a. Semiotics: looking for meaning in signs
i. A dyadic approach: signifier and signified
One first semiotic model comes from Ferdinand de Saussure, a Swiss linguist and semiotician, often considered the founder of XXth-century linguistics (Wintle, 2002). Although he is considered one of the founders of semiotics – which he actually referred to as semiology – his ideas are in fact derived from Aristotelian and Neo-Platonist theories that can be traced back to the Middle Ages. To quote Munteanu (1996):
“[A]s for the constitution of Saussurian semiotic theory, the importance of the Augustinian thought contribution (correlated to the Stoic one) has also been recognized. Saussure did not do anything but reform an ancient theory in Europe, according to the modern conceptual exigencies”
159
As a linguist, Saussure in his analysis focuses of course on linguistic signs, such as words, lexemes, and graphemes. His analysis is called dyadic, meaning it splits every sign into two semiotic layers, namely the signifiant (“signifier” in English), and the signifié (“signified”).
In Saussure’s words (1977), the signified is the concept of the sign, and the signifier is the sound pattern. He then qualifies his words by writing:
“The sound pattern is not actually a sound; for a sound is something physical. A sound pattern is the hearer’s psychological impression of a sound, as given to him by the evidence of his senses. This sound pattern may be called a ‘material’
element only in that it is the representation of our sensory impressions. The sound pattern may thus be distinguished from the other element associated with it in a linguistic sign. This other element is generally of a more abstract kind: the concept.”
The distinction is conventionally made now by accepting that the signifier is the form that is taken by the sign, and the signified refers to the concept linked to it. As time went by, Saussure’s ideas became more and more practical (Chandler, 2007). For instance, both the signifier and the signified were, to him, purely mental constructs (what he refers to as the
“impression of a sound” for instance): one would think of the image of a boat (here, the signified), and would mentally link it to the word “boat” (the signifier), without having to say the word aloud. Further academic works, such as the ones from Jakobson (1990), gave the signifier a much more physical dimension, as something that can be seen, touched, smelled, tasted or heard.
The combination of signifier and signified, both forming a relationship called signification, is a sign. Both signifier and signified are essential for the sign to exist: there cannot be a signifier without any meaning, or a signified without any form (Saussure, 1977).
However, the same signifier could represent different signifieds: the word open could be used both for the sign in front of a shop, or written on a push-button inside a lift: “push to open the door” (Chandler, 2007). Similarly, the concept open can be linked to various signifiers. The Saussurean analysis of signs favours the spoken word to the written one, considering the latter as a secondary, dependant on the former, yet comparable sign system (Saussure, 1977), turning the written word into the signifier, and speech into the signified. Considering that Saussure identified semiotics as “a science which studies the role of signs as part of social life” (1977), it is rather ironic to consider that his vision of linguistic signs is this immaterial (not to mention
“abstract”, a term that he despised (ibid.)). The philosopher and writer Susanne Langer (1951),
160
although unrelated to Saussurean semiotics, helps shed some light as to why the vision is this disconnected to the physical world:
“Symbols are not proxy for their objects but are vehicles for the conception of objects. […] In talking about things we have conceptions of them, not the things themselves; and it is the conceptions, not the things, that symbols directly mean.
Behaviour towards conceptions is what words normally evoke; this is the typical process of thinking. […] If I say “Napoleon”, you do not bow to the conqueror of Europe as though I had introduced him, but merely think of him”
The immateriality of the Saussurean signs is because the sheer value of words resides in the lack of value words have in themselves, which allows signs to convey communicative transparency. Saussure uses the analogy of a chessboard and its pieces to illustrate his idea:
each piece has a different value depending on its position on the board, the same way that “the sign is more than the sum of its parts” (Chandler, 2007). Because of this lack of intrinsic value, relations between signs are essential to create meaning; he writes “everything depends on relations” (1977), meaning that no sign can, by itself, convey meaning; only relations between signs can do so. This means that both the signifier and the signified are essentially relational concepts. This can be difficult to grasp, for one may think that a flower is a flower, and that the word “flower” does have a meaning by itself. The Saussurean logic would yet argue that the meaning of “flower” can only be understood through its relation to an environment of words, such as “garden”. There are therefore two levels of relationships in Saussurean signs: the relationship between signifier and signified, and the relationship between signs. To sum it up, while signification (in the sense of “what is signified”) undoubtedly relies on the relationship between the two parts of a given sign, the value of said sign is defined by the relationships between the sign and other signs within the system as a whole. Saussure (ibid.) illustrates this distinction with the example of the word “sheep”:
“The French word mouton may have the same meaning as the English word sheep; but it does not have the same value. There are various reasons for this, but in particular the fact that the English word for the meat of this animal, as prepared and served for a meal, is not sheep but mutton. The difference in value between sheep and mouton hinges on the fact that in English there is also another word mutton for the meat, whereas mouton in French covers both.”
The difference between sheep and mutton is a good example of the Saussurean conception of differential meaning. Therefore, each sign, of any semiological system exists in distinction from the other ones. Advertising is a good example of this differentiation of signs:
161
when advertisers position a product, they generally do not do it through the relationship of advertising signifiers and real world, physical referents, but rather use the differentiation of each sign from the others to which it is associated. The differences between signs are mostly negative: a sign exists by contrast with other signs of a similar system. For instance, if one had to explain the colour blue to someone who does not share the same language, instead of picking various blue items, one would find it easier to choose the blue item out of a collection of similar items, only different by colour.
One last trait of the Saussurean signs is their arbitrariness, which is also the principal feature of language, according to Saussure (1977). The idea of an artificial relationship between signifier and signified is not new, since it was debated by Plato in his Cratylus. This arbitrary relationship between signifier and signified is what Hockett (1958) refers to as the “design feature” of language, and could explain the exceptional changeability of language (Lyons, 1977). The arbitrariness refers to the fact that there is no natural, nor inherent connection between a concept and how we word it, may it be in speech or in writing, as Saussure writes,
“the signs used in writing are arbitrary. The letter t, for instance, has no connection with the sound it denotes”. It is here important to note that this concept could not be applied to Japanese writing! Indeed, as we have seen before, many of the Japanese kanji, by their very nature and history, defy this rule of arbitrariness. The artificiality of the signifier/signified relationship also comprises the fact that different languages use different signified to refer to the same signifier (a tree is “un arbre” in French, and “木” in Japanese), and the acknowledgement that it is impossible to define a signifier better suited than another. Arbitrariness is a rather radical statement, in so as it defines languages as autonomous from reality. As Chandler (2007) states:
“The Saussurean model, with its emphasis on internal structures within a sign-system, can be seen as supporting the notion that language does not reflect reality but rather constructs it.”
The arbitrary characteristic of signs does help to explain the scope for their understanding (and the importance of context). There is no one-to-one link between signified and signifier; signs have manifold meanings, rather than unique ones. Within the frame of a single language, one signifier may allude various signifieds (with the clear example of puns) and many signifiers may account for a single signified, which is the case with synonyms. Critics such as Jakobson (1990) point out that some signifier/signified relationships are not completely artificial, with the example of onomatopoeia, particularly those of the sounds made by animals we are accustomed to. This point of view is valid, yet weakened by the variation of
162
interpretation of the same sound that is made around the world (the ガオー gaō made by a Japanese lion is quite far from its English counterpart, “rawr”). Saussure’s answer to critics is the recognition of a certain nuance in the arbitrariness of the relationships. If linguistic signs were to be completely arbitrary, language would not be a system and its communicative purpose would be destroyed. While the sign is not defined out of the language, it must undergo some intra-linguistic determination. For instance, signifiers must come about as well-formed sets of sounds which agree with existing patterns within the language in question: a new word in English that would need to be pronounced with the guttural French [r] sound would not match the language. Besides, a compound noun such as “screwdriver” is not completely arbitrary since it is an eloquent association of two existing signs. Saussure (1977) then concedes with the idea of degrees of arbitrariness:
“The fundamental principle of the arbitrary nature of the linguistic sign does not prevent us from distinguishing in any language between what is intrinsically arbitrary – that is, unmotivated – and what is only relatively arbitrary. Not all signs are absolutely arbitrary. In some cases, there are factors which allow us to recognize different degrees of arbitrariness, although never to discard the notion entirely. The sign may be motivated to a certain extent.”
Here, then, Saussure adjusts his stance slightly and refers to signs as being “relatively arbitrary”. Natural languages are not randomly defined, unlike past creations such as the Morse Code. Nor does the arbitrary nature of the sign make it politically or socially ‘neutral’
(understood as “free of connotation”): in Western culture, “white” has come to be a privileged (but stereotypically ‘invisible’) signifier (Dyer, 1997). As Lévi-Strauss stated, the sign is arbitrary a priori but no longer is a posteriori – after the sign has come into historical existence, it cannot be arbitrarily changed (Lévi-Strauss, 1963). As the sign becomes a historical reality and is used, it acquires a history of its own, and a connotative dimension. Saussure writes that
“a language is always an inheritance from the past” which, as a user, one has “no choice but to accept”.
The arbitrariness principle does not, of course, imply that anyone can arbitrarily choose any signifier for a given signified. The signifier/signified relation is not a matter of individual choice; should it be the case, any communication would be rendered impossible. As an individual, a mother tongue is a “given”: no one creates a system for themselves. Saussure considered the language system as a non-negotiable ‘contract’ into which one is born. The ontological arbitrariness that it encompasses becomes imperceptible to us as we come to accept it as natural.
163
ii. Peirce’s trichotomy
Another approach to semiotics is the triadic one, conventionally theorised in the United States by the scholar Charles S. Peirce. Where the Saussurean conception of a sign could be considered as a “self-contained dyad”, the Peircean consideration can be defined as “something that relates to something else for someone in some respect or capacity” (Merrell, 2001). Let us first explain this rather hazy statement. As it is understood in the word triadic, the Peircean approach divides the sign in three parts: the representamen, which is the form which the sign takes (its physical, material one in the case of a tangible object), the interpretant, which concerns the sense that one makes of the sign, and the object, which stands for “something beyond the sign to which it refers” (Chandler, 2007). To quote Peirce (1932):
“A sign… [in the form of a representamen] is something which stands to somebody for something in some respect or capacity. It addresses somebody, that is, creates in the mind of that person an equivalent sign, or perhaps a more developed sign.
That sign which it creates I call the interpretant of the first sign. The sign stands for something, its object. It stands for that object, not in all respects, but in reference to a sort of idea, which I have sometimes called the ground of the representamen.”
All three elements are essential to define a sign, which is the result of what is depicted (the object), how it is depicted (the representamen), and how it is interpreted (the interpretant).
Before we go any further into explanations, the conception of object needs to be slightly qualified. Peirce (along with Merrell, 2001) create a distinction between the “real” object and the “semiotic object”, the latter being the one which will be discussed from now on. According to Peirce, our knowledge of the world can never be absolute, only an approximation of the “real”
world as it is, or as it is becoming. Expanding such a claim, the object that one can see, touch, taste, hear or smell, the “semiotically real object” can never be exactly similar to the “really real object”, for the world is far too complex for our minds to grasp it all. Therefore, an object can never be more than “semiotically real”. Object and representamen are mediated by the interpretant, which both creates an interrelation between them both, and with the interpretant.
A proper, balanced sign offers these three, interdependent parts. Munday, retrieved in Chandler (2007), illustrates the three parts of a sign as a box:
164
“The three elements that make up a sign function like a label on an opaque box that contains an object. At first the mere fact that there is a box with a label on it suggests that it contains something, and then when we read the label we discover what that something is. The process of semiosis, or decoding the sign, is as follows.
The first thing that is noticed (the representamen) is the box and label; this prompts the realization that something is inside the box (the object). This realization, as well as the knowledge of what the box contains, is provided by the interpretant. ‘Reading the label’ is actually just a metaphor for the process of decoding the sign. The important point to be aware of here is that the object of a sign is always hidden. We cannot actually open the box and inspect it directly.
The reason for this is simple: if the object could be known directly, there would be no need of a sign to represent it. We only know about the object from noticing the label and the box and then ‘reading the label’ and forming a mental picture of the object in our mind. Therefore, the hidden object of a sign is only brought to realization through the interaction of the representamen, the object and the interpretant.”
Eco (1976) illustrated the Peircean model as follows:
If a parallel had to be made between the dyadic Saussurean conception of sign and the Peircean version, the Peircean representamen would roughly equal Saussure’s signifier. The interpretant could potentially be the signified, albeit the difference that it is itself a sign in the mind of the interpreter. To understand this and Jakobson’s remark “the meaning of the sign is the sign it can be translated into” (1971) and the principle of unlimited semiosis, which we will come back to later, let us remember that to Peirce, one cannot understand the world, nor the reality, in its entire complexity. Since our knowledge is only partial, any interpretation can be reinterpreted, which brings the concept of unlimited semiosis, which Umberto Eco conjectures in his Theory of Semiotics (1976). Because one’s interpretation of a sign becomes a
interpretant
object representamen
165
representation of the sign, this representation can be reinterpreted, leading to another representamen, and so on. Reasonably anguished by such a perspective, Eco finds an answer, in the form of what he calls an “energetic interpretant”. Essentially, to him, the interpretant produced by an object is double in nature. There is on one hand the emotional interpretant, the mental sign, the mark that, in each and everyone’s mind, creates the link between an object and a sign. Interpretations, within affective interpretants, have consequences remaining within the framework of interpretation and change of representations, without altering behaviour in any way. “Energetic interpretant” is, on the other hand, the one creating a change of habit (Eco, 1984). When this seemingly infinite sequence of representations of representations leaves the mental milieu to enter a more applied one, causing a change in behaviour, “our way of acting within the world is either transitorily or permanently changed” (ibid.). This new mindset, this practical aspect, is the final interpretant that ends the endless cycle of semiosis, proposing a tangible result to hang onto.
One key difference between the dyadic and triadic model is Peirce’s object (or referent).
The inclusion of a more tangible term goes in contradiction with the Saussurean model, which was more intellectual (cf. supra). Of course, the Peircean object is not restricted to physical things, and encompasses more intangible concepts, but the choice of words definitely left space for materiality. The separation of objects and interpretant derives from Peirce’s belief that
‘meaning’ comes from both ‘reference’ (the object) and ‘sense’ (the representamen and the interpretant), which would allow a broader consideration of what is a sign (Bruss, 1978).
The Saussurean idea of signs is that of an arbitrary relationship between signifier and signified (cf. supra). Peirce’s model embraces a broader vision, with both arbitrary, or conventional signs, and natural signs, which resemble the object they depict. During the IVth century CE, St. Augustine, in the second book of his De doctrina christiana (translated in 1995 by Green), defined natural signs as ones that could be interpreted as such thanks to an immediate link to what they signified, even when no conscious purpose had formed them as such. For instance, smoke indicates fire, and footprints can indicate that an animal or a person has passed by. Peirce offers more than just a separation between natural and arbitrary signs, and has even built a typology of them, classified in different “modes of relationship” between sign vehicles and what is signified (Hawkes, 1977). The terminology ‘mode’ is rather definite, even though one could easily slip and characterise the ensuing categories as ‘types’. A sign can be iconic, indexical or symbolic, or even a combination of all. For instance, “a map is indexical in pointing to the locations of things, iconic in representing the directional relations and distances between
166
landmarks, and symbolic in using conventional symbols (the significance of which must be learned)” (Chandler, 2007).
• Icon
The icon is the simplest mode, since it is a pattern that physically resembles what it
‘stands for’. Icons have qualities which “resemble” those of the stuffs they characterise, which cause them to “excite analogous sensations in the mind” (Peirce, 1932). Simply because an interpretant bears a resemblance to that which it portrays does not necessarily make it purely iconic. Langer (1951) argues that “the picture is essentially a symbol, not a duplicate, of what it represents”. Pictures look like what they epitomise only in some aspects. What we tend to identify in an image are analogous relations of parts to a whole (ibid.). There is of course a lexical issue with terms such as “icon”, index” and “symbol” (cf. infra) since their semiotic, technical meaning differs from their everyday, vulgarised ones. An icon, in popular terms, refers to three things:
- Religious icons, which are visual artworks of holy images, designed to be venerated.
- Computer icons, those small images that signify specific functions. Icons can actually convey an iconic mode, but according to the ones that are referred to, computer icons can also be indexical, or symbolic (cf. infra).
- The noun “icon”, and the related adjective “iconic”, can be used to refer to a person or an item which is famous among a given community or culture.
In the Peircean sense, the major feature of iconicity is purely alleged semblance. Peirce (1932) stated that an iconic sign represents its object “mainly by its similarity”. Note that notwithstanding the name, icons are not automatically visual (cf. infra). A sign is an icon
“insofar as it is like that thing and used as a sign of it” (ibid.). On Saussurean terms, it would be a signifier that imitates the signified. For instance, a picture or a portrait of your face is an icon of you. The little picture of a printer in any computer software is an icon of the printing function (however, the actual word ‘print’ is not an icon, since the letters of the words and the act of printing in itself do not resemble each other). Words also can be, at least partly iconic.
Onomatopoeias, onomatopoetic words such as “bow-wow”, “splash” or “hiccup” resemble the sounds they represent- at least a little. And the bird called the whippoorwill produces a call resembling this English phrase, so whippoorwill is an iconic word. In Japanese it is common in hypocoristic speech to use animal sounds to refer to the animals that make the sounds. For example,ワンワン wan-wanmay mean ‘dog’, and モーモーmō-mōmay mean ‘cow’. Even if a word is, by itself, not iconic, it can be pronounced in an iconic way: by saying “Kazuya took
167
a carrot and went chop chop chop chop”, instead of “he chopped it”, the discourse gives “chop”
an iconic dimension. Imitative gestures, such as stretching one’s hand, palm facing outwards, to mimic a stop, are also iconic. Because iconicity relies on resemblance, there is debate over how similar a word must be to the signified to be iconic.
Semioticians commonly claim that there are no ‘pure’ icons (Chandler, 2007). In his 1976 Theory of semiotics, Eco provides an extensive critique of iconicity. All artists use stylistic conventional techniques and these are, of course, customarily and historically inconstant. Peirce (1932) wrote that although “any material image” (a painting, for instance) may be considered as looking like what it aims at depicting, it is “largely conventional in its mode of representation”. He then claims:
“We say that the portrait of a person we have not seen is convincing. So far as, on the ground merely of what I see in it, I am led to form an idea of the person it represents, it is an icon. But, in fact, it is not a pure icon, because I am greatly influenced by knowing that it is an effect, through the artist, caused by the original’s appearance… Besides, I know that portraits have but the slightest resemblance to their originals, except in certain conventional respects, and after a conventional scale of values, etc.”
• Index
An index is defined by some sensory feature. Indexicality, because of its intricacy, is probably the most unfamiliar sign system. It is a mode in which the representamen, or the signifier, is not arbitrarily, but rather directly connected, in some way, to the interpretant, or the signified. This relationship is either physical or causal, and do not depend on the intention of the subject. Typically, it can be understood that a sign A, something that can be seen, heard, smelt, touched or tasted, connects with, infers or points towards B, something that would be of interest to an animal (Gasser, 2012). Indexicality is then automatically rooted in the physical world, since Peirce considers the object of in this sign-meaning creation “necessarily existant”
(Peirce, 1932), and their connection “a matter of fact” (ibid.). Whereas iconicity is categorised by similarity, indexicality is defined by contiguity. Peirce (ibid.) writes: “Psychologically, the action of indices depends upon association by contiguity, and not upon association by resemblance or upon intellectual operations”. To Bruss (1978), indexicality is “a relationship rather than a quality. Hence the signifier need have no particular properties of its own, only a demonstrable connection to something else. The most important of these connections are spatial co-occurrence, temporal sequence, and cause and effect”.
168
All animals use several kinds of indexical signs when dealing with the world. The more intellectually developed animals are good at learning and exploiting more advanced indices: a cat will use and learn many more indexical signs than a frog, a fish or an ant -- which tend to be restricted to ones acquired innately (ibid.). For instance, it is commonly thought in France that when swallows are flying low, a storm is coming. This indexical sign has scientific roots (as cumulonimbi drag the hot air up, only cold air is left underneath, and midges, which are swallows’ privileged food, must fly lower to benefit from warmer temperatures; the swallows simply follow the midges) which are often ignored for the sign in itself. There, the link between the swallows (the signifier) and the storm (the signified) is observed and inferred; the same way one infers fire when seeing smoke, sound when hearing echo, flowers when smelling them…
Other indexical signs can be medical (a symptom is an indexical sign of a disease, or an infection), measured (time is measured by a clock, temperature by a thermometer), signalised (a knock on the door means someone would like to come in), personal trademarks (slang, catchphrases or handwriting), etc. All these examples depend on a certain statistical uniformity of part A (the signal pattern) with part B (the behaviorally relevant state). The exploitation of this regularity requires first, detecting property A (which is not necessarily simple) and either learning (or innately knowing) its correlation with the B. The indexical connection needs not be perfect. It isn't always warmer closer to the sea surface, swallows flying low don't always mean the rain is coming this way, and even a stoplight can be broken sometimes, which means that even if it does not turn green or red, it can be either safe or dangerous to cross the street. This doesn't detract from the usefulness of these signs to guide one’s life in a confusing and only partly predictable world. Words are said to be indexical when they directly point to their meaning - without depending on any relationship to other words.
Thus, words like here, there, I, me, you, this, etc. For all of these there is an implied pointing gesture. A connection can be found with the Latin root of the word index, since it originally only referred to the index finger, which is traditionally used to point at things. Of the three modes, only indexicality can serve as evidence of an object’s existence.
• Symbol
The two precedent semiotic modes can regroup natural signs, since there is either a physical resemblance (iconic) between signified and signifier, or at least a direct connection (indexical) between the two. The symbolic mode is much more conventional, if not arbitrary (Peirce, 1933). A symbolic relationship between a representamen and an interpretant needs to be agreed upon, and learnt by the language users. Nowadays language is usually observed as a (mostly) symbolic sign-system, yet Saussure avoided defining linguistic signs as ‘symbols’
169
precisely because of the risk of misunderstanding with popular usage (Chandler, 2007). For instance, let us consider the word ‘kitty’. On the one hand, this noun could be considered as an index for a cat, for English speakers. One could easily point at a cat and say “kitty”. Yet it is quite rare for the utterance of a word to connect with the thing it refers to. Sometimes such a correlation exists, of course, but a word in any language is vastly more complex and sophisticated even for language-learning infants. Some signs that show that “kitty” is not indexical to the cat is that, for instance, the word “kitty” can be used even when a cat is not here, proving that the connection between the animal and the word is rather weak. Such a connection is at work when language users refer to entities that do not exist, such as “fairy”,
“god”, “unicorn” … None of these could be indices, for no one could point at them. However, mentioning a symbolic sign often enables the utterance of other words with which the original word (here, “kitty”) shares strong connections, such as “cat”, “cute”, “soft”, “fluffy”, “cat food”, etc. (Gasser, 2012). This suggests that the word “kitty” may be somehow physically linked to these other words in the brain. It also infers that “kitty” finds some meaning in the very stimulation of just these specific words (and their related emotional dimension) when the word
“kitty” is spoken (Chandler, 2007).
These word-to-word connections, also referred to as word-associates, are critical for anchoring the meaning of a word without any necessary correlation in space and time between the signal (the sound of the word) and its meaning. Because of the directness of the connection between representamen and interpretant, indices do not require any such set of relationships to work as signs. Symbols, like most words in a human language, are (a) easily detachable from their context, and (b) are closely associated with large sets of other words. They allow humans to have a vocabulary that goes beyond their immediate environment: even children who have grown in an area where the climate is very hot know the word “snow”, and even if they have never actually experienced the phenomenon, can link it to other words such as “cold”, “white”,
“flake”, “ice”, etc. (Merrell, 2001). Thanks to these connections, the users get an idea of what experiencing snow is like - enough to read and produce the words fittingly. This is the enormous power of human symbols: once one has learnt a basic lexicon, based in part on indexical relationships, one can use it as a bootstrap to many other new concepts and words (ibid., Gasser, 2012). Given the opportunity of transgenerational cultural transmission, human knowledge and understanding have become cumulative and have grown at a very rapid rate, in comparison to the creation and transmission of innate knowledge. Symbols can also be nonverbal: although words stand for the archetypic symbol, non-words, such as flags, mascots or totem animals, fonts and logos are also symbols, for they directly link to other words. Mathematical and logical symbols also get their meaning from their relation to other symbols. Thus, π pi is defined as the