Whereas syntagmatic analysis studies the 'surface structure' of a text, paradigmatic analysis seeks to identify the various paradigms (or pre-existing sets of signifiers) which underlie the manifest content of texts. This aspect of structural analysis involves a consideration of the positive or negative connotations of each signifier (revealed through the use of one signifier rather than another), and the existence of 'underlying' thematic paradigms (e.g. binary oppositions such as public/private). 'Paradigmatic relations' are the oppositions and contrasts between the signifiers that belong to the same set from which those used in the text were drawn.
Semioticians often focus on the issue of why a particular signifier rather than a workable alternative was used in a specific context: on what they often refer to as 'absences'. Saussure noted that a characteristic of what he called 'associative' relations - what would now be called paradigmatic relations - was that (in contrast to syntagmatic relations) such relations held 'in absentia' - in the absence from a specific text of alternative signifiers from the same paradigm (Saussure 1983, 122; Saussure 1974, 123). He also argued that signs take their value within the linguistic system from what they are not (Saussure 1983, 115; Saussure 1974, 117). We have popular sayings in English concerning two kinds of absences: we refer to 'what goes without saying' and 'what is conspicuous by its absence'. What 'goes without saying' reflects what it is assumed that you 'take for granted' as 'obvious'. In relation to the coverage of an issue (such as in 'factual' genres) this is a profoundly ideological absence which helps to 'position' the text's readers, the implication being that 'people like us already agree what we think about issues like that'. As for the second kind of absence, an item which is present in the text may flout conventional expectations, making the conventional item 'conspicuous by its absence' and the unexpected item 'a statement'. This applies no less to cultural practices. If a man wears a suit at his office it says very little other than that he is conforming to a norm. But if one day he arrives in jeans and a tee-shirt, this will be interpreted as 'making a statement'. Analysing textual absences can help to reveal whose interests are served by their omission. Such analysis pays particular attention to the issue of which questions are left unasked.
Paradigmatic analysis involves comparing and contrasting each of the signifiers present in the text with absent signifiers which in similar circumstances might have been chosen, and considering the significance of the choices made. It can be applied at any semiotic level, from the choice of a particular word, image or sound to the level of the choice of style, genre or medium. The use of one signifier rather than another from the same paradigm is based on factors such as technical constraints, code (e.g. genre), convention, connotation, style, rhetorical purpose and the limitations of the individual's own repertoire. The analysis of paradigmatic relations helps to define the 'value' of specific items in a text.
Some semioticians refer to the 'commutation test' which can be used in order to identify distinctive signifiers and to define their significance - determining whether a change on the level of the signifier leads to a change on the level of the signified. Its origins lie in a linguistic test of substitution applied by the Prague Structuralists (including Roman Jakobson). In order to identity within a language its phonemes and their 'distinctive features' (for example, voiced/unvoiced; nazalized/not nazalized), linguists experimented with changes in the phonetic structure of a word in order to see at what point it became a different word. The original commutation test has evolved into a rather more subjective form of textual analysis. Roland Barthes refers to using the commutation test to divide texts into minimal significant units, before grouping these units into paradigmatic classes (Barthes 1967, 48). To apply this test a particular signifier in a text is selected. Then alternatives to this signifier are considered. The effects of each substitution are considered in terms of how this might affect the sense made of the sign. This might involve imagining the use of a close-up rather than a mid-shot, a subtitution in age, sex, class or ethnicity, substituting objects, a different caption for a photograph, etc. It could also involve swapping over two of the existing signifiers, changing their original relationship. The influence of the substitution on the meaning can help to suggest the contribution of the original signifier and also to identify syntagmatic units (Barthes 1967, III 2.3; Barthes 1985, 19-20). The commutation test can identify the sets (paradigms) and codes to which the signifiers used belong. For instance, if changing the setting used in an advertisement contributes to changing the meaning then 'setting' is one of the paradigms; the paradigm set for the setting would consist of all of those alternative signifiers which could have been used and which would have shifted the meaning. Arriving at a party in a Nissan Micra 'says something different' from arriving in an Alfa Romeo. Wearing jeans to a job interview will be interpreted differently from 'power dressing'.
The commutation test may involve any of four basic transformations, some of which involve the modification of the syntagm. However, the consideration of an alternative syntagm can itself be seen as a paradigmatic substitution.
These four basic tranformational processes were noted as features of perception and recall (Allport & Postman 1945; Newcomb 1952: 88-96). They correspond exactly to the four general categories to which Quintilian (circa 35-100 AD) assigned the rhetorical figures (or tropes) as 'deviations' from 'literal' language (Nöth 1990, 341).
Structuralists emphasize the importance of relations of paradigmatic opposition. The primary analytical method employed by many semioticians involves the identification of binary or polar semantic oppositions (e.g. us/them, public/private) in texts or signifying practices. Such a quest is based on a form of 'dualism'. Note that the slanting line linking and separating the two terms in such pairings is sometimes referred to by semioticians as 'the bar', a term employed by Jacques Lacan (Lacan 1977, 149).
Dualism seems to be deeply-rooted in the development of human categorization. Jakobson and Halle observe that 'the binary opposition is a child's first logical operation' (Jakobson & Halle 1956, 60). Whilst there are no opposites in 'nature', the binary oppositions which we employ in our cultural practices help to generate order out of the dynamic complexity of experience. At the most basic level of individual survival humans share with other animals the need to distinguish between 'own species and other, dominance and submission, sexual availability or lack of availability, what is edible and what is not' (Leach 1970, 39). The range of human distinctions is far more extensive than those which they share with other animals since it is supported by the elaborate system of categorization which language facilitates. The British anthropologist Sir Edmund Leach reflects that 'a speechless ape presumably has some sort of feelings for the opposition "I"/"Other", perhaps even for its expanded version "We"/"They", but the still more grandiose "Natural"/"Supernatural" ("Man"/"God") could only occur within a linguistic frame... The recognition of a distinction Natural/Supernatural (Real/Imaginary) is a basic marker of humanity' (Leach 1982, 108-9).
People have believed in the fundamental character of binary oppositions since at least classical times. For instance, in his Metaphysics Aristotle advanced as primary oppositions: form/matter, natural/unnatural, active/passive, whole/part, unity/variety, before/after and being/not-being. But it is not in isolation that the rhetorical power of such oppositions resides, but in their articulation in relation to other oppositions. In Aristotle's Physics the four elements of earth, air, fire and water were said to be opposed in pairs. For more than two thousand years oppositional patterns based on these four elements were widely accepted as the fundamental structure underlying surface reality.
The elements of such frameworks appeared in various combinations, their shifting forms driven in part by the tensions inherent within such schemes. The theory of the elements continued to enjoy widespread influence until the time of scientists such as Robert Boyle (1627-91).
|Element||Quality||Humour||Body fluid||Organ||Season||Cardinal point||Zodiac signs||Planet
||hot and moist
||sanguine (active and enthusiastic)
||Gemini, Libra, Aquarius
||hot and dry
||choleric (irritable and changeable)
||Aries, Leo, Sagittarius
||cold and dry
||melancholic (sad and brooding)
||Taurus, Virgo, Capricorn
||cold and moist
||phlegmatic (apathetic and sluggish)
||Cancer, Scorpio, Pisces
Lyons comments that 'binary opposition is one of the most important principles governing the structure of languages' (Lyons 1977, 271). Saussure, of course, emphasized the differences between signs rather than their similarities. Opposites (or antonyms) clearly have a very practical function compared with synonyms: that of sorting. Roman Jakobson built on Saussure's work, proposing that linguistic units are bound together by a system of binary oppositions. Such oppositions are essential to the generation of meaning: the meaning of 'dark' is relative to the meaning of 'light'; 'form' is inconceivable except in relation to 'content'. It is an open question whether our tendency to think in opposites is determined by the prominence of oppositions in language or whether language merely reflects a universal human characteristic.
The various conventionally-linked terms with which we are familiar within a culture might more appropriately be described as paired 'contrasts', since they are not always direct 'opposites' (although their use often involves polarization). Distinctions can be made between various types of 'oppositions', perhaps the most important being the following:
This is basically a distinction between digital and analogue oppositions: digital differences are either/or; analogue distinctions are 'more-or-less'. We may note here that most of the oppositions in English are 'morphologically related' - that is, one term is a negative which is formed by the addition of a prefix such as un- or -in (e.g. formal/informal). Despite this, most of the commonly used oppositions in English (and in many other languages) are apparently morphologically unrelated (e.g. good/bad) (and thus more arbitrary). In English, most morphologically unrelated oppositions are comparative (gradable) and many morphologically related oppositions are not, but there many exceptions to this pattern - including terms which may be paired with another which is either morphologically related or unrelated (e.g. friendly/unfriendly and friendly/hostile). Positive and negative terms can be distinguished even in morphologically unrelated oppositions (such as good/bad) by such cues as their most common sequence - a point to which we will return (Lyons 1977, 275-277). There is no logical necessity for morphologically unrelated oppositions, as Syme explains to Winston in the dystopian novel Nineteen Eighty-Four written by George Orwell in 1949:
John Lyons suggests that the reason why we tend to use morphologically unrelated forms in comparative oppositions is to emphasize the semantic distinction involved: '"good" and "bad" are more obviously different lexemes than "friendly" and "unfriendly"' (Lyons 1977, 277). He adds that 'gradable opposites manifest the property of polarity more strikingly than do other opposites' (ibid., 279). Furthermore, in everyday discourse we frequently treat comparative terms as if they were discrete categories (ibid., 278). For whatever reasons we seem to favour categorization which is 'black and white'.
It is a feature of culture that binary oppositions come to seem 'natural' to members of
a culture. Many pairings of concepts (such as male/female and mind/body)
are familiar to members of a culture and may
seem commonsensical distinctions for everyday communicational purposes even
if they may be regarded as 'false dichotomies' in critical contexts.
Rudyard Kipling satirized the apparently universal tendency to divide the people we know
directly or indirectly into 'Us' and 'Them' ('We and They',
Kipling 1977, 289-290):
The opposition of self/other (or subject/object) is psychologically fundamental.
The neo-Freudian psychoanalyst Jacques Lacan wrote in 1957 (in 'The Insistence of the Letter in
the Unconscious') that 'the unconscious is structured like a language'
(cf. Lacan 1977, 159, 298).
The mind imposes some degree of constancy on the dynamic flux of experience by defining
'the Self' in relation to 'the Other'.
Initially, in the primal realm of 'the Real' (where there is no absence, loss or
lack), the infant has no centre of identity and experiences no clear boundaries
between itself and the external world.
The child emerges from the Real and enters
'the Imaginary' at the age of about six- to eighteen-months, before the acquisition of speech.
This is a private psychic realm in which the construction of the Self as subject is
initiated. In the realm of visual images, we find our sense of self reflected back by an Other
with whom we identify. For Lacan, this does not reflect a dichotomy between Self and Other,
because not only is Self always defined in terms of Other, but paradoxically, Self is
Other. He describes a defining moment in the Imaginary which he calls 'the mirror
phase', when seeing one's mirror image (and being told by one's mother, 'That's you!')
induces a strongly-defined illusion of a coherent and self-governing personal identity.
This marks the child's emergence from a matriarchal state of 'nature' into the patriarchal
order of culture.
As the child gains mastery within the pre-existing 'Symbolic
order' (the public domain of verbal language), language (which can be mentally manipulated)
helps to foster the individual's sense of a conscious Self residing in an 'internal world'
which is distinct from 'the world outside'.
However, a degree of individuality and autonomy is
surrendered to the constraints of linguistic conventions, and
the Self becomes a more fluid and ambiguous relational signifier rather than a
relatively fixed entity. Subjectivity is dynamically constructed through discourse.
Emile Benveniste argued that 'language is possible only because each speaker sets himself
up as a subject by referring to himself as "I" in his discourse. Because of this, "I" posits
another person, the one who, being as he is completely exterior to "me", becomes my echo
to whom I say "you" and who says "you" to me'... Neither of these terms can be considered
without the other; they are complementary... and at the same time they are reversible'
(Benveniste 1971, 225).
The entry into the Symbolic order may be illustrated with Freud's description
(in Beyond the Pleasure Principle, 1920) of the
fort-da game played by his grandson at the age of about eighteen months.
The child was alternately throwing away and pulling back a cotton-reel, whilst
attempting to say the words 'fort!' (gone away!) and 'da!' (there it is!) - thus creating the
shortest possible narrative form.
According to Freud this represented a symbolization of the
mother leaving and returning. It turns a paradigmatic substitution into an elementary
syntagm and demonstrates the lure of repetition and difference.
Its focus on
absence/presence has made it a favourite of post-structuralist theorists such as Lacan and
Derrida. It can stand for anything which we have lost or fear losing, and for the pleasure or
hope of its recovery. It is thus symbolic of the loss of (amongst other things) the
imagined oneness of being in the Imaginary.
Romantics may (at least retrospectively) identify with a childhood sense of growing separation
from that which can be described. They tend to echo the poet Shelley (1815) in a vision of primal
experience as a mystical sense of oneness, of being within a universal continuum: 'Let us
recollect our senses as children. What a distinct and intense apprehension we had of the world
and of ourselves... We less habitually distinguished all that we saw and felt from ourselves.
They seemed as it were to constitute one mass'
(Forman 1880, 261). The
Romantic sense of loss in mediation is perhaps most powerfully represented in Rousseau's
interpretation of our use of tools as involving the loss of a primal unity with the world.
Such Romantic visions emphasize the unity of the knower and the known.
Childhood or primal experience is
portrayed by Romantics as virtually 'unmediated'. And yet all but the most naive epistemology
suggests that our experience of the world is unavoidably mediated. Indeed, without the
separation of Self from Other there would be no 'me' who could hark back to a
pre-lapsarian myth of oneness.
'Male' and 'female' are not 'opposites', and yet cultural myths routinely encourage us to treat
them as such.
Guy Cook offers a simple example of how images of masculinity and femininity
can be generated through a series of binary oppositions in a literary text
(Cook 1992, 115).
He instances two consecutive speeches from the beginning of a scene in
Shakespeare's Romeo and Juliet:
All nice people, like us, are We
And everyone else is They:
But if you cross over the sea,
Instead of over the way,
You may end by (think of it!)
Looking on We
As only a sort of They!
The opposition of self/other (or subject/object) is psychologically fundamental. The neo-Freudian psychoanalyst Jacques Lacan wrote in 1957 (in 'The Insistence of the Letter in the Unconscious') that 'the unconscious is structured like a language' (cf. Lacan 1977, 159, 298). The mind imposes some degree of constancy on the dynamic flux of experience by defining 'the Self' in relation to 'the Other'. Initially, in the primal realm of 'the Real' (where there is no absence, loss or lack), the infant has no centre of identity and experiences no clear boundaries between itself and the external world.
The child emerges from the Real and enters 'the Imaginary' at the age of about six- to eighteen-months, before the acquisition of speech. This is a private psychic realm in which the construction of the Self as subject is initiated. In the realm of visual images, we find our sense of self reflected back by an Other with whom we identify. For Lacan, this does not reflect a dichotomy between Self and Other, because not only is Self always defined in terms of Other, but paradoxically, Self is Other. He describes a defining moment in the Imaginary which he calls 'the mirror phase', when seeing one's mirror image (and being told by one's mother, 'That's you!') induces a strongly-defined illusion of a coherent and self-governing personal identity. This marks the child's emergence from a matriarchal state of 'nature' into the patriarchal order of culture.
As the child gains mastery within the pre-existing 'Symbolic order' (the public domain of verbal language), language (which can be mentally manipulated) helps to foster the individual's sense of a conscious Self residing in an 'internal world' which is distinct from 'the world outside'. However, a degree of individuality and autonomy is surrendered to the constraints of linguistic conventions, and the Self becomes a more fluid and ambiguous relational signifier rather than a relatively fixed entity. Subjectivity is dynamically constructed through discourse. Emile Benveniste argued that 'language is possible only because each speaker sets himself up as a subject by referring to himself as "I" in his discourse. Because of this, "I" posits another person, the one who, being as he is completely exterior to "me", becomes my echo to whom I say "you" and who says "you" to me'... Neither of these terms can be considered without the other; they are complementary... and at the same time they are reversible' (Benveniste 1971, 225).
The entry into the Symbolic order may be illustrated with Freud's description (in Beyond the Pleasure Principle, 1920) of the fort-da game played by his grandson at the age of about eighteen months. The child was alternately throwing away and pulling back a cotton-reel, whilst attempting to say the words 'fort!' (gone away!) and 'da!' (there it is!) - thus creating the shortest possible narrative form. According to Freud this represented a symbolization of the mother leaving and returning. It turns a paradigmatic substitution into an elementary syntagm and demonstrates the lure of repetition and difference. Its focus on absence/presence has made it a favourite of post-structuralist theorists such as Lacan and Derrida. It can stand for anything which we have lost or fear losing, and for the pleasure or hope of its recovery. It is thus symbolic of the loss of (amongst other things) the imagined oneness of being in the Imaginary.
Romantics may (at least retrospectively) identify with a childhood sense of growing separation from that which can be described. They tend to echo the poet Shelley (1815) in a vision of primal experience as a mystical sense of oneness, of being within a universal continuum: 'Let us recollect our senses as children. What a distinct and intense apprehension we had of the world and of ourselves... We less habitually distinguished all that we saw and felt from ourselves. They seemed as it were to constitute one mass' (Forman 1880, 261). The Romantic sense of loss in mediation is perhaps most powerfully represented in Rousseau's interpretation of our use of tools as involving the loss of a primal unity with the world. Such Romantic visions emphasize the unity of the knower and the known. Childhood or primal experience is portrayed by Romantics as virtually 'unmediated'. And yet all but the most naive epistemology suggests that our experience of the world is unavoidably mediated. Indeed, without the separation of Self from Other there would be no 'me' who could hark back to a pre-lapsarian myth of oneness.
'Male' and 'female' are not 'opposites', and yet cultural myths routinely encourage us to treat them as such. Guy Cook offers a simple example of how images of masculinity and femininity can be generated through a series of binary oppositions in a literary text (Cook 1992, 115). He instances two consecutive speeches from the beginning of a scene in Shakespeare's Romeo and Juliet:
Wilt thou be gone? It is not yet near day;
It was the nightingale, and not the lark,
That pierc'd the fearful hollow of thine ear;
Nightly she sings on yond pomegranate tree.
Believe me, love, it was the nightingale.
It was the lark, the herald of the morn,
No nightingale. Look, love, what envious streaks
Do lace the severing clouds in yonder east;
Night's candles are burnt out, and jocund day
Stands tiptoe on the misty mountain tops.
I must be gone and live or stay and die.
||(Romeo and Juliet III, v)
Cook notes the following gendered oppositions:
Such oppositions tend to retreat to transparency in reading or watching the play. The gendered character of the echoes and parallels is consequently quite surprising when the text is submitted to this kind of analysis. And yet these oppositions do not seem to be purely analytical constructions. Indeed, we may also note that Juliet emphasizes sound whilst Romeo relies on vision (yet another stereotypically gendered association). Through the endless repetition of such subtle patterns - in countless variations - mythologies such as that of heterosexual romance are generated and sustained.
Paired signifiers are seen by structuralist theorists as part of the 'deep [or 'hidden'] structure' of texts, shaping the preferred reading. Such linkages seem to become aligned in some texts and codes so that additional 'vertical' relationships (such as male/mind, female/body) acquire apparent links of their own - as feminists and queer theorists have noted (Silverman 1983, 36; Grosz 1993, 195; Chaplin 1994, 11; Butler 1999, 17). As Kaja Silverman notes, 'a cultural code is a conceptual system which is organized around key oppositions and equations, in which a term like "woman" is defined in opposition to a term like "man", and in which each term is aligned with a cluster of symbolic attributes' (Silverman 1983, 36).
This notion can be traced to Claude Lévi-Strauss's discussion of analogical relationships which generate systems of meaning within classification systems. Structuralist theorists such as Lévi-Strauss have argued that binary oppositions form the basis of underlying 'classificatory systems' within cultures - constituting fundamental organizing metaphors and metonyms. He saw certain key binary oppositions as the invariants or universals of the human mind, cutting across cultural distinctions. Lévi-Strauss wrote:
Lévi-Strauss undertook synchronic studies of systems of cultural practices, seeking to identify underlying semantic oppositions in relation to such phenomena as myths, totemism and kinship rules. Individual myths and cultural practices defy interpretation, making sense only as a part of a system of differences and oppositions expressing fundamental reflections on the relationship of nature and culture. This is expressed in terms of the relations between humankind and various other phenomena, such as: animals, plants, supernatural beings, heavenly bodies, forms of food and so on. Certain binary distinctions based on the form of human body are universal and seem fundamental - notably male/female and right/left. 'Such natural pairs are invariably loaded with cultural significance - they are made into the prototype symbols of the good and the bad, the permitted and the forbidden' (Leach 1970, 44). Lévi-Strauss argues that within a culture 'analogical thought' leads to some oppositions (such as edible/inedible) being perceived as metaphorically resembling the 'similar differences' of other oppositions (such as native/foreign) (Lévi-Strauss 1974).
Lévi-Strauss reported three stages in his analytical method:
For Lévi-Strauss, myths represent a dreamlike working-over of a fundamental dilemma or contradiction within a culture which can be expressed in the form of a pair of oppositions. The development of the myth constitutes a repeated reframing of this tension through layers of paired opposites which are transformations of the primary pair. These layers begin with classifications based on physical perception and become increasingly more generalized. Claude Lévi-Strauss has demonstrated how cooking transforms Nature into Culture: South American myths oppose the raw to the cooked (Lévi-Strauss 1970). He comments on his theorizing: 'In order to construct this system of myths about cooking, we found ourselves obliged to use oppositions between terms all more or less drawn from sensory qualities: raw and cooked, fresh and rotten, and so forth. Now we find that the second step in our analysis reveals terms still opposed in pairs, but whose nature is different to the degree that they involve not so much a logic of qualities as one of forms: empty and full, container and contents, internal and external, included and excluded, etc.' (cited in Jameson 1972, 118-119).
In a major review of the anthropological literature, Lévi-Strauss famously and provocatively declared that 'exchange, as a total phenomenon, is from the first a total exchange, comprising food, manufactured objects and that most precious category of goods, women' (Lévi-Strauss 1969, 60-1). We have referred already to his reflections on the significance of our preparation of food. His observations on the social phenomenon of exchange are distinctive because he argued that exogamy (marrying outside the group) and more generally 'the relations between the sexes' are a form of communication (ibid., 493-4). Language, economics and sexuality - thus arguably the basis of all communication - draw on three fundamental oppositions: addressor/addressee; buyer/seller; masculine/feminine (Coward & Ellis 1977, 58). As Lévi-Strauss noted, social exchanges involve the exchange of 'social values' (Lévi-Strauss 1969, 62). The production of subject positions in relation to these key oppositions can be seen as a primary mechanism for the reproduction of society and its values.
Lévi-Strauss even turned his attention to the textual codes of literature in what is probably the most famous structuralist textual analysis of all. In collaboration with the linguist Roman Jakobson, he undertook an analysis of the sonnet 'Les Chats' by Charles Baudelaire (1821-1867). This involved a detailed outline of the oppositions of parts of speech, poetic forms, semantic features and so on (Lane 1970, 202-221). Since this is such a frequently-cited analysis, the poem and an English rendering are reproduced here for the reader's convenience. The commentators helpfully note, by the way, that L'Érèbe is a 'shady region bordering on Hell' and that Erebus is 'brother of the night' (Lane 1970, 213).
||Les amoureux fervents et les savants austères
||Fervent lovers and austere savants
||Aiment également, dans leur mûre saison,
||Cherish alike, in their mature season,
||Les chats puissants et doux, orgueil de la maison,
||Cats powerful and gentle, pride of the house,
||Qui comme eux sont frileux et comme eux sédentaires.
||Like them they feel the cold, like them are sedentary.
||Amis de la science et de la volupté,
||Friends of science and of sensuality,
||Ils cherchent le silence et l'horreur des ténèbres;
||They seek silence and the horror of the dark;
||L'Érèbe les eût pris pour ses courriers funèbres,
||Erebus would take them for his funereal couriers,
||S'ils pouvaient au servage incliner leur fierté.
||If they'd to servitude incline their pride.
||Ils prennent en songeant les nobles attitudes
||They take on when dreaming the noble postures
||Des grands sphinx allongés au fond des solitudes,
||Of great sphinxes stretched out in the depths of solitude,
||Qui semblent s'endormir dans un rêve sans fin;
||Seeming to sleep in a dream without end;
||Leurs reins féconds sont pleins d'étincelles magiques,
||Their fecund loins are full of magic sparks,
||Et des parcelles d'or, ainsi qu'un sable fin,
||And particles of gold, as well as fine sand,
||Étoilent vaguement leurs prunelles mystiques.
||Vaguely star their mystic pupils.
In a headnote to the paper, Lévi-Strauss notes that the poem consisted of 'superimposed levels: phonology, phonetics, syntax, prosody, semantics etc.' (Lane 1970, 202). The authors demonstrate that 'the different levels on which we touched blend, complement each other or combine' (ibid., 217). For instance, they note a link between the grammatical and semantic levels: 'All beings in the sonnet are masculine but the cats and their alter ego, les grands sphinx, are of an androgynous nature. This very ambiguity is emphasized throughout the sonnet by the paradoxical choice of feminine substantives [nouns] for so-called masculine rhymes' (ibid., 221). Here is a breakdown of the rhyme scheme which, together with the text, may assist interested readers to note patterns for themselves.
|Line||Rhyme word||English equivalent||Rhyme scheme||Rhyme form||Grammatical function||Singular/plural form|
We have already noted the asssociation of feminine nouns with masculine rhymes. In reflecting on patterns in this rhyme scheme, the reader may also notice, as Lévi-Strauss and Jakobson pointed out, the curious circumstance that in this sonnet 'all the substantives [nouns] are feminine' and that 'all feminine rhymes are plural' (Lane 1970, 205, 220). The authors argue that 'for Baudelaire, the image of the cat is closely linked to that of the woman', citing the association of 'puissants et doux' with women in other poetry. Lévi-Strauss and Jakobson emphasize the importance of binary oppositions. At the semantic level, other than what they see as 'the oscillation between male and female' in the poem, they argue that another key opposition is animate/inanimate. At a linguistic level a fundamental opposition is metaphor/metonymy. Again, readers may care to identify such oppositions for themselves. The authors argue that the poem seeks to 'resolve' the oppositions which it generates at various levels (ibid., 218-9). Whilst widely-cited, this analysis is also understandably criticized as arid by those whom structuralism leaves cold. Being an archetypical structuralist analysis, it confines itself to structural relations within the text (Riffaterre 1970).
More broadly, aesthetic 'movements' can be interpreted in terms of paradigms of characteristic oppositions. Each movement can be loosely identified in terms of a primary focus of interest: for instance, realism tends to be primarily oriented towards the world, neo-classicism towards the text and romanticism towards the author (which is not to suggest, of course, that such goals have not been shared by other movements). Such broad goals generate and reflect associated values. Within a particular movement, various oppositions constitute a palette of possibilities for critical theorists within the movement. For instance, the codes of romanticism are built upon various implicit or explicit articulations of such oppositions as: expressive/instrumental, feeling/thought, emotion/reason, spontaneity/deliberation, passion/calculation, inspiration/effort, genius/method, intensity/reflection, intuition/judgement, impulse/intention, unconsciousness/design, creativity/construction, originality/conventionality, creation/imitation, imagination/learning, dynamism/order, sincerity/facticity, natural/artificial and organic/mechanical. The alignment of some of these pairs generates further associations: for instance, an alignment of spontaneity/deliberation with sincerity/facticity equates spontaneity with sincerity. More indirectly, it may also associate their opposites, so that deliberation reflects insincerity or untruthfulness. Romantic literary theorists often proclaimed spontaneity in expressive writing to be a mark of sincerity, of truth to feeling - even when this ran counter to their own compositional practices (Chandler 1995, 49ff). Even within 'the same' aesthetic movement, various theorists construct their own frameworks, as is illustrated in Abrams' study of romantic literary theory (Abrams 1971). Each opposition (or combination of oppositions) involves an implicit contrast with the priorities and values of another aesthetic movement: thus (in accord with the Saussurean principle of negative differentiation) an aesthetic movement is defined by what it is not. The evolution of aesthetic movements can be seen as the working-out of tensions between such oppositions. Similarly, within textual analysis, it has been argued that the structure of particular texts (or myths) works to position the reader to privilege one set of values and meanings over the other. Sometimes such oppositions may appear to be resolved in favour of dominant ideologies but poststructuralists argue that tensions between them always remain unresolved.
One aesthetic movement, that of Surrealism, can be seen as centrally concerned with the resolution of opposites. Charles Forceville argues that:
As we shall see shortly, this Surrealist mission has much in common with poststructuralist goals.
Paradigmatic analysis has also been applied to popular culture. Exploring a basic opposition of wilderness/civilization, Jim Kitses analysed the film genre of the western in relation to a series of oppositions: individual/community; nature/culture; law/gun sheep/cattle (Kitses 1970). John Fiske makes considerable analytical use of such oppositions in relation to mass media texts (Fiske 1987). Umberto Eco analysed the James Bond novels in terms of a series of oppositions: Bond/villain; West/Soviet Union; anglo-saxon/other countries; ideals/cupidity; chance/planning; excess/moderation; perversion/innocence; loyalty/disloyalty (Eco 1966).
Binary oppositions can be traced even in visual images. Jean-Marie Floch compares and contrasts the logos of the two major computer companies, IBM and Apple, revealing their differences to be based on a series of associated binary oppositions, the most obvious of which are listed here (Floch 2000, 41). The contrast could hardly involve a clearer opposition. Appropriately, Apple's logo seems to be defined purely in opposition to the more established/establishment image of IBM.
A past chairman of the Apple Products division is quoted as saying, 'Our logo is a great mystery: it is a symbol of pleasure and knowledge, partially eaten away and displaying the colours of the rainbow, but not in the proper order. We couldn't wish for a more fitting logo: pleasure, knowledge, hope and anarchy' (Floch 2000, 54). Clearly, the bitten apple refers both to the story of the Tree of Knowledge in the Garden of Eden and to the association of IBM with the east coast and 'the Big Apple' of New York. The psychedelic mixed-up rainbow (green, yellow, orange, red, violet and blue) signifies the west coast hippie era of the 1960s, with its associations of idealism and 'doing your own thing'. Thus, despite representing a binary opposition to the IBM logo, the multi-coloured Apple logo seeks to signify a rejection of the binarism reflected in the 'black-and-white' (or rather monochrome) linearity of IBM's logo. Competing companies clearly need to establish distinct identities, and such identities are typically reflected in their logos. This example may tempt the reader to compare the visual identities of other competing corporations.
Oppositions are rarely equally weighted. The Russian linguist and semiotician Roman Jakobson introduced the theory of markedness: 'Every single constituent of any linguistic system is built on an opposition of two logical contradictories: the presence of an attribute ("markedness") in contraposition to its absence ("unmarkedness")' (cited in Lechte 1994, 62). The concept of markedness can be applied to the poles of a paradigmatic opposition: paired signs consist of an 'unmarked' and a 'marked' form. This applies, as we shall see, both at the level of the signifier and at the level of the signified. The 'marked' signifier is distinguished by some special semiotic feature (Nöth 1990, 76). In relation to linguistic signifiers, two characteristic features of marked forms are commonly identified: these relate to formal features and generic function. The more 'complex' form is marked, which typically involves both of the following features:
In English, linguistically unmarked forms include the present tense of verbs and the singular form of nouns. The active voice is normally unmarked, although in the restricted genre of traditional academic writing the passive voice is still often the unmarked form.
The markedness of linguistic signs includes semantic marking: a marked or unmarked status applies not only to signifiers but also to signifieds. According to 'the binary thesis' 'a signified's content is determined by a series of binary contrasts in which one term is marked and the other unmarked' (Holdcroft 1991, 127). With morphologically related pairings there is an obvious relation between formal and semantic marking, and John Lyons suggests that distributional marking in oppositions is probably determined by semantic marking (Lyons 1977, 307). One form of semantic marking relates to specificity. The unmarked term is often used as a generic term whilst the marked term is used in a more specific sense. General references to humanity used to use the term 'Man' (which in this sense was not intended to be sex-specific), and of course the word 'he' has long been used generically. In English the female category is generally marked in relation to the male, a point not lost on feminist theorists (Clark & Clark 1977, 524). Lyons notes, however, that it is not always the female term which is marked - he refers to several farmyard animals as exceptions - bull, cock, ram and drake - suggesting that this is perhaps because such animals are normally reared in smaller numbers (Lyons 1977, 308).
Where terms are paired the pairing is rarely symmetrical but rather hierarchical. With apologies to George Orwell we might coin the phrase that 'all signifieds are equal, but some are more equal than others'. With many of the familiarly paired terms, the two signifieds are accorded different values. The unmarked term is primary, being given precedence and priority, whilst the marked term is treated as secondary or even suppressed and excluded as an 'absent signifier'. When morphological cues (such as un- or -in) are lacking, the 'preferred sequence' or most common order of paired terms usually distinguishes the first as a semantically positive term and the second as a negative one (Lyons 1977, 276; Malkiel 1968). 'Term B' is referred to by some theorists as being produced as an 'effect' of 'Term A'. The unmarked term is presented as fundamental and originative whilst the marked term 'is conceived in relation to it' as derivative, dependent, subordinate, supplemental or ancillary (Culler 1985, 112; Adams 1989, 142). This framing ignores the fact that the unmarked term is logically and structurally dependent on the marked term to lend it substance. Even the arch-structuralist Lévi-Strauss acknowledged that 'the very notion of opposition implies that the two forms were originally conceived of as complementary terms, forming a part of the same classification' (in Lane 1970, 202). Derrida demonstrated that within the oppositional logic of binarism neither of the terms (or concepts) makes sense without the other. This is what he calls 'the logic of supplementarity': the 'secondary' term which is represented as 'marginal' and external is in fact constitutive of the 'primary' term and essential to it (Derrida 1976). The unmarked term is defined by what it seeks to exclude. Consequently, the boundaries of foundational oppositions, seemingly 'absolute', have to be policed because 'transgressions' are inevitable (Eagleton 1983, 133).
In the pairing of oppositions or contraries, Term B is defined relationally rather than substantively. The linguistic marking of signifiers in many of these pairings is referred to as 'privative' - consisting of suffixes or prefixes signifying lack or absence - e.g. non-, un- or -less. In such cases, Term B is defined by negation - being everything that Term A is not. For example, when we refer to 'non-verbal communication', the very label defines such a mode of communication only in negative relation to 'verbal communication'. Indeed, the unmarked term is not merely neutral but implicitly positive in contrast to the negative connotations of the marked term. For the French psychoanalyst Jacques Lacan the marked term in the pairing of men/women is negatively defined within 'the symbolic order' in terms of the absence or lack of a privileged signifier associated with control and power - the phallus (though see feminist critiques of Lacan's phallocentrism, e.g. Lovell 1983, 44-45). The association of the marked term with absence and lack is of course problematized by those who have noted the irony that the dependence of Term A on Term B can be seen as reflecting a lack on the part of the unmarked term (Fuss 1991, 3).
The unmarked form is typically dominant (e.g. statistically within a text or corpus) and therefore seems to be 'neutral', 'normal' and 'natural'. It is thus 'transparent' - drawing no attention to its invisibly privileged status, whilst the deviance of the marked form is salient. Where it is not totally excluded, the 'marked' form is foregrounded - presented as 'different'; it is 'out of the ordinary' - an extraordinary deviational 'special case' which is something other than the standard or default form of the unmarked term (Nöth 1990, 76; Culler 1989, 271). Unmarked/marked may thus be read as norm/deviation. It is notable that empirical studies have demonstrated that cognitive processing is more difficult with marked terms than with unmarked terms (Clark & Clark 1977). Marked forms take longer to recognize and process and more errors are made with these forms.
On the limited evidence from frequency counts of explicit verbal pairings in written text, I would suggest that whilst it is very common for one term in such pairings to be marked, in some instances there is not a clearly marked term. For instance, in general usage there seems to be no inbuilt preference for one term in a pairing such as old/young (one is just as likely to encounter young/old). Furthermore, the extent to which a term is marked is variable. Some terms seem to be far more clearly marked than others: frequency counts based on texts on the World-Wide Web suggest that in the pairing public/private, for instance, private is very clearly the marked term (accorded secondary status). How strongly a term is marked also depends on contextual frameworks such as genres and sociolects, and in some contexts a pairing may be very deliberately and explicitly reversed when an interest group seeks to challenge the ideological priorities which the markedness may be taken to reflect. Not all of the pairs listed will seem to be 'the right way round' to everyone - you may find it interesting to identify which ones seem counter-intuitive to you and to speculate as to why this seems so.
However 'natural' familiar dichotomies and their markedness may seem, their historical origins or phases of dominance can often be traced. For instance, perhaps the most influential dualism in the history of Western civilization can be attributed primarily to the philosopher René Descartes (1596-1650) who divided reality into two distinct ontological substances - mind and body. This distinction insists on the separation of an external or 'real' world from an internal or 'mental' one, the first being material and the second non-material. It created the poles of objectivity and subjectivity and fostered the illusion that 'I' can be distinguished from my body. Furthermore, Descartes' rationalist declaration that 'I think, therefore I am' encouraged the privileging of mind over body. He presented the subject as an autonomous individual with an ontological status prior to social structures (a notion rejected by poststructural theorists). He established the enduring assumption of the independence of the knower from the known. Cartesian dualism also underpins a host of associated and aligned dichotomies such as reason/emotion, male/female, true/false, fact/fiction, public/private, self/other and human/animal. Indeed, many feminist theorists lay a great deal of blame at Descartes' door for the orchestration of the ontological framework of patriarchal discourse. One of the most influential of theorists who have sought to study the ways in which reality is constructed and maintained within discourse by such dominant frameworks is the French historian of ideas, Michel Foucault, who focused on the analysis of 'discursive formations' in specific historical and socio-cultural contexts (Foucault 1970; Foucault 1974).
The strategy of 'deconstruction' which was adopted by the post-structuralist philosopher Jacques Derrida (1976) sought to challenge the phonocentric privileging of speech over writing in western culture and to demonstrate the instability of this opposition (Derrida 1976; Derrida 1978). Derrida also challenged the privileging of the signified over the signifier, seeing it as a perpetuation of the traditional opposition of matter and spirit or substance and thought. He noted that within such discourse the material form is always subordinated to the less material form. Derrida sought to blur the distinction between signifier and signified, insisting that 'the signified always already functions as a signifier' (Derrida 1976, 7). He similarly challenged other loaded oppositions such as presence over absence, nature over culture, masculine over feminine and literal over metaphorical. Other 'critical theorists' have similarly sought to 'valorize term B' in the semiotic analysis of textual representations, though most are content with simply reversing the valorization rather than more radically seeking to destabilize the oppositional framework. This strategy is reflected in the way in which some activists in minority groups have hijacked the dominant language of the majority - as in the case of a campaign against homophobia which was launched by the Terrence Higgins Trust in the UK in September 1999 under the slogan 'It's prejudice that's queer'. The posters used neatly inverted heterosexist notions by substituting homophobia for homosexuality: 'I can't stand homophobes, especially when they flaunt it'; 'My son is homophobic, but I hope it's just a phase'; and 'homophobes shouldn't be left alone with kids'. This strategy of ironic reversal had been foreshadowed in the wittily subversive formulation that 'we don't yet know what causes heterosexuality' (found in gay webpages).
Following on from Derrida's deconstruction of Saussure's Course in General Linguistics, Robert Hodge and Gunther Kress have offered a useful visual mapping of Saussure's model of semiotics in terms of its own explicit oppositions. The diagram shown below is based on theirs. The leftmost terms represent those which were privileged by Saussure whilst those on the right represent those which he marginalizes in the Course. Seeking to revalorize those terms which Saussure had devalorized, Hodge and Kress build their own more explicitly social and materialist framework for semiotics on 'the contents of Saussure's rubbish bin'. Their agenda for an 'alternative semiotics' is based on:
The concept of markedness can be applied more broadly than simply to paradigmatic pairings of words or concepts. Whether in textual or social practices, the choice of a marked form 'makes a statement'. Where a text deviates from conventional expectations it is 'marked'. Conventional, or 'over-coded' text (which follows a fairly predictable formula) is unmarked whereas unconventional or 'under-coded' text is marked. Marked or under-coded text requires the interpreter to do more interpretative work.
The existence of marked forms is not simply a structural feature of semiotic systems. Kathryn Woodward argues that 'it is through the marking out of... differences that social order is produced and maintained' (Woodward 1997, 33). Unmarked forms reflect the naturalization of dominant cultural values. The French feminist Hélène Cixous has emphasized the gendered character of binary oppositions, which are consistently weighted in favour of the male (cited in Woodward 1997, 36 and Allen 2000, 152). As Trevor Millum notes:
Applying the concept of marked forms to mass media genres, Merris Griffiths, then one of my own research students, examined the production and editing styles of television advertisements for toys. Her findings showed that the style of advertisements aimed primarily at boys had far more in common with those aimed at a mixed audience than with those aimed at girls, making 'girls' advertisements' the marked category in commercials for toys. Notably, the girls' ads had significantly longer shots, significantly more dissolves (fade out/fade in of shot over shot), less long shots and more close-ups, less low shots, more level shots and less overhead shots. The gender-differentiated use of production features which characterized these children’s commercials reflected a series of binary oppositions - fast vs. slow, abrupt vs. gradual, excited vs. calm, active vs. passive, detached vs. involved. Their close association in such ads led them to line up consistently together as ‘masculine’ vs. ‘feminine’ qualities. The 'relative autonomy' of formal features in commercials seems likely to function as a constant symbolic reaffirmation of the broader cultural stereotypes which associate such qualities with gender - especially when accompanied by gender-stereotyped content. Readers may care to reflect on the way in which 'dark goods' and 'light goods' have traditionally been sold in high-street electrical shops. Dark goods such as televisions, video-recorders, camcorders and sound-systems were primarily targetted at men and the sales staff focused on technical specifications. Light goods such as refrigerators, washing-machines and cookers were targetted at women and the sales staff focused on appearance. The extent to which this particular pattern still survives in your own locality may be checked by some investigative 'window-shopping'.
'Binarism' has been defined as 'the passion of those who tend to see everything as divided into two categories' (Hervey 1982, 24). There is a delightfully ironic quip (variously attributed) that 'The world is divided into those who divide people into two types, and those who don't'. The interpretive usefulness of simple dichotomies is often challenged on the basis that life and (perhaps by a misleading 'realist' analogy) texts are 'seamless webs' and thus better described in terms of continua. But it is useful to remind ourselves that any interpretive framework cuts up its material into manageable chunks. The test of its appropriateness can surely only be assessed in terms of whether it advances our understanding of the phenomenon in question.
The structuralist semiotician Algirdas Greimas introduced the semiotic square (which he adapted from the 'logical square' of scholastic philosophy) as a means of analysing paired concepts more fully (Greimas 1987, xiv, 49). The semiotic square is intended to map the logical conjunctions and disjunctions relating key semantic features in a text. Fredric Jameson notes that 'the entire mechanism... is capable of generating at least ten conceivable positions out of a rudimentary binary opposition' (in Greimas 1987, xiv). Whilst this suggests that the possibilities for signification in a semiotic system are richer than the either/or of binary logic, but that they are nevertheless subject to 'semiotic constraints' - 'deep structures' providing basic axes of signification.
The symbols S1, S2, Not S1 and Not S2 represent positions within the system which may be occupied by concrete or abstract notions. The double-headed arrows represent bilateral relationships. The upper corners of the Greimasian square represent an opposition between S1 and S2 (e.g. white and black). The lower corners represent positions which are not accounted for in simple binary oppositions: Not S2 and Not S1 (e.g. non-white and non-black). Not S1 consists of more than simply S2 (e.g. that which is not white is not necessarily black). In the horizontal relationships represent an opposition between each of the left-hand terms (S1 and Not S2) and its paired right-hand term (Not S1 and S2). The terms at the top (S1, S2) represent 'presences', whilst their companion terms (Not S1 and Not S2) represent 'absences'. The vertical relationships of 'implication' offer us an alternative conceptual synthesis of S1 with Not S2 and of S2 with Not S1 (e.g. of white with not-black or of black with not-white). Greimas refers to the relationships between the four positions as: contrariety or opposition (S1/S2); complementarity or implication (S1/Not S2 and S2/Not S1); and contradiction (S1/Not S1 and S2/Not S2). Varda Langholz Leymore offers an illustrative example of the linked terms 'beautiful' and 'ugly'. In the semiotic square the four related terms (clockwise) would be 'beautiful', 'ugly', 'not beautiful' and 'not ugly'. The initial pair is not simply a binary opposition because 'something which is not beautiful is not necessarily ugly and vice versa a thing which is not ugly is not necessarily beautiful' (Langholz Leymore 1975, 29). The same framework can be productively applied to many other paired terms, such as 'thin' and 'fat'.
Occupying a position within such as framework invests a sign with meanings.
The semiotic square can be used to highlight 'hidden' underlying themes in a text
or practice. Using a slightly adapted version of the square shown here,
Fredric Jameson outlines how it might be applied to Charles Dickens'
novel, Hard Times.
In his foreword to an English translation of a book by Greimas, Jameson reflects on his own use of the technique. He suggests that the analyst should begin by provisionally listing all of the entities to be coordinated and that even apparently marginal entities should be on this initial list. He notes that even the order of the terms in the primary opposition is crucial: we have already seen how the first term in such pairings is typically privileged. He adds that ' the four primary terms... need to be conceived polysemically, each one carrying within it its own range of synonyms... such that... each of the four primary terms threatens to yawn open into its own fourfold system' (in Greimas 1987, xv-xvi). Jameson suggests that Not S2, the negation of the negation, 'is always the most critical position and the one that remains open or empty for the longest time, for its identification completes the process and in that sense constitutes the most creative act of the construction' (ibid., xvi). Using the earlier example of aesthetic movements and their dominant focuses, the reader might find it interesting to apply the semiotic square to these. To recap, it was suggested that realism tends to be primarily oriented towards the world, neo-classicism towards the text and romanticism towards the author. We may assign the concepts of world, text and author to three corners of the square - a fourth term is conspicuous by its absence. Jameson's caveats about the order and formulation of terms may be useful here.
Turning to other contexts, in relation to children's toys Dan Fleming offers an accessible application of the semiotic square (Fleming 1996, 147ff). Gilles Marion has used the Greimasian square to suggest four purposes in communicating through clothing: wanting to be seen; not wanting to be seen; wanting not to be seen; and not wanting not to be seen (cited in draft publication by David Mick). Most recently, Jean-Marie Floch has used the grid to illustrate an interesting exploration of the 'consumption values' represented by Habitat and Ikea furniture (Floch 2000, 116-144). However, the Greimasian analysis of texts in terms of the semiotic square has been criticized as easily leading to reductionist and programmatic decodings. Worse still, some theorists seem to use the square as little more than an objective-looking framework which gives the appearance of coherence and grand theory to loose argument and highly subjective opinions.
Critics of structuralist analysis note that binary oppositions need not only to be related to one another and interpreted, but also to be contextualised in terms of the social systems which give rise to texts (Buxton 1990, 12). Those who use this structuralist approach sometimes claim to be analysing the 'latent meaning' in a text: what it is 'really' about. Unfortunately, such approaches typically understate the subjectivity of the interpreter's framework. Illuminating as they may sometimes be, any inexplicit oppositions which are identified are in the mind of the interpreter rather than contained within the text itself (Culler 1975; Adams 1989, 139). Yet another objection is that 'the question of whether categories like sacred/profane and happiness/misery are psychologically real in any meaningful sense is not posed and the internal logic of structuralism would suggest it need not be posed' (Young 1990, 184).