein, Frankfurt/M. 1978. Press. Cabe, Patrick, (1980) Picture perception in Bachelard, Gaston (1949) La psychoanalyse nonhuman subjects, in The perception du feu. Paris: Gallimard (pocket edition, of pictures, Volume Il: Dürer’s devices. 1971). Hagen, Margaret, (ed.), 305-343. New Bahn, Paul G. (1998) Prehistoric art. York: Academic Press. Cambridge: Cambridge University Callaghan, Tara, & Rankin, Mary Press. (2002) Emergence of graphic symbol Bartlett, F.C., (1932) Remembering. A study in experimental and social functioning and the question of domain specificity. InChild Development, 73-2, psychology. Cambridge University 359-376. Press, Cambridge. Reprint 1967. Cassirer, Ernst, (1942) Zur Logik der — (1958) Thinking. An experimental Kulturwissenschaften. Göteborg: and social study. Allen & Unwin Ltd., Elanders. London. – (1945) An essay on man. New Haven: Bednarik, Robert G. (1991): The direct Yale University Press. dating of rock art. Mimeographed text _ (1957) The Philosophy of symbolic forms. reproduced from the program of the III. The phenomenology of knowledge. Yinchuan conference, October 1991. Translated by Ralph Mannheim. New Bentele, Günter (1984) Zeichen und

120 Haven: Yale University Press. evolution, 111-139. Oxford: Oxford – (1995) Nachgelassene Manuskripte University Press. und Texte Band 1: Zur Metaphysik der Deely, John, (1982) Introducing Semiotic. symbolischen Formen. Krois, J.M., eds. Its history and doctrine. Bloomington: Hamburg: Felix Meiner Verlag. Indiana University Press. Chafe, Wallace (1977) The recall and – (1994) New beginnings. Early modern verbalization of past experience, in philosophy and postmodern thought. Current issues in linguistic theory, Toronto: University of Toronto Press. Cole, R.W., ed., 215-246. Bloomington Degerando, J. M., (1800) Des signes et de & London: Indiana University Press. l’art de penser considérés dans leurs Chesney, Shirley (1991): Max Raphael’s rapports mutuels. Goujon fils, Paris contribution to the study of prehistoric 1800 (An Vl11), volumes I-IV. symbol systems, in Rock art and – (2001) Four ages of understanding. prehistory, P. Bahn, & A. Rosenfeld, Toronto: University of Toronto Press. (eds.), 14-22. Oxford: Oxbow DeLoache, Judy, (2000) Dual representation Monographs. and young children’s use of scale Clark, Andy (1997) Being there. Putting models. In Child Development, 71, 2, brain, body, and world together again. 329-338. Cambridge, Mass., The MIT Press. – (2004) Becoming symbol-minded. In – (2003) Natural-born Cyborgs. Oxford: Trends in cognitive sciences, 8, 2, 66- Oxford University Press. 70. Clark, Herbert (1996) Using language. – & Burns, Nancy (1994) Early Cambridge: Cambridge University understanding of the representational Press. function of pictures. In Cognition 52, Clayton, N. S. & Dickinson, A. (1998). 83-110. Episodic-like memory during cache – & Miller, Kevin, & Rosengren, Karl recovery by scrub-jays. Nature, 395, (1997) The credible shrinking room. In 272-274. Psychological Science,8, 4, 308-313. Dascal, Marcelo (1978) La sémiologie de Deregowski, Jan, (1970) A note on Leibniz. Paris: Aubier Montaigne. the possible determinant of ‘split – (1983) Signs and cognitive processes: representation’ as an artistic style”, in notes for a chapter in the history of International Journal of Psychology, semiotics. In History of Semiotics. A. 5:1, 21-26. Eschbach and J. Trabant (eds.), 169- – (1972) Pictorial representation and 190. Amsterdam/Philadelphia: John culture, in Scientific American, 225:5, Benjamins. 82-88. – (1998) Hobbes’s challenge. In Descartes, – (1973) Illusion and culture, in Illusion Leibniz e a Modernidade. L. R. dos in Nature and Art. Gregory, R. L., Santos, P. M. S. Alves and A. Cardoso & Gombrich, E. H., (eds.), 160-191. (eds.), 369-398. Lisboa: Editora Colibri London Duckworth & Co, Ltd. Third impression 1980. (Quoted from English web version: http://www.tau.ac.il/humanities/ – (1976) On seeing a picture for the first philos/dascal/papers/hobbes- time, in Leonardo 9:1, 19-23. challenge.htm – (1977) /Answer to Kennedy’s comments ) Deacon, Terry (1997) The Symbolic on Deregowski 1976/, in Leonardo, Species: The Co-Evolution of Language 10:1, 86. and the Brain. New York: Norton. _ (1984) Distortion in art. The eye and – (2003) Universal grammar and semiotic the mind. London: Routledge & Kegan constraints. In Christiansen, Morton Paul. H, & Simon Kirby (eds) ., Language Donald, Merlin (1991) Origins of the

121 Modern Mind. Three Stages in the Frazer, James (1922) The Golden Evolution of Culture and Cognition, Bough. Fascimile reproduction 1993: Cambridge, Mass.: Harvard University Hertfordshire: Wordsworth. Press. Gallagher, Shaun (2005) How the body – (2001) A Mind So Rare. The Evolution shapes the mind. Oxford: Clarendon of Human Consciousness. New York: Press. Norton. Gardner, Howard, (1980) Artful scribbles. Eco, Umberto, (1984) Semiotics and the New York Basic Books. philosophy of language. Bloomington: – (1982) Art, mind, and brain. New York Indiana University Press. Basic Books. – (1998) Réflexions à propos du débat sur – & Wolf, Dennie, (1983) Waves and l’iconisme (1968-1998) . In Visio, 3, 1, streams of symbolization: notes on the 9-32. development of symbolic capacities in – (1999) Kant and the platypus. New York: young children, in Sloboda, John, & Harcourt Brance & Co. Rogers, Don, eds., The Acquisition of Edelmann, Gerald (1992:) Bright Air, Symbolic Skills ; pp. 19-42. brilliant fire. On the matter of the mind. Gibson , James (1966) The senses London: Penguin Books. considered as perceptual systems. –, & Tonini, Guilio, (2000:) A universe Boston: Houghton Mifflin Co. of consciousness. New York: Basic – (1971) The information available in Books. pictures, Leonardo, 4:1: 27-35. Elkins, James, (1996) The Impossibility of – (1978) The ecological approach to visual close reading. In Current Anthropology perception of pictures, Leonardo, 11:3, 37, 2 185-226. 227-235. — (1997) Our beautiful, dry, and distant – (1980) A prefatory essay on the perception texts. Art history as writing. University of surfaces versus the perception of Park: Pennsylvania State University. markings on a surface, in The perception Emmeche, Claus (2001) Does a robot have of pictures, Volume I: Alberti’s window. an Umwelt? In Semiotica 134-1/4, 653- Hagen, Margaret, (ed.), xi-xvii. New 693. York: Academic Press. – (2002) The chicken and the Orphean – (1982) Reasons for realism. Reed, Edward, egg. On the function of meaning and & Jones, Rebecca, (eds) . Hillsdale New the meaning of function. In Sign System Jersey: Lawrence Earlbaum Ass. Studies 30, 1, 15-32. Gombrich, E. H., (1960) Art and Illusion. – & Hoffmeyer, Jesper (1991) From London: Phaidon Press. Second edition language to nature – the semiotic 1962. metaphor in biology. In Semiotica 84- – (1963) Meditations on a Hobby Horse. 1/2, 1-42. London: Phaidon Press. Fagot, Joel, ed., (2000) Picture perception Gomez de Liano, Ignacio, (1982) El idioma in animals. Hove: Psychology Press. de la imaginación. Madrid: Taurus. Fauconnier, Gilles, (1994) Mental spaces. Goodman, Nelson (1968) Languages of New York : Cambridge University art. London: Oxford University Press. Press. Second edition. Goody, Jack, (1977) The domestication ot Fauconnier, Gilles, & Sweetser, Eve (eds), the savage mind. Cambridge: Cambridge (1996) Spaces, Worlds, and Grammar. University Press. Chicago & London : The University of – (1982) Cooking, cuisine, and class. Chicago Press. Cambridge: Cambridge University Foley, Robert A. (1991). The silence of the Press. past, in Nature, vol 353. September 12, Greimas, A.J. (1970) Du sens. Paris: 114-115. Seuil.

122 Grice, Paul (1989) Studies in the way of Translated by Aubrey de Selincourt. words. Cambridge: Harvard University Harmondsworth: Penguin. Press. Hjelmslev, Louis (1937) La catégorie des Gurwitsch, Aron (1957:) Théorie du champ cas. Copenhague: Acta Jutlandica. de la conscience. Bruges: Desclée de Hochberg, Julian, (1972) The representation Brouver. of things and people, in Art, Perceptlon, – (1964) The field of consciousness. and Reality, Gombrich, E. H., Hochberg, Pittsburgh: Duquesne University Press J., & Black, M.(eds,) 47-94. Baltimore – (1966) Studies in Phenomenology and and London: John Hopkins University Psychology. Evanston: Northwestern Press. University Press. – (1978a) Perception. Second, revised – (1974a) Phenomenology and the theory edition. Englewood Cliffs: Prentice- of science. Evanston: Northwestern Hall, Inc. University Press. – (1978b) Art and perception, in Handbook – (1974b) On thematization, in Research in of Perception, vol. X; Carterene, E., & phenomenology, IV, 85-50. Friedman, M., (eds.), 225-258. N. Y. – (1979) Human encounters in the social and London: Academic Press. world. Pittsburgh: Duquesne University – & Brooks, Virginia (1962) Picture Press perception as un unlearned ability. In – (1985) Marginal consciousness. Athens, American Journal of Psychology, 76, Ohio: Ohio University Press. 624-628. Gusdorf, Georges (1966), Les sciences Hockett, Charles, (1977) The view from humains et la pensée occidentale. XIII language. Athens: Univ. of Georgia volumes (or more), Payot, Paris 1966. Press. Hagen, Margaret, (1979) A new theory of Hoffman, Donald, (1998) Visual the psychology of representational art, intelligence. how we create what we in Nodine, C., & Fisher, D., eds., 1979; see. New York: Norton. pp 196-212. – (2004) Images as interstellar messages. – (1980a) ed., The perception of pictures. In Between worlds: the art and science Volume I: Alberti’s window. Volume II: of interstellar message composition. Dürer’s devices. New York: Academic Vakoch, D., ed. Harvard: MIT Press. Press. Hornborg, Alf (1999) Money and the – (1980b) Generative theory: a perceptual semiotics of ecosystem dissolution. In theory of pictorial representation, in Journal of material culture, 4, 2, 143- The perception of pictures. Volume II: 162. Dürer’s devices. Hagen, M., ed., ed., pp – (2001a) Symbolic technologies. 3-46. New York: Academic Press. Marchines and the Marxian notion of Hägerstrand, Torsten (1983) In search fetischism, in Anthropological theory, for the sources of concepts, in The 1, 4, 473-496. practice of geography, Buttimer, Anne, - (2001b) Vital signs: An ecosemiotic (ed.), 238-256. London and New York: perspective on the human ecology of Longman. Amazonia: in Sign Systems Studies, Halbwachs, Maurice (1925) Les cadres 29:1, 121-152. sociaux de la mémoire. Paris: PUF. Husserl, Edmund (1913) Logische Second edition 1952. Untersuchungen. Tübingen: Niemeyer. – (1950) La mémoire collective. Paris: PUF. Fünfte Auflage 1968. Second revised edition 1968. – (1939) Erfahrung und Urteil. Prag: Hermerén, Göran (1983) Aspects of Academia Verlagsbuchhandlung. aesthetics. Lund: Gleerups. – (1962a) Die Krisis der europäischen Herodotus (1954) The Histories. Wissenschaften und die transzendentale

123 Phänomenologie. Husserliana VI. The Body, language, and mind. Zlatev, J., Hague: Nijhoff. Ziemke, T., & Frank, R., eds. Berlin & – (1962b) Phänomenologische Psychologie. New York: Mouton de Gruyter Husserliana IX. The Hague: Nijhoff. Jones, Rebecca, & Hagen, Margaret, – (1973) Ding und Raum. Husserliana (1980) A perspective on cross-cultural XVI. The Hague: Nijhoff. picture perception, in The perception – (1980) Phantasie, Bildbewusstsein, of pictures, Volume Il: Dürer’s devices. Erinnerung. Husserliana XXIII. The Hagen, Margaret, (ed.), 193-226. New Hague: Nijhoff. York: Academic Press. Innes, Harold, (1950) Empire and Kennedy, John M., (1974a) A psychology communication. Toronto: University of ot picture perception. San Francisco: Toronto Press, 1950. 2nd ed. 1972. JosseyBas, Inc. Itakura, S. (1994) Recognition of line- – (1974b) Icons and information, in Media drawing representations by chimpanzees and symbols, Olson, David, (ed.), 211- (Pan troglodytes). In Journal of General 240. Chicago: National society for the Psychology, 121, 189-197. study of education. Ivins, William M., (1953) Prints and visual Kintsch, Walter (1974) The representation communication. Cambridge, Mass., of meaning in memory. Hillsdale: Harvard University Press. Erlbaum Ass. Jakobson, Roman, (1975) Coup d’œil – (1977) Memory and cognition. New sur le développment de la sémiotique. York: Wiley. Bloomington: Indiana University Krampen, Martin (1991) Children’s Press. Drawings. Iconic coding of the — (1990) On language. Waugh, Linda, environment. New York: Plenum Press. & Monville-Borston, Monique (eds.). Kuhlmeier, Valerie, & Boysen, Sarah, (2001) Cambridge., Mass.: Harvard University The effect of response contingencies Press. on sclae model task performance by Janet, Pierre, (1928) L’évolution de la chimanzee, in Journal of comparative mémoire et de la notion du temps. Paris: psychology, 115, 3, 300-306 Chahine. – (2002) Chimpanzee recocognize spatial Jappy, Tony (2000): “Iconicity, and object crrespondeces between a scale Hypoiconicity”. In The Digital model and its referent, in Psychological Encyclopaedia of Charles S. Peirce, Science, 13, 1, 60-63. Quiroz, João, ed. http://www. – & Mukobi, Kimberly (1999) Scale- digitalpeirce.org/jappy/hypjap.htm model comprehension by Chimpanzee, Johansen, Jørgen Dines, (1993) Dialogic in Journal of comparative psychology, semiosis. Bloomington: Indiana 113, 4, 396-402. University Press. Küng, Guido (1973) Husserl on pictures Johnson, Mark (1987) The Body in the and intentional objects. In Review of Mind. Chicago & London: University metaphysics. June 1973, XXXVI, 4, of Chicago Press. 670-680. _ (2005) The philosophical significance of Lakoff, George (1987) Women, fire, and image schemas. In: From Perception to dangerous things. Chicago: University Meaning: Image Schemas in Cognitive of Chicago Press. Linguistics, Beate Hampe (ed.), Berlin: – & Johnson, Mark (1999) Philosophy in Mouton de Gruyter. the Flesh. New York: Basic Books. Johnson, Mark, and Tim Rohrer, (in press) Lévi-Strauss, Claude (1958) Anthropologie We are life creatures: Embodiment, structurale /I/. Paris: Plon. American Pragmatism, and the Liungman, Carl. G., (1991) Dictionary of cognitive organism. To appear in In symbols. Santa Barbara, California: 124 ABC Clio. Hjelmquist, & L.G. Nilsson, (eds.), Li Fushun (1992): A survey of the 233-252. Amsterdam, North Holland. methods of dating rock art in China. In Neisser, Ulric (1976) Cognition and reality. International Newsletter on Rock Art, San Francisc: Freeman & Co. no 2; 15-19. Nordbladh, Jarl (1973) Om bilderna som Lotman, J.M., (1976) Culture and meddelanden i samhället, in Bild & information, in Dispositio. Revista miljö. Konstvetenskapliga studier, 71- hispánica de semiótica literaria, 3, 1, 85. Gothenburg. 213-215. – (1977) Images as messages in — (1979) Culture as collective intellect society. Prolegomena to the study of and problems of artificial intelligence, Scandinavian Petroglyphs and Semiotics, in Russian Poetics in translation, 6, in New Directions in Scandinavian University of Essex, 1979; 84-96 Archaeology, Kristainsen, Kristian, & — & Uspenskij, B.A., Ivanov, V.V., Paludan-Müller, Carsten, (eds.), 63-78. Toporov, V.N., & Pjatigorskij, A.M. Copenhagen: The National Museum of (1975:) Thesis on the semiotic study Denmark. of culture. Lisse: The Peter de Ridder – (1978) Some problems concerning the Press. relation between rock art, religion and Luckman, Thomas (1980) Lebenswelt und society, in Acts of the International Geschichte. Paderborn: Schöning. Symposium on Rock art, Hankø 6-12 Mallery, Garrick, (1880-81): Introduction August 1972, 185-210. Oslo: Instituttet to the study of sign language among for Sammenlignende Kulturforskning. the North American Indians and A – (1980) Glyfer och rum. Gothenburg: collection of gesture signs. In Aboriginal Institute of Archeology. sign languages of the Americas and Parmentier, Richard J. (1985) Signs’s Australia I., Umiker-Sebeok, J., & place in medias res: Peirce’s concept Sebeok, Th., (eds.), 1-76; 77-406. New of semiotic mediation. In Mertz, York & London: Plenum Press 1978. Elizabeth, & Parmentier, Richard J. – (1881). Sign language among North (eds.) Semiotic mediation. Sociocultural American Indians. Reprinted: The and psychological perspectives; 23- 48. Hague & Paris: Mouton 1972. Orlando: Academic Press. Mandler, Jean (2004) The foundations Peirce, Charles Sanders, (1931-58) of mind. Oxford: Oxford University Collected Papers I-VIII. Hartshorne, C, Press. Weiss, P, & Burks, A, (eds.) . Cambridge, Manetti, G. (1993) Theories of the Sign in MA: Harvard University Press. Classical Antiquity. Bloomington, In. ; – (1998) The Essential Peirce, volume Indiana University Press. II. Ed. By the Peirce Edition Project. Masuda, Y. (1980) Information society as Bloomington and Indianapolis: Indiana post-industrial society. Washington, University Press. D.C., World future society. Piaget, Jean (1945) La formation du symbole Metz, Christian, (1990) Photograph and chez l’enfant. Neuchatel: Delachaux & fetish, in Sqioers. Carol, ed., The Niestlé. Third edition 1967. critical image. Essays on contemporary – (1967a) La psychologie de l’intelligence. photography, 155-164. Seattle: Bay Paris: Armand Colin. Press. – (1967b) Biologie et connaissance. Paris: Mithen, Steven (1996) The Prehistory of Gallimard. Pocket edition 1973 the Mind. London: Thames & Hudson. – (1970) Epistémologie des sciences de Muter, Paul (1986). Blissymbolics, l’homme. Paris: Gallimard. cognition and the handicapped. In Pick, Anne, & Pick, Herbert, (1978) Communication and Handicap, E. Culture and perception, in Handbook

125 of Perception, vol. X; Carterene, E., & Searle, John (1969) Speech Acts. Friedman, M., (eds.), 19-39. N. Y. and Cambridge: Cambridge University London: Academic Press. Press. Pincavet, Fr., (1891) Les Idéologues. – (1995) The construction of social reality. Essai sur l’histoire des idées et des New York: The Free Press. theories scientifiques, philosophiques, Singer, Milton (1984) Man’s glassy essence. religieuses, etc., en France depuis 1789. Bloominington, IN., Indiana University Paris: Félix Alcan. Press. Ponzio, Agosto (1993) Signs, dialogue, and Smith, Barry, & Varzi, Achille (1999) The ideology. Amsterdam: Benjamins. niche, in Noûs, 33:2, 198-222. Popper, Karl, (1972) Objective knowledge. Sonesson, Göran, (1978) Tecken och Oxford : Clarendon. handling. Lund. Doxa. Posner, Roland, (1989) What is culture? – (1987) Bildbetydelser i Toward a semiotic explication of informationssamhället. Lund: anthropological concepts, in The Nature Institutionen för konstvetenskap. of Culture. Koch, Walter, ed., 240-295. — (1988) Methods and models in pictorial Bochum. Brochmeyer. semiotics. Semiotics project: Lund Prieto, Luis (1966). Messages et signaux. University. Paris: PUF. – (1989a) Pictorial concepts. Inquiries into Reed, Edward S. (1996) The necessity of the semiotic heritage and its relevance experience. New Haven: Yale University for the analysis of the visual world. Press. Lund: Aris/Lund University Press. Renfrew, Colin (1982) Toward an – (1989b) Semiotics of photography. On archaeology of mind. Cambridge: tracing the index. Semiotics project. Cambridge University Press. Lund: University. Renfrew, Colin, & Bahn, Paul, (1991). – (1992a) Bildbetydelser. Inledning till Archaeology. Theories, methods, and bildsemiotiken som vetenskap. Lund: practice. London: Thames and Hudson Studentlitteratur. Ltd. – (1992b) The semiotic function and Rossi-Landi, Ferruccio (1983) Language as the genesis of pictorial meaning. In work & trade. A semiotic homology for Tarasti, Eero (ed.), Center/Periphery linguistics & economics. Massachusetts: in representations and institutions. Bergin & Garvey. Translation of Italian Proceedings from the Conference of original 1968. The International Semiotics Institute, Rumelhart, David, & Norman, D. (1978) Imatra, Finland, July 16-21, 1990., 211- Accretion, tuning, and restructuring, 156. Imatra: Acta Semiotica Fennica. in Semantic factors in cognition, – (1992c) Comment le sens vient aux Cotton, J., & Klatzky, L., eds . 37-54. images. Un autre discours de la méthode. Hillsdale: Erlbaum Ass. In Carani, Marie (ed.) , De l’histoire Saussure, F. (1973). Cours de linguistique de l’art à la sémiotique visuel, 29-84. générale. Edition critique par Ullio de Québéc: Les éditions du Septentrion/ Mauro. Paris: Payot. CÉLAT. Schütz, Alfred, (1932) Der sinnhafte – (1993a) Pictorial semiotics, Gestalt Aufbau der sozialen Welt. Springer, psychology, and the ecology of Vienna 1932. New edition Suhrkamp, perception. In Semiotica 99: 3/4, 319- Frankfurt/M.1974. 399. – (1967) Collected papers I: The problem – (1993b) Die Semiotik des Bildes, of social reality. Nijhoff, The Hague. Zeitschrift für Semiotik 15: 1-2, 131- Scruton, Roger (1974) Art and imagination. 164. London: Meuthen & Co. – (1993c) The multiple bodies of man.

126 Project for a semiotics of the body. in Pätli, Ü., eds., 88-127. Tartu, Tartu Degré, 74, d-d42. University Press. – (1994a) Prolegomena to the semiotic – (2000a) Iconicity in the ecology of analysis of prehistoric visual displays. semiosis, In Troels Deng Johansson, In Semiotica, 100: 3/, 267-332. 47-64. - Martin Skov & Berit Brogaard (eds.), – (1994b) Sémiotique visuelle et écologie Iconicity - A Fundamental Problem in sémiotique, in RSSI, 14, 1-2, printemps Semiotic, 59-80. Aarhus: NSU Press. 1994, pp. 31-48. 47-64. – (2000b) Ego meets Alter. In Semiotica – (1995) On pictorality. The impact of the 128-3/4, 2000: Essays in honour of perceptual model in the development Vilmos Voigt, Bernard, Jeff (ed.), 537- of visual semiotics. In Sebeok, Th., & 559. Umiker-Sebeok, J. (eds) , The semiotic – (2001a:) From Semiosis to Ecology. web 1992/93: Advances in visual On the theory of iconicity and its semiotics, 67-108. Mouton de Gruyter, consequences for the ontology of the Berlin & New York. Lifeworld. In Quinn, Andrew W., (ed.), – (1996a) An essay concerning images. Cultural cognition and space Cognition/ From rhetoric to semiotics by way of Cognition culturelle et cognition ecological physics. In Semiotica, 109- spatiale. VISIO, 6: 2-3: 85-110. 1/2, 41-140. – (2001b) De l’iconicité de l’image à — (1996b) Commentary to Elkins, James, l’iconicité des gestes, Actes du congrès The Impossibility of close reading. In ORAGE 2001 ORAlité et Gestualité, Current Anthropology 37, 2, 215-216. Aix-en-Provence, Juin 18-22, 2001. – (1997) The ecological foundations of Paris : L’Harmattan. iconicity, in Rauch, Irmgard, & Carr, – (2003a) The Symbolic Species revisited. Gerald F., (eds.) , Semiotics Around Considerations on the semiotic turn the World: Synthesis in Diversity. in cognitive science and biology Proceedings of the Fifth International SGBWP3. Working Papers of the Congress of the IASS, Berkeley, June Project “Language, gesture, and pictures 12-18, 1994. 739-742. Berlin & New from the point of view of semiotic York: Mouton de Gruyter. development”, Lund University, Lund. – (1998a) /entries/, in Encyclopaedia of _ (2003b) Why the mirror is a sign – and Semiotics. Bouissac, Paul, ed., New why the television picture is no mirror. York & London: Oxford University Two episodes in the critique of the Press. iconicity critique. In Proceedings of – (1998b) That there are many kinds of the 5th International Interdisciplinary pictorial signs, Visio, 3, 1, 33-54. Symposium of the Austrian Society for — (1998c) On the notion of text in cultural Semiotic Studies “Picture Language semiotics, in Σηµειωτικ′η. Trudy po - Visualization – Diagrammatics: znakovym sistemam Sonesson, Göran Iconicity/Das 5. Wiener Symposium Sign System Studies 26, Torop, Peeter, ‘Bildsprache - Visualisierung - Lotman, Michaeil, & Kull, Kalevi, Diagrammati”: Ikonizität , November (eds.), 83-114. Tartu, Tartu University 2002, published as S. European Journal Press for semiotic studies, volume 15:2-4, – (1999a) Post-photography and beyond. 2003, pp. 217-232. From mechanical reproduction to digital – (Forthcoming a) From Iconicity to production. In Visio 4, 1, 11-36. Pictorality. A view from ecological – (1999b) The life of signs in society semiotics. To appear in VISIO: thematic – and out of it. In Semiotiké. Trudy po issue: Iconicity revisited/Iconicité znakovym sistemam/Sign system studies revisité, Sonesson, Göran (ed) . 26, Torop, P., Lotman, M., Kull, K., – (Forthcoming b), The signs of the body and the body of signs. In Body, Eleanor Rosch (1991) The embodied language, and mind. Zlatev, J., Ziemke, mind. Cognitive science and human T., & Frank, R., eds. Berlin & New experience. Cambridge, Mass.: MIT York: Mouton de Gruyter Press. Sperber, Dan (1982) Le savoir des Vygotsky, Lev (199) Tänkande och språk. anthropologues. Paris: Harmann. Göteborg: Daidalos. — Wilson, Deirdre (1986) Relevance. von Däniken, Erik (1973) Meine Welt in Second, revised editon 1995. Bildern. Düsseldorf & Wien: Econ Stern, William (1914) Psychologie der Verlag. frühen Kindheit. Tenth, unmodified Wertsch, James (2002) Voices of collective edition: Heidelberg: Quelle & Meyer remembering. Cambridge: Cambridge 1971. University Press, Tilley, Christopher, (1991) Material culture White, Randall (2000) Prehistoric art. New and text. The art of ambiguity. London York: Harry N. Abrahams, Inc. & New York: Routledge. Winner, Ellen (1982) Invented Worlds. Trevarthen, Colwyn, & Logotheti, Katerina, Cambridge: Harward University Press (1989) Child and culture: Genesis of Yates, Frances (1966) The art of memory. co-operative knowing, in Cognition and London: Routlege & Kegan Paul. social worlds. Gellatly, Angus, Rogers, Ziemke, Tom , & Sharkey, Noel E. (2001) Don, & Sloboda, John, eds., 37-56. A stroll through the worlds of robots and Oxford: Clarendon Press. animals: Applying Jakob von Uexküll’s Tomasello, Michael (1999) The Cultural theory of meaning to adaptive robots Origins of Human Cognition. and artificial life. Semiotica, 134(1-4) , Cambridge, MA: Harvard University 701-746. Press. Zlatev, Jordan (2002) Mimesis: The —, Carpenter, Malinda, Call, Joseph, “missing link” between signals and Behne, Tanya, & Moll, Henrike (2005) symbols in phylogeny and ontogeny. Understanding and sharing intentions: In Mimesis, Sign and the Evolution of the origins of cultural cognition. In Language, A. Pajunen (ed.), 93-122. Behavioral and Brain Sciences 28, 675- Turku: University of Turku Press. 735. – (2003) Meaning = Life (+ Culture). An Vakoch, D. (1999) The view from a distant outline of a unified biocultural theory of star. In Mercury, March/April, 26-39, meaning. Evolution of communication quoted from http://www.astrosociety. 4:2, 253-296. org/pubs/mercury/9902/vakoch.html _ (2005) What’s in a schema? Bodily – (2001) Messages from ET. Published on mimesis and the grounding of language. http://www.space.com/searchforlife/ In: From Perception to Meaning: Image vakoch_messages_010706.html Schemas in Cognitive Linguistics, – (2003) Messages from the moon, to Mars, Beate Hampe (ed.). Berlin: Mouton de to the stars. Published on http://seti.org/ Gruyter. seti/our_projects/interstellar_messages/ – (in press a) On intersubjectivity and overview/history/messages.html mimetic schemas. To appear in van Dijk, Teun, (1980) Macrostructures. Intellectica, thematic issue on culture Hillsdale: Erlbaum Ass. and cognition. – & Kintsch, W. (1978) Cognitive – (in press b) Intersubjectivity, language psychoiogy and discourse, in Current and the mimesis hierarchy. To appear trends in textlinguistics, Dressler, W., in The Shared Mind, Zlatev, Jordan, ed., 61-80. Berlin & New York: de Racine, Timothy, Sinha, Chris, & Gruyter. Itkonen, Esa, eds. Varela, Francisco, Thompson, Evan and – Persson, Thomas, & Gärdenfors, Peter, (2005) Bodily mimesis as “the missing link” in human cognitive evolution. Lund: Lund University Cogntive Studies 121.