Semiotics for Beginners: Modality and Representation

Daniel Chandler

Modality and Representation

Whilst semiotics is often encountered in the form of textual analysis, it also involves philosophical theorising on the role of signs in the construction of reality. Semiotics involves studying representations and the processes involved in representational practices, and to semioticians, 'reality' always involves representation.

To semioticians, a defining feature of signs is that they are treated by their users as 'standing for' or representing other things. Jonathan Swift's satirical account of the fictional academicians of Lagago highlights the inadequacies of the commonsense notion that signs stand directly for physical things in the world around us.

A Scheme for entirely abolishing all Words whatsoever... was urged as a great Advantage in Point of Health as well as Brevity. For it is plain, that every Word we speak is in some Degree a Diminution of our Lungs by Corrosion, and consequently contributes to the shortning of our Lives. An Expedient was therefore offered, that since Words are only Names for Things, it would be more convenient for all Men to carry about them, such Things as were necessary to express the particular Business they are to discourse on. And this Invention would certainly have taken Place, to the great Ease as well as Health of the Subject, if the Women in conjunction with the Vulgar and Illiterate had not threatned to raise a Rebellion, unless they might be allowed the Liberty to speak with their Tongues, after the manner of their Ancestors; such constant irreconcilable Enemies to Science are the common People. However, many of the most Learned and Wise adhere to the New Scheme of expressing themselves by Things, which hath only this Inconvenience attending it, that if a Man's Business be very great, and of various kinds, he must be obliged in Proportion to carry a greater bundle of Things upon his Back, unless he can afford one or two strong Servants to attend him. I have often beheld two of those Sages almost sinking under the Weight of their Packs, like Pedlars among us; who, when they met in the Streets, would lay down their Loads, open their Sacks, and hold Conversation for an Hour together; then put up their Implements, help each other to resume their Burthens, and take their Leave.

But for short Conversations a Man may carry Implements in his Pockets and under his Arms, enough to supply him, and in his House he cannot be at a loss: Therefore the Room where Company meet who practise this Art, is full of all Things ready at Hand, requisite to furnish Matter for this kind of artificial Converse. Another great Advantage proposed by this Invention, was that it would serve as a Universal Language to be understood in all civilized Nations, whose Goods and Utensils are generally of the same kind, or nearly resembling, so that their Uses might easily be comprehended. And thus Embassadors would be qualified to treat with foreign Princes or Ministers of State to whose Tongues they were utter Strangers.

(Jonathan Swift [1726/1735]: Gulliver's Travels, Part III, 'A Voyage to Laputa, Balnibarbi, Luggnagg, Glubbdubdrib and Japan', Chapter V)

The proposal by the academicians of Lagago to substitute objects for words highlights problems with the simplistic notion of signs being direct substitutes for things. The academicians adopted the philosophical stance of naive realism in assuming that words simply mirror objects in an external world. They believed that 'Words are only Names for Things', a stance involving the assumption that 'things' necessarily exist independently of language prior to them being 'labelled' with words. According to this position (which accords with a still widespread popular misconception of language) there is a one-to-one correspondence between word and referent (sometimes called language-world isomorphism), and language is simply a nomenclature - an item-by-item naming of things in the world. As Saussure put it, this is 'the superficial view taken by the general public' (Saussure 1983, 16, 65; Saussure 1974, 16, 65).

Within the lexicon of a language, it is true that most of the words are 'lexical words' (or nouns) which refer to 'things', but most of these things are abstract concepts rather than physical objects in the world. Only 'proper nouns' have specific referents in the everyday world, and only some of these refer to a unique entity (e.g. Llanfairpwllgwyngyllgogerychwyrndrobwllllantysiliogogogoch - the name of a Welsh village). As Rick Altman notes, 'A language made up entirely of proper nouns, like the one used in horse racing forms, offers significant representational benefits; every name clearly corresponds to and identifies a single horse' (Altman 1999, 87). However, the communicative function of a fully-functioning language requires the scope of reference to move beyond the particularity of the individual instance. Whilst each leaf, cloud or smile is different from all others, effective communication requires general categories or 'universals'. Anyone who has attempted to communicate with people who do not share their language will be familiar with the limitations of simply pointing to things. You can't point to 'mind', 'culture' or 'history'; these are not 'things' at all. The vast majority of lexical words in a language exist on a high level of abstraction and refer to classes of things (such as 'buildings') or to concepts (such as 'construction'). Language depends on categorization, but as soon as we group instances into classes (tokens into types), we lose any one-to-one correspondence of word and thing (if by 'things' we mean specific objects). Furthermore, other than lexical words, the remaining elements of the lexicon of a language consist of 'function words' (or grammatical words, such as 'only' and 'under') which do not refer to objects in the world at all. The lexicon of a language consists of many kinds of signs other than simply nouns. Clearly, language cannot be reduced to the naming of things.

The less naive realists might note at this point that words do not necessarily name only physical things which exist in an objective material world but may also label imaginary things and also concepts. Peirce's referent, for instance, is not limited to things which exist in the physical world but may include non-existent objects and ideas. However, as Saussure noted, the notion of words as labels for concepts 'assumes that ideas exist independently of words' (Saussure 1983, 65; Saussure 1974, 65). and for him, 'no ideas are established in advance... before the introduction of linguistic structure' (Saussure 1983, 110; cf. 114-115, 118; Saussure 1974, 112; cf. 116, 120). It remains a rationalist and 'nomenclaturist' stance on language when words are seen as 'labels' for pre-existing ideas as well as for objects. It is reductionist: reducing language to the purely referential function of naming things. When we use language, its various kinds of signs relate to each other in complex ways which make nonsense of the reduction of language to a nomenclature. Referentiality may be a function of language but it is only one of its functions. Furthermore, as Vivien Burr puts it, 'whatever the nature of the "real" world, we cannot assume that the words in our language refer to it or describe it' (Burr 1995, 60). The philosophically flawed assumption that it is a necessary condition of a sign that the signifier has a referent has been termed the 'referential illusion' by Roland Barthes (Barthes 1957) and the 'referential fallacy' by Michael Riffaterre (Genosko 1994, 38, 51; Allen 2000, 115).

A radical response to realists is that things do not exist independently of the sign systems which we use; 'reality' is created by the media which seem simply to represent it. Language does not simply name pre-existing categories; categories do not exist in 'the world' (where are the boundaries of a cloud; when does a smile begin?). We may acknowledge the cautionary remarks of John Lyons that such an emphasis on reality as invariably perceptually seamless may be an exaggeration. Lyons speculates that 'most of the phenomenal world, as we perceive it, is not an undifferentiated continuum' ; and our referential categories do seem to bear some relationship to certain features which seem to be inherently salient (Lyons 1977, 247; my emphasis; cf. ibid., 260)

Saussure's model of the sign involves no direct reference to reality outside the sign. This was not a 'denial' of extralinguistic reality as such but a reflection of his understanding of his own role as a linguist. Saussure accepted that in most scientific disciplines the 'objects of study' were 'given in advance' and existed independently of the observer's 'point of view'. However, he stressed that in linguistics, by contrast, 'it is the viewpoint adopted which creates the object' (Saussure 1983, 8; Saussure 1974, 8). Whilst such a statement might go without comment in a discipline with an acknowledged self-sufficiency (such as mathematics), in the context of human language one can understand how it might be criticized as an idealist model. In the Saussurean model the signified is only a mental concept; concepts are mental constructs, not 'external' objects. However, as Rodowick notes, 'stressing the relation of difference or nonidentity between an object and the form or substance of its expression need not imply the absence of representation' (Rodowick 1994, 162). A concept may, of course, refer to something in experiential reality but the Saussurean stance is a denial of the 'essentialist' argument that signifieds are distinct, autonomous entities in an objective world which are definable in terms of some kind of unchanging 'essence' (Culler 1985, 24). Saussurean semiotics asserts the non-essential nature of objects. Just like signifiers, signifieds are part of the sign system; signifieds are socially constructed. According to the Whorfian stance, the signified is an arbitrary product of our culture's 'way of seeing'. The Saussurean perspective 'tends to reverse the precedence which a nomenclaturist accords to the world outside language, by proposing that far from the world determining the order of our language, our language determines the order of the world' (Sturrock 1986, 17).

In contrast to the Saussurean model, Peirce's model of the sign explicitly features the referent - something beyond the sign to which the sign vehicle refers (though not necessarily a material thing). However, it also features the interpretant which leads to an 'infinite series' of signs, so at the same time Peirce's model also seems to suggest the relative independence of signs from any referents (Silverman 1983, 15). For Peirce, reality can only be known via signs. If representations are our only access to reality, determining their accuracy is a critical issue. Peirce adopted from logic the notion of 'modality' to refer to the truth value of a sign, acknowledging three kinds: actuality, (logical) necessity and (hypothetical) possibility (Hodge & Kress 1988, 26). Furthermore, his classification of signs in terms of the mode of relationship of the sign vehicle to its referent reflects their modality - their apparent transparency in relation to 'reality' (the symbolic mode, for instance, having low modality). Peirce asserted that, logically, signification could only ever offer a partial truth because it if offered the complete truth it would destroy itself by becoming identical with its object (cited in Grayson 1998, 40).

Theorists who veer towards the extreme position of philosophical idealism (for whom reality is purely subjective and is constructed in our use of signs) may see no problem with the Saussurean model. Indeed, the Saussurean model has itself been described as 'idealist' (Culler 1985, 117). Those drawn towards philosophical realism (for whom a single objective reality exists indisputably 'outside' us) would challenge it. According to this stance, reality may be 'distorted' by the media which we use to apprehend it but such media play no part in 'constructing' the world. Even those who adopt an intermediate constructionist (or constructivist) position - that language and other media play a major part in 'the social construction of reality' - may tend to object to an apparent indifference towards social reality in Saussure's model. Those on the political left in particular would object to its sidelining of the importance of the material conditions of existence. Umberto Eco provocatively asserts that 'semiotics is in principle the discipline studying everything which can be used in order to lie' (Eco 1976, 7).

From the perspective of social semiotics the original Saussurean model is understandably problematic. Whatever our philosophical positions, in our daily behaviour we routinely act on the basis that some representations of reality are more reliable than others. And we do so in part with reference to cues within texts which semioticians (following linguists) call 'modality markers'. Such cues refer to what are variously described as the plausibility, reliability, credibility, truth, accuracy or facticity of texts within a given genre as representations of some recognizable reality. Gunther Kress and Theo van Leeuwen acknowledge that

A social semiotic theory of truth cannot claim to establish the absolute truth or untruth of representations. It can only show whether a given 'proposition' (visual, verbal or otherwise) is represented as true or not. From the point of view of social semiotics, truth is a construct of semiosis, and as such the truth of a particular social group, arising from the values and beliefs of that group. (Kress and van Leeuwen 1996, 159).

From such a perspective, reality has authors; thus there are many realities rather than the single reality posited by objectivists. This stance is related to Whorfian framings of relationships between language and reality. Constructionists insist that realities are not limitless and unique to the individual as extreme subjectivists would argue; rather, they are the product of social definitions and as such far from equal in status. Realities are contested, and textual representations are thus 'sites of struggle'.

Modality refers to the reality status accorded to or claimed by a sign, text or genre. More formally, Robert Hodge and Gunther Kress declare that 'modality refers to the status, authority and reliability of a message, to its ontological status, or to its value as truth or fact' (Hodge & Kress 1988, 124). In making sense of a text, its interpreters make 'modality judgements' about it, drawing on their knowledge of the world and of the medium. For instance, they assign it to fact or fiction, actuality or acting, live or recorded, and they assess the possibility or plausibility of the events depicted or the claims made in it.

Clearly, the extent to which a text may be perceived as 'real' depends in part on the medium employed. Writing, for instance, generally has a lower modality than film and television. However, no rigid ranking of media modalities is possible. John Kennedy showed children a simple line drawing featuring a group of children sitting in a circle with a gap in their midst (Kennedy 1974). He asked them to add to this gap a drawing of their own, and when they concentrated on the central region of the drawing, many of them tried to pick up the pencil which was depicted in the top right-hand corner of the drawing! Being absorbed in the task led them to unconsciously accept the terms in which reality was constructed within the medium. This is not likely to be a phenomenon confined to children, since when absorbed in narrative (in many media) we frequently fall into a 'suspension of disbelief' without compromising our ability to distinguish representations from reality. Charles Peirce reflected that 'in contemplating a painting, there is a moment when we lose the consciousness that it is not the thing, the distinction of the real and the copy disappears' (Peirce 1931-58, 3.362).

Whilst in a conscious comparison of a photographic image with a cartoon image of the same thing the photograph is likely to be judged as more 'realistic', the mental schemata involved in visual recognition may be closer to the stereotypical simplicity of cartoon images than to photographs. People can identify an image as a hand when it is drawn as a cartoon more quickly than when they are shown a photograph of a hand (Ryan & Schwartz 1956). This underlines the importance of perceptual codes in constructing reality. Umberto Eco argues that through familiarity an iconic signifier can acquire primacy over its signified. Such a sign becomes conventional 'step by step, the more its addressee becomes acquainted with it. At a certain point the iconic representation, however stylized it may be, appears to be more true than the real experience, and people begin to look at things through the glasses of iconic convention' (Eco 1976, 204-5).

Modality cues within texts include both formal features of the medium and content features such as the following (typical high modality cues are listed here as the first in each pair), though it is their interaction and interpretation, of course, which is most important.

	Formal features 3D-flat detailed-abstract colour-monochrome edited-unedited moving-still audible-silent Content features possible-impossible plausible-implausible familiar-unfamiliar current-distant in time local-distant in space
Cornelius Gijsbrechts (1670): A Cabinet of Curiosities with an Ivory Tankard (showing reverse of cupboard door) Oil on Canvas, Statens Museum for Kunst

The media which are typically judged to be the most 'realistic' are photographic - especially film and television. James Monaco suggests that 'in film, the signifier and the signified are almost identical... The power of language systems is that there is a very great difference between the signifier and the signified; the power of film is that there is not' (Monaco 1981, 127-8). This is an important part of what Christian Metz was referring to when he described the cinematic signifier as 'the imaginary signifier'. In being less reliant than writing on symbolic signs, film, television and photography suggest less of an obvious gap between the signifier and its signified, which make them seem to offer 'reflections of reality' (even in that which is imaginary). But photography does not reproduce its object: it 'abstracts from, and mediates, the actual' (Burgin 1982a, 61). Whilst we do not mistake one for the other we do need to remind ourselves that a photograph or a film does not simply record an event, but is only one of an infinite number of possible representations. All media texts, however 'realistic', are representations rather than simply recordings or reproductions of reality. As the film theorist D N Rodowick puts it, 'Rather than reproducing the "world" spontaneously and automatically, as the ideology of realism would have the spectator believe, the cinematic apparatus always operates selectively, limiting, filtering and transforming the images that are its raw material' (Rodowick 1994, 77).

The film theorist Andr� Bazin refers to the 'reproductive fallacy' as being that the only kind of representation which can show things 'as they really are' is one which is (or appears to be) exactly like that which it represents in every respect. Except in the case of digitally-sourced reproductions, texts are constructed from different materials from that which they represent, and representations cannot be replicas. For Bazin, aesthetic realism depended on a broader 'truth to reality' (Bazin 1974, 64; Lovell 1983, 81).

Modality judgements involve comparisons of textual representations with models drawn from the everyday world and with models based on the genre; they are therefore obviously dependent on relevant experience of both the world and the medium. Robert Hodge and David Tripp's semiotic study of Children and Television focuses on the development of children's modality judgements (Hodge & Tripp 1986).

Ien Ang (1985) argues that watching television soap operas can involve a kind of psychological or emotional realism for viewers which exists at the connotative rather than the denotative level. Viewers find some representations emotionally or psychologically 'true-to-life' (even if at the denotative level the treatment may seem 'unrealistic'). I would argue that especially with long-running soaps (which may become more 'real' to their fans over time) what we could call generic realism is another factor. Viewers familiar with the characters and conventions of a particular soap opera may often judge the programme largely in its own generic terms rather than with reference to some external 'reality'. For instance, is a character's current behaviour consistent with what we have learnt over time about that character? The soap may be accepted to some extent as a world in its own right, in which slightly different rules may sometimes apply. This is of course the basis for what Coleridge called the 'willing suspension of disbelief' on which drama depends.

Robert Hodge and Gunther Kress argue that:

Different genres, whether classified by medium (e.g. comic, cartoon, film, TV, painting) or by content (e.g. Western, Science Fiction, Romance, news) establish sets of modality markers, and an overall value which acts as a baseline for the genre. This baseline can be different for different kinds of viewer/reader, and for different texts or moments within texts. (Hodge & Kress 1988, 142)

What are recognized as 'realistic' styles of representation reflect an aesthetic code. Over time, certain methods of production within a medium and a genre become naturalized. The content comes to be accepted as a 'reflection of reality'. In the case of popular television and film, for instance, the use of 'invisible editing' represents a widespread set of conventions which has come to seem 'natural' to most viewers. In 'realistic' texts what is foregrounded is the 'content' rather than the 'form' or style of production. As in the dominant mode of 'scientific' discourse, the medium and codes are discounted as neutral and transparent and the makers of the text retreat to invisibility. Consequently, 'reality' seems to pre-exist its representation and to 'speak for itself'; what is said thus has the aura of 'truth'. John Tagg argues that

The signifier is treated as if it were identical with a pre-existent signified and... the reader's role is purely that of a consumer... Signifier and signified appear not only to unite, but the signifier seems to become transparent so that the concept seems to present itself, and the arbitrary sign is naturalized by a spurious identity between reference and referents, between the text and the world. (Tagg 1988, 99)

However, Tagg adds that such a stance need not involve positing 'a closed world of codes' (ibid., 101) or the denial of the existence of what is represented outside the process which represents it (ibid., 167). He stresses 'the crucial relation of meaning to questions of practice and power', arguing that 'the Real is a complex of dominant and dominated discourses which given texts exclude, separate or do not signify' (ibid., 101).

The Belgian surrealist Ren� Magritte (1898-1967) painted La Trahison des Images (The Treachery of Images) in 1936. It depicts a side-on view of a smoker's pipe and the text 'Ceci n'est pas une pipe' ('This is not a pipe'). The image provided here is similar but the text is different - this is not a reproduction of Magritte's painting(!). Both examples - Magritte's painting and this version - give us pause for thought. Each 'realistically' depicts an object which we easily recognize. If it were a language lesson or a child's 'reading book' (the style reminds me of old-fashioned Ladybird books for children), we might expect to see the words 'This is a pipe'. To depict a pipe and then provide a label which insists that 'this is not a pipe' initially seems perverse. Is it purely irrational or is there something which we can learn from this apparent paradox? What could it mean? As our minds struggle to find a stable, meaningful interpretation we may not be too happy that there is no single, 'correct' answer to this question - although those of us who are relatively 'tolerant of ambiguity' may accept that it offers a great deal of food for thought about levels (or modes) of reality. The indexical word 'this' can be seen as a key to the interpretation of this painting: what exactly does the word 'this' refer to? Anthony Wilden suggests several alternative interpretations:

• this [pipe] is not a pipe;
• this [image of a pipe] is not a pipe;
• this [painting] is not a pipe;
• this [sentence] is not a pipe;
• [this] this is not a pipe;
• [this] is not a pipe.

  (Wilden 1987, 245)

Although we habitually relate the 'meaning' of texts to the stated or inferred purposes of their makers, Magritte's own purposes are not essential to our current concerns. It suits our purposes here to suggest that the painting could be taken as meaning that this representation (or any representation) is not that which it represents. That this image of a pipe is 'only an image' and that we can't smoke it seems obvious - nobody 'in their right mind' would be so foolish as to try to pick it up and use it as a functional pipe (although many readers will have heard by now of the unfortunate, deluded man who 'mistook his wife for a hat'). However, we do habitually refer to such realistic depictions in terms which suggest that they are nothing more nor less than what they depict. Any representation is more than merely a reproduction of that which it represents: it also contributes to the construction of reality. Even 'photorealism' does not depict unmediated reality. The most realistic representation may also symbolically or metaphorically 'stand for' something else entirely. Furthermore, the depiction of a pipe is no guarantee of the existence of a specific pipe in the world of which this is an accurate depiction. Indeed, it seems a fairly generalized pipe and could therefore be seen (as is frequently true of language lessons, children's encyclopedia entries and so on) as an illustration of the 'concept' of a pipe rather than of a specific pipe. The label seeks to anchor our interpretation - a concept to which we will return later - and yet at the same time the label is part of the painting itself rather than a title attached to the frame. Magritte's painting could be seen as a kind of defamiliarization: we are so used to seeing things and attaching labels to them that we seldom look deeper and do not see things in their specificity. One function of art (and of surrealistic art in particular) is 'to make the familiar strange' (as the Russian formalists put it).

Alfred Korzybski (1879-1950), the founder of a movement known as 'General Semantics', declared that 'the map is not the territory' and that 'the word is not the thing' (Korzybski 1933; cf. Chase 1938 and Hayakawa 1941). The non-identity of sign and thing is, of course, a very basic Saussurean principle. However, whilst Saussure's model is anti-realist, the General Semanticists adopted the realist stance that language comes 'between' us and the objective world and they sought to reform our verbal behaviour to counteract the linguistic distortion of 'reality'. They felt that one reason for the confusion of signifiers and referential signifieds was that we sometimes allow language to take us further up the 'ladder of abstraction' than we think we are. Here is a homely example of levels of verbal abstraction in relation to a cow called 'Bessie':

1. The cow known to science ultimately consists of atoms, electrons etc. according to present-day scientific inference...
2. The cow we perceive is not the word but the object of experience, that which our nervous system abstracts (selects)...
3. The word 'Bessie' (cow) is the name we give to the object of perception of level 2. The name is not the object; it merely stands for the object and omits reference to many characteristics of the object.
4. The word 'cow' stands for the characteristics we have abstracted as common to cow, cow, cow... cow. Characteristics peculiar to particular cows are left out.
5. When Bessie is referred to as 'livestock' only those characteristics she has in common with pigs, chickens, goats, etc. are referred to.
6. When Bessie is included among 'farm assets' reference is made only to what she has in common with all other saleable items on the farm.
7. When Bessie is referred to as an 'asset' still more of her characteristics are left out.
8. The word 'wealth' is an extremely high level of abstraction, omitting almost all reference to the characteristics of Bessie.
   (McKim 1972, 128; the origins of this example are in Korzybski, via Hayakawa 1941, 121ff)

The ladder metaphor is consistent with how we routinely refer to levels of abstraction - we talk of thinkers with 'their heads in the clouds' and of 'realists' with their 'feet on the ground'. As we move up the ladder we move from the particular to the general, from concrete reality to abstract generalization. The General Semanticists were of course hard-headed realists and what they wanted was for people to keep their feet firmly planted on the ground. In alerting language users to levels of abstraction, the General Semanticists sought to avoid the confusion of higher logical types with lower logical types. 'A map' is of a higher (more general) logical type than 'the territory', and linguistic representation in particular lends itself to this process of abstraction. Clearly we can learn more about a place by visiting it than by simply looking at a map of it, and we can tell more about a person by meeting them than by merely looking at a photograph of them. Translation from lower levels to higher levels involves an inevitable loss of specificity - like earth being filtered through a series of increasingly fine sieves or like photocopies being repeatedly made of the 'copies' which they produce. Being alert for the consequent losses, absences or exclusions is important to the semiotician as well as the 'general semanticist'. Whilst the logician may be able to keep such levels separate, in most acts of communication some 'slippage' occurs routinely, although we are normally capable of identifying what kind of messages we are dealing with, assigning them to appropriate levels of abstraction. Semioticians observe that some kind of 'translation' is unavoidable in human communication. Claude L�vi-Strauss declared that 'understanding consists in the reduction of one type of reality to another' (L�vi-Strauss 1961, 61). Algirdas Greimas observed that 'signification is... nothing but... transposition from one level of language to another, from one language to a different language, and meaning is nothing but the possibility of such transcoding' (cited in Jameson 1972, 215-216).

Whilst it can be useful to consider abstraction in terms of levels and logical typing, the implicit filter metaphor in the General Semanticists' 'ladder of abstraction' is too uni-dimensional. Any given 'object' of perception could be categorised in a variety of ways rather than in terms of a single 'objective' hierarchy. The categories applied depend on such factors as experience, roles and purposes. This raises issues of interpretation. For instance, looking at an advertisement featuring a woman's face, some viewers might assume that the image stood for women in general, others that she represented a particular type, role or group, and yet others might recognise her as a particular individual. Knowing the appropriate level of abstraction in relation to interpreting such an image would depend primarily on familiarity with the relevant cultural codes.

The General Semanticists set themselves the therapeutic goal of 'purifying' language in order to make its relationship to reality more 'transparent', and from such roots sprang projects such as the development of 'Basic English' (Ogden 1930). Whatever reservations we may have about such goals, Korzybski's popularization of the principle of arbitrariness could be seen as a useful corrective to some of our habits of mind. As a caveat Korzybski's aphorism seems unnecessary: we all know that the word 'dog' cannot bark or bite, but in some circumstances we do behave as if certain signifiers are inseparable from what they stand for. 'Commonsense' still leads us routinely to identify sign and thing, representation with what it represents. Terence Hawkes notes that 'Saussure points out that native speakers tend to assume a necessary "fitness", an unquestionable "identity" between signifier and signified, between "the sound image" made by the word "tree" and the concept of an actual tree. This assumption is the basis of language's anaesthetic function' (Hawkes 1977, 70).

In his massively influential book The Interpretation of Dreams, Sigmund Freud argued that 'dream-content is, as it were, presented in hieroglyphics, whose symbols must be translated... It would of course be incorrect to read these symbols in accordance with their values as pictures, instead of in accordance with their meaning as symbols' (Freud 1938, 319). He also observed that 'words are often treated in dreams as things' (ibid., 330). Magritte played with our habit of identitifying the signifier with the signified in a series of drawings and paintings in which objects are depicted with verbal labels which 'don't belong to them'. In his oil-painting entitled The Interpretation of Dreams we are confronted with images of six familiar objects together with verbal labels. Such arrangements are familiar, particularly in the language-learning context suggested by the blackboard-like background. However, we quickly realize that the words do not match the images under which they appear. If we then rearrange them in our minds, we find that the labels do not correspond to any of the images. The relation between the image of an object and the verbal label attached to it is thus presented as arbitrary.

The confusion of the representation with the thing represented is a feature of schizophrenia and psychosis (Wilden 1987, 201). 'In order to able to operate with symbols it is necessary first of all to be able to distinguish between the sign and the thing it signifies' (Leach 1970, 43). However, the confusion of 'levels of reality' is also a normal feature of an early phase of cognitive development in childhood. Jerome Bruner observed that for pre-school children thought and the object of thought seem to be the same thing, but that during schooling one comes to separate word and thing (Bruner 1966). The substitution of a sign for its referent (initially in the form of gestures and imitative sounds) constitutes a crucial phase in the infant's acquisition of language. The child quickly discovers the apparently magical power of words for referring to things in their absence - this property of displacement being a key 'design feature' of language (Piaget 1971, 64; Hockett 1958; Hockett 1960; Hockett 1965). Helen Keller, who became blind and deaf at the age of eighteen months, was gradually taught to speak by her nurse (Keller 1945). At the age of nine whilst playing with water she felt with her hand the motions of the nurse's throat and mouth vibrating the word 'water'. In a sudden flash of revelation she cried out words to the effect that 'everything has a name!'. It is hardly surprising that even in middle childhood children sometimes appear to have difficulty in separating words from what they represent. Piaget illustrates the 'nominal realism' of young children in an interview with a child aged nine-and-a-half:

"Could the sun have been called 'moon' and the moon 'sun'? - 'No.' 'Why not?' - 'Because the sun shines brighter than the moon...' 'But if everyone had called the sun 'moon', and the moon 'sun', would we have known it was wrong?' - 'Yes, because the sun is always bigger, it always stays like it is and so does the moon.' 'Yes, but the sun isn't changed, only its name. Could it have been called... etc.?' - 'No... Because the moon rises in the evening, and the sun in the day.'
(Piaget 1929: 81-2)

Thus for the child, words do not seem at all arbitrary. Similarly, Sylvia Scribner and Michael Cole found that unschooled Vai people in Liberia felt that the names of sun and moon could not be changed, one of them expressing the view that these were God-given names (Scribner & Cole 1981, 141).

The anthropologist Claude Levy-Bruhl claimed that people in 'primitive' cultures had difficulty in distinguishing between names and the things to which they referred, regarding such signifiers as as an intrinsic part of their signifieds (cited in Olson 1994, 28). The fear of 'graven images' within the Judeo-Christian tradition and also magical practices and beliefs such as Voodoo are clearly related to such a phenomenon. Emphasizing the epistemological significance of writing, David Olson argues that the invention (around 4000 years ago) of 'syntactic scripts' (which superceded the use of tokens) enabled referential words to be distinguished more easily from their referents, language to be seen as more than purely referential, and words to be seen as (linguistic) entities in their own right. He suggests that such scripts marked the end of 'word magic' since referential words came to be seen as representations rather than as instrinsic properties or parts of their referents. However, in the Middle Ages words and images were still seen as having a natural connection to things (which had 'true names' given by Adam at the Creation). Words were seen as the names of things rather than as representations. As Michel Foucault (1926-84) has shown, only in the early modern period did scholars come to see words and other signifiers as representations which were subject to conventions rather than as copies (Foucault 1970). By the seventeenth century clear distinctions were being made between representations (signifiers), ideas (signifieds) and things (referents). Scholars now regarded signifiers as referring to ideas rather than directly to things. Representations were conventionalized constructions which were relatively independent both of what they represented and of their authors; knowledge involved manipulating such signs. Olson notes that once such distinctions are made, the way is open to making modality judgements about the status of representations - such as their perceived truth or accuracy (Olson 1994, 68-78, 165-168, 279-280). Whilst the seventeenth century shift in attitudes towards signs was part of a search for 'neutrality', 'objectivity' and 'truth', in more recent times, of course, we have come to recognize that 'there is no representation without intention and interpretation' (Olson 1994, 197).

It is said that someone once asked an astronomer how he had discovered the name of a previously unknown star! Sophisticated literates are able to joke about the notion that names 'belong' to things. In one of Aldous Huxley's novels an old farmworker points out his pigs: '"Look at them, sir,' he said, with a motion of his hand towards the wallowing swine. 'Rightly is they called pigs' (Chrome Yellow, Chapter 5). Literate adults may not often seem to be prey to this sort of nominal realism. However, certain signifiers become regarded by some as far from 'arbitrary', acquiring almost magical power - as in relation to 'graphic' swearing and issues of prejudice - highlighting the point that signifiers are not socially arbitrary. Children are just as aware of this: many are far from convinced by adult advice that 'sticks and stones may break my bones, but names can never hurt me'. By adding his own label to an advertisement, Thomas Streeter at the University of Vermont alerts us to the way in which referents can be redefined by the skilful use of verbal anchors. So reminding ourselves that 'the word is not the thing' may be more useful than it might at first seem to be.

The literary theorist Catherine Belsey argues that

Language is experienced as a nomenclature because its existence precedes our "understanding" of the world. Words seem to be symbols for things because things are inconceivable outside the system of differences which constitutes the language. Similarly, these very things seem to be represented in the mind, in an autonomous realm of thought, because thought is in essence symbolic, dependent on the differences brought about by the symbolic order. And so language is 'overlooked', suppressed in favour of a quest for meaning in experience and/or in the mind. The world of things and subjectivity then become the twin guarantors of truth. (Belsey 1980, 46)

Hamlet refers to: 'the purpose of playing, whose end, both at the first and now, was and is, to hold, as 'twere, the mirror up to nature' (Shakespeare, Hamlet, III, ii), and being 'true-to-life' is probably still a key criterion in judgements of literary worth. However, Belsey comments:

The claim that a literary form reflects the world is simply tautological. If by 'the world' we understand the world we experience, the world differentiated by language, then the claim that realism reflects the world means that realism reflects the world constructed in language. This is a tautology. If discourses articulate concepts through a system of signs which signify by means of their relationship to each other rather than to entities in the world, and if literature is a signifying practice, all it can reflect is the order inscribed in particular discourses, not the nature of the world. (ibid.)

The medium of language comes to acquire the illusion of 'transparency': this feature of the medium tends to blind its users to the part it plays in constructing their experiential worlds. 'Realistic' texts reflect a mimetic purpose in representation - seeking to imitate so closely that which they depict that they may be experienced as virtually identical (and thus unmediated). Obviously, purely verbal signifiers cannot be mistaken for their real world referents. Whilst it is relatively easy for us to regard words as conventional symbols, it is more difficult to recognize the conventionality of images which resemble their signifieds. Yet even an image is not what it represents - the presence of an image marks the absence of its referent. The difference between signifier and signified is fundamental. Nevertheless, when the signifiers are experienced as highly 'realistic' - as in the case of photography and film - it is particularly easy to slip into regarding them as identical with their signifieds. In contrast even to realistic painting and drawing, photographs seem far less obviously 'authored' by a human being. Just as 'the word is not the thing' and 'the map is not the territory' nor is a photograph or television news footage that which it depicts. Yet in the 'commonsense' attitude of everyday life we routinely treat high modality signifiers in this way. Indeed, many realistic filmic narratives and documentaries seem to invite this confusion of representation with reality (Nichols 1981, 21). Thus television is frequently described as a 'window on the world' and we usually assume that 'the camera never lies'. We know of course that 'the dog in the film can bark but it cannot bite' (Hall 1980, 131) (though, when 'absorbed', we may 'suspend disbelief' in the context of what we know to be enacted drama). However, we are frequently inclined to accept 'the evidence of our own eyes' even when events are mediated by the cameras of journalists. Highly 'realistic' representations in any medium always involve a point-of-view. Representations which claim to be 'real' deny the unavoidable difference between map and territory. Lewis Carroll satirized the logical consequences of neglecting the importance of this difference:

'That's another thing we've learned from your Nation,' said Mein Herr, 'map-making. But we've carried it much further than you. What do you consider the largest map that would be really useful?'

'About six inches to the mile.'

'Only six inches!' exclaimed Mein Herr. 'We very soon got to six yards to the mile. Then we tried a hundred yards to the mile. And then came the grandest idea of all! We actually made a map of the country, on the scale of a mile to the mile'

'Have you used it much?' I enquired.

'It has never been spread out, yet,' said Mein Herr: 'the farmers objected: they said it would cover the whole country, and shut out the sunlight! So we now use the country itself, as its own map, and I assure you it does nearly as well.'
(Lewis Carroll, Sylvie and Bruno Concluded, Chapter 11)

In the sense that there is always an unavoidable difference between the represented and its representation, 'the camera always lies'. We do not need to adopt the 'scientific' realism of the so-called General Semanticists concerning the 'distortion of reality' by our signifying systems, but may acknowledge instead that reality does not exist independently of signs, turning our critical attention to the issue of whose realities are privileged in particular representations - a perspective which, avoiding a retreat to subjectivism, pays due tribute to the unequal distribution of power in the social world.

Whilst Saussurean semioticians (with language as their model) have emphasized the arbitrary relationship of the signifier to the signified, some subsequent theorists have stressed 'the primacy of the signifier' - Jacques Lacan even praised Lewis Carroll's Humpty Dumpty as 'the master of the signifier' for his declaration that 'when I use a word, it means just what I choose it to mean - neither more nor less'. Many postmodernist theorists postulate a complete disconnection of the signifier and the signified. An 'empty' or 'floating signifier' is variously defined as a signifier with a vague, highly variable, unspecifiable or non-existent signified. Such signifiers mean different things to different people: they may stand for many or even any signifieds; they may mean whatever their interpreters want them to mean. In such a state of radical disconnection between signifier and signified, 'a sign only means that it means' (Goldman & Papson 1994, 50). Such a disconnection is perhaps clearest in literary and aesthetic texts which foreground the act and form of expression and undermine any sense of a 'natural' or 'transparent' connection between a signifier and a referent. However, Jonathan Culler suggests that to refer to an 'empty signifier' is an implicit acceptance of its status as a signifier and is thus 'to correlate it with a signified' even if this is not known; 'the most radical play of the signifier still requires and works through the positing of signifieds' (Culler 1985, 115). Shakespeare famously referred to 'a tale told by an idiot, full of sound and fury, signifying nothing' (Macbeth V, iii). The notion of the 'floating signifier' can be found around the year 1950 in L�vi-Strauss (see Lechte 1994, 26-7, 64, 73). Roland Barthes referred specifically to non-linguistic signs as being so open to interpretation that they constituted a 'floating chain of signifieds' (Barthes 1977, 39). The first explicit reference to an 'empty signifier' of which I am aware is that of Barthes in his essay 'Myth Today' (Barthes 1957; cf. Culler 1975, 19). Barthes defines an empty signifier as one with no definite signified. There are some similarities with the linguistic concept of an 'empty category' (Lechte 1994, 64) and with Hjelmslev's figurae or non-signifying sign elements (ibid., 137; see Articulation).

Whereas Saussure saw the signifier and the signified (however arbitrary their relationship) as being as inseparable as the two sides of a piece of paper, poststructuralists have rejected the stable and predictable relationship embedded in his model. The French psychoanalyst Jacques Lacan wrote of 'the incessant sliding of the signified under the signifier' (Lacan 1977, 154) - he argued that there could be no anchoring of particular signifiers to particular signifieds - although this in itself is hardly contentious in the context of psychoanalysis. Jacques Derrida refers also to the 'freeplay' of signifiers: they are not fixed to their signifieds but point beyond themselves to other signifiers in an 'indefinite referral of signifier to signified' (Derrida 1978, 25). He championed the 'deconstruction' of western semiotic systems, denying that there were any ultimate determinable meanings. Whilst for Saussure the meaning of signs derives from how they differ from each other, Derrida coined the term diff�rance to allude also to the way in which meaning is endlessly deferred. There is no 'transcendent signified' (Derrida 1978, 278-280; Derrida 1976, 20). These notions were anticipated by Peirce in his version of 'unlimited semiosis', although he emphasized that in practice this potentially endless process is inevitably cut short by the practical constraints of everyday life (Gallie 1952, 126). Unlike Peirce, postmodernist theories grant no access to any reality outside signification. For Derrida, 'il n'y a riens hors du texte' ('there is nothing outside the text') - although this assertion need not necessarily be taken 'literally' (Derrida 1976, 158, 163). For materialist marxists and realists, postmodernist idealism is intolerable: 'signs cannot be permitted to swallow up their referents in a never-ending chain of signification, in which one sign always points on to another, and the circle is never broken by the intrusion of that to which the sign refers' (Lovell 1983, 16). Some theorists note that an emphasis on the unavoidability of signification does not necessitate denying any external reality. David Sless comments that 'I am not suggesting that the only things in the universe are signs or texts, or that without signs nothing could exist. However, I am arguing that without signs nothing is conceivable' (Sless 1986, 156). We may note in passing that since the phrase 'the empty (or free-floating) signifier' has become something of an academic 'sound-bite' the term itself is ironically in danger of being an empty signifier.

The notion of reality as degenerative is found in the Romantic mythology of a primal state of unmediatedness (referring to children before language or human beings before The Fall) (Chandler 1995, 31-2). In his book The Image, Daniel Boorstin charted the rise of what he called 'pseudo-events' - events which are staged for the mass media to report (Boorstin 1961). However, any 'event' is a social construction - bounded 'events' have no objective existence, and all news items are 'stories' (Galtung & Ruge 1981).

We might posit three key historical shifts in representational paradigms in relation to Peirce's differential framing of the referential status of signs:

• an indexical phase - the signifier and the referent are regarded as directly connected;
• an iconic phase - the signifier is not regarded as part of the referent but as depicting it transparently;
• a symbolic phase - the signifier is regarded as arbitrary and as referring only to other signs.

Such a schematization bears some similarity to that of the postmodernist Jean Baudrillard. Baudrillard interprets many representations as a means of concealing the absence of reality; he calls such representations 'simulacra' (or copies without originals) (Baudrillard 1984). He sees a degenerative evolution in modes of representation in which signs are increasingly empty of meaning:

These would be the successive phases of the image:
1. It is the reflection of a basic reality.
2. It masks and perverts a basic reality.
3. It masks the absence of a basic reality.
4. It bears no relation to any reality whatever: it is its own pure simulacrum. (Baudrillard 1988, 170)

Baudrillard argues that when speech and writing were created, signs were invented to point to material or social reality, but the bond between signifier and signified became eroded. As advertising, propaganda and commodification set in, the sign began to hide 'basic reality'. In the postmodern age of 'hyper-reality' in which what are only illusions in the media of communication seem very real, signs hide the absence of reality and only pretend to mean something. For Baudrillard, simulacra - the signs which characterize late capitalism - come in three forms: counterfeit (imitation) - when there was still a direct link between signifiers and their signifieds; production (illusion) - when there was an indirect link between signifier and signified; and simulation (fake) - when signifiers came to stand in relation only to other signifiers and not in relation to any fixed external reality. It is hardly surprising that Douglas Kellner has criticized Baudrillard as a 'semiological idealist' who ignores the materiality of sign production (cited in Stam 2000, 306). Baudrillard's claim that the Gulf War never happened is certainly provocative (Baudrillard 1995).

Such perspectives, of course, beg the fundamental question, 'What is "real"?' The semiotic stance which problematizes 'reality' and emphasizes mediation and convention is sometimes criticized as extreme 'cultural relativism' by those who veer towards realism - such critics often object to an apparent sidelining of referential concerns such as 'accuracy' (e.g. Gombrich 1982, 188, 279, 286). However, even philosophical realists would accept that much of our knowledge of the world is indirect; we experience many things primarily (or even solely) as they are represented to us within our media and communication technologies. Since representations cannot be identical copies of what they represent, they can never be neutral and transparent but are instead constitutive of reality. As Judith Butler puts it, we need to ask, 'What does transparency keep obscure?' (Butler 1999, xix). Semiotics helps us to not to take representations for granted as 'reflections of reality', enabling us to take them apart and consider whose realities they represent.

Contents

• Contents Page
• Preface
• Introduction
• Signs
• Modality and representation
• Paradigms and syntagms
• Syntagmatic analysis
• Paradigmatic analysis
• Denotation, connotation and myth
• Rhetorical tropes
• Codes
• Modes of address
• Encoding/Decoding
• Articulation
• Intertextuality
• Criticisms of semiotic analysis
• Strengths of semiotic analysis
• D.I.Y. semiotic analysis
• Glossary of key terms
• Suggested reading
• Semiosis bookstore
• References
• Index
• Semiotics: The Basics