Intro      Writing       Les mots et les noms       The unspeakable Other       Painting       Mathematics       The Thing Itself       Bibliography

Les mots et les noms

The dilemma posed by Still Life, which faces Byatt and Alexander in the novel, is not a new one and is currently established at the heart of post-structural and deconstructive literary theory [1]. The ambition neatly to replicate things with words is at best optimistic folly: witness the opening sentences of Ferdinand de Saussure's 1916 Course in General Linguistics: 'Some people regard language, when reduced to its elements, as a naming-process only — a list of words, each corresponding to the thing that it names… This conception… assumes that ready-made ideas exist before words' (65). It is a 'rather naïve approach' (ibid), a logocentricism now reviled.

The text is explicitly aware that this view of language has fallen into disrepute, and Raphael Faber's lecture on Mallarmé, 'Les mots et les noms' [2], provides a succinct summary of its history. This can be divided into three stages. Firstly, early religious notions had language established by divine right: language 'had once been thought of as Adamic naming' (SL, 245) says Raphael in his lecture, and elsewhere in the novel this is repeated: 'Adam in the Garden named the flora and fauna', (SL, 365); 'In the Renaissance there was a belief that language was a God-given symbolic system' (SL, 366). Secondly, this absolute unity of word and thing began to crumble with increasing self-consciousness about language: Raphael gives 'a neat and brilliant history of words unfitting themselves from objects' (SL, 245). Later, in his rooms, Frederica says, ' 'In the Mallarmé lecture you said that we can't reproduce — le bois intrinsèque et dense des arbres [3].'' (SL, 257) Although the earlier view is the religious one, this later schism is called a 'myth' in the text: 'So powerful was the myth of this divorce of names from things that it was almost impossible not to see Shakespeare's verse as qualitatively different from Keats's in this way…' (SL, 345) Finally, one is left with absence: unable to reproduce the world of things in words, we are left with 'a garden once full of imagined blooms, colours, lights, solidity, now inhabited by the ghosts of these things.' (SL, 245-6)

While this is also its prehistory, post-structuralism postdates the setting of this lecture by a decade, obliquely referenced later in the same chapter: 'It is now fashionable intellectually to write of Desire and the Other, of the desire of a text for itself, or for another, of language for an ungraspable referent.' (SL, 256) The late seventies and opening years of the eighties were the time, in literary theory, of textual desire; references to 'the desire of a text', to language, and to the referent recall Barthes, a key post-structuralist, in Plaisir du Texte: for example, 'The text you write must prove to me that it desires me' (Barthes, 1973: 13), and 'The text is a fetishist object and this fetish desires me' (ibid: 45).

One must not confuse the terminology, however, as 'an ungraspable referent' appears to do: where Byatt's novel speaks of 'words' and 'things', post-structuralism recognises three terms, 'the triple cluster of referent, signified, and signifier' (Kristeva, 1974: 68). The signifier is the sound-image or written word, such as 'tri:' or tree. The signified is the associated concept. Taken together, these two constitute the linguistic sign. The referent, meanwhile, is deictic, like the words 'you', or 'tomorrow'; to bring the referent into the essay I would not only have to identify the specific tree I am watching wave its branches through the window, but uproot that very tree and insert it into this paper. The referent is not ungraspable; it is the very thing that can be grasped; it is material, extralinguistic. The signified, on the other hand, is psychological, a concept. A failure to distinguish between the two lies at the heart of Still Life's contradictions: failing to separate the idea of something, the signified, from an individual material item, the referent, leads one to fall into the trap of thinking that neat equivalence is possible, that 'ready-made ideas exist before words' (Saussure, 1916: 65). On the contrary, as Saussure established and post-structuralism later enshrined, 'Without language, thought is a vague, uncharted nebula. There are no pre-existing ideas, and nothing is distinct before the appearance of language.' [4] (ibid: 112) Further, within a language, words 'limit each other reciprocally' (ibid: 116): 'concepts are purely differential and defined not by their positive content but negatively by their relations with the other terms of the system.' (ibid: 117) Post-structural theory takes this negativity further than did Saussure, insisting that it is the difference between signifiers that constitutes meaning: the signified, as formless concept, does not have existence prior to its signifier. Hence Barthes is able to speak of 'a galaxy of signifiers' (1970: 12) producing meaning, and Kristeva can say that 'the Saussurean cleavage (signifier/signified) is forever unbridgeable' (1977: 165).

Whereas before it might have seemed obvious that 'Words give us a clear, ordinary little picture of things' , post-structuralism brings to the foreground how language creates that 'little image', hemming signifiers in with other signifiers. Alexander's aspirations look very different from this vantage point: he had hoped to write verse in which 'a yellow chair was the thing itself, a yellow chair' (SL, 2) — but what is meant by 'the thing itself, a yellow chair'? Not the referent, which is extralinguistic, one cannot write a chair. The signified then, the concept — but 'initially the concept is nothing,… it is a value determined by its relations with other similar values' (Saussure, 1916: 117). Having reached for some 'thing itself' which is more than the signifier, but can be contained by a signifier, but will not insert itself within the network of signifiers, Alexander is right to conclude 'it couldn't be done.' (SL, 2)

Byatt's hope of writing the novel 'innocently, without recourse or reference to other people's thoughts' (SL, 131) is equally doomed by Saussurean linguistics even before simile, metaphor, or figure of speech enter the text. It is language that carves up the nebula, the formless chaos, that will become thought, and language is inherently social, it 'never exists apart from the social fact' (Saussure, 1916: 77); it requires and exists through a linguistic community. By using language we are using used thoughts. Post-structuralism theorises the social foundation of language much more richly, notably in the writings of Lacan and Kristeva, but we need not look even so far to understand why Byatt's plan 'turned out to be impossible: one cannot think at all without a recognition and realignment of ways of thinking and seeing we have learned over time.' (SL, 131)

The contradictions in the text now come clear, synthesised by post-structural theory. Metaphor, simile, and analogy, in the traditional sense, are avoidable; I need not compare thee to a summer's day; one need not see Marcus's interest in the cleistogamous flower as psychoanalytically, causally linked to his own sexuality. The metaphoricity which the text finds inherent in etymology is a red herring: synchronously, the relationship between a signifier and its signified is arbitrary; there is no essential 'treeness' to the word 'tree', and so the claim that there is metaphoric meaning inherent in every word is nonsense. Diachronically, however, language is also the fossilisation of different stages in the life of a linguistic community. It provides evidence of past thought, and so unravelling 'panick-grass' from its Latin name, Panicum, from panis, bread, is meaningful [5]. What is at work in this supposed metaphoricity is a combination of two aspects of language: on the synchronous level, signifiers existing within the network of signifiers; and on the diachronic level, the effect of language being not ours to invent but passed on to us from a linguistic community. This must be emphasised: signifiers exist in a network of signifiers; language is always already social. Thus we cannot escape association, the slippage of meaning along the chain of signifiers, though we may evade explicit metaphor; thus we cannot avoid other people's thoughts, though we may not cite another person.

We have resolved the text's contradictions, and made clear the absurdity of its aims — representation of the thing itself, really! And what 'thing' would that be? — through the basic tenets of post-structuralism. It comes as a shock, therefore, to discover that this wise theory is itself obsessed with nothing other than the same problem: the representation of the thing itself — only this time, we dare not even name what we are looking for.

---

1 Chronologically, post-structuralism slightly preceeds deconstruction, which is the narrower term of the two. The arguments foundational to post-structuralism usually apply equally to deconstruction. For simplicity's sake I will henceforth use the term 'post-structuralism' except when referring to a specifically deconstructive practice. 

2 'Names and Nouns' is Byatt's translation in the text (SL, 245)

3 'the dense, intrinsic wood of the trees'

4 This, and the relationship between signifier and signified, is elaborated in Part Two, Chapter Four, of Saussure's Course in General Linguistics. As Saussure notes, 'If words stood for pre-existing concepts, they would all have exact equivalents in meaning from one language to the next' (116), which is not so.

5 Sunflower need not be interpreted diachronically, as both sun and flower are current English words. Saussure discusses this kind of relationship between similar words as part of the network of meaning, and notes that 'The value of just any term is accordingly determined by its environment' (Saussure, 1916: 116).