Roger Poole on the seductions and dangers of structuralism

  • Structuralism and Since: From Lévi-Strauss to Derrida edited by John Sturrock
    Oxford, 190 pp, £5.50, January 1980, ISBN 0 19 215839 2

John Sturrock’s little book is the best single guide to its subject that has yet appeared. Structuralism and Since demands, though, that its title be taken literally. It traces, technically and without concession to idle curiosity, the course of ‘structuralism’ in its modern phase – from the moment when it achieved new importance in the work of Lévi-Strauss in the early Sixties, through its development and extensions in the work of Roland Barthes, Michel Foucault, Jacques Lacan and Jacques Derrida, to its present position, which is known either as ‘post-structuralism’ or simply as ‘deconstruction’.

It is significant that it should bear the date 1979, though, for it is very much a book which closes and summarises the ideas of two decades, not a book which opens a new decade of inquiry. The clarity and incisiveness with which all five experts write on their chosen author is due in part to this time factor: 20 years have passed, and the stirring and often ambiguous movements of the Sixties and Seventies can now get what seems like a definitive ‘placing’. Indeed, the sheer expertise of these five essays, the sense of there being nothing significant to add after a 20-page summary, makes one realise how much time has passed since Foucault published, in 1966, what was then an almost incomprehensible book, Les Mots et les Choses, and since Lacan published, in the same year, what was agreed on all sides to be an absolutely incomprehensible book, Ecrits.

To read Dan Sperber on Lévi-Strauss, Hayden White on Foucault, or Malcolm Bowie on Lacan, is to realise that the terrain, until recently so inhospitable, has been expertly mapped, the rough places made plain, and pleasant bowers and seats arranged for the ascending traveller at just the conceptual altitude where he will feel the need for them. The subject of structuralism, and its offshoot, ‘post-structuralism’, has been mastered. It has become a matter of knowledge almost, rather than opinion. Brilliantly expounded, with cracking pace and unflappable self-confidence, the book is a mine of information and an indispensable primer to anyone who comes to the subject fresh and ready to make a new conquest, just as it is an extraordinarily adept configuration of the field for those who may come to the subject weary from old failures to understand, or convinced that it is either marginal or obscure.

What is strikingly original is that the five expositors, each of them a well-known expert in his field, have chosen to expound the central contentions (the ‘facts’, the ‘subject’, the ‘materials’ which have been the concern of the five mages) in terms of the rhetorical assumptions and tropes that the mages have used when writing. It was one of the central contentions of structuralism in its early days that what we had to study was not so much discrete ‘things’ in isolation, but bundles of relationships, binary pairs, transformations within a system, and so on. Consequently, each one of the five mages developed a special or characteristic rhetorical cast, slant or habit with respect to his ‘materials’, and it is in terms of these characteristic rhetorical slants that the story of their achievement is now told. And it makes a great deal of sense.’ Meta-treatments are here subject to meta-description, and the gain in ‘primary’ comprehension is enormous.

It is very fitting that Dan Sperber, approaching the vast oeuvre of Claude Lévi-Strauss, should begin with a simple matter, the trope (which is almost a habit of thought) which Lévi-Strauss uses most consistently in his work: ‘One of his favourite figures is a fairly rare form of “abstract for concrete” substitution, or synecdoche, whereby a quality is used as an equivalent for the person or thing which possesses it: a calabash is referred to as a “container”, the beverage in it as the “contained”.’ But this is only part of the story of this ‘tropical’ way of processing thought: ‘These abstract synecdoches become the instruments of a second favourite figure of speech: antithesis ... The more elaborate synecdoches enable the antithesis to be further developed into a chiasmus, or “symmetrical inversion” in Lévi-Strauss’s terms.’

How often, in reading through one of his analyses, has one not been thrown by some sudden swerve, some unexpected transformation in the structure? This goes a long way to explaining why. Dan Sperber suggests that there is some particularly satisfying ‘fit’ between this rhetorical cast of mind and the thought processes of ‘Untamed Thinking’: ‘There is an interesting relationship between Lévi-Strauss’s way of thinking and that of people who tell myths. It is not one of similarity but complementarity: Lévi-Strauss tends, as we have seen, to represent a concrete object by one of its abstract properties; this makes him particularly apt at unravelling the thought of people who tend, contrarily, to represent an abstract property by some concrete object which possesses it, i.e. people given to using a “concrete for abstract” form of synecdoche.’

This is, of course, not all there is to it, and Dan Sperber has some hard things to say about this use of synecdoche. For instance, later on, he suggests that Lévi-Strauss achieves some of his more spectacular results through an abuse of synecdoche. He also claims that Lévi-Strauss does not often refer to a formula for myth which he says is basic to it, and has no interest in finding the ‘minimal elements’ of myth, which other, less inspired, workers could then move from as the basis of a unified scientific enterprise. He has not defined ‘mythemes’, nor has he put forward a grammar of myths. But with his pointing-up of synecdoche, Dan Sperber has illuminated a great deal.

Likewise, when Hayden White comes to account for the kind, or status, of the analysis that Michel Foucault has been producing (which is not philosophy, not sociology, not philology), he finds it useful to isolate a form of analysis which Foucault almost constantly uses – catachresis. ‘Wherever Foucault looks, he finds nothing but discourse: and wherever discourse arises, he finds a struggle between those who claim the “right” to discourse and those who are denied the right to their own discourse.’ So much, so to speak, for the ‘what’ of Foucault’s interests: the analysis of what he calls ‘the discourse of power’. But what of the ‘how’?

It is not surprising, then, that Foucault’s own discourse tends to assume the form of what the critic Northrop Frye calls the ‘existential projection’ of a rhetorical trope into a metaphysics. This rhetorical trope is catachresis, and Foucault’s style not only displays a profusion of the various figures sanctioned by catachresis, such as paradox, oxymoron, chiasmus, hysteron proteron, metalepsis, prolepsis, antonomasia, paronomasia, antiphrasis, hyperbole, litotes, irony, and so on; his own discourse stands as an abuse of everything for which ‘normal’ or ‘proper’ discourse stands. It looks like history, like philosophy, like criticism, but it stands over against these discourses as ironic antithesis ... Foucault’s ‘discourse about discourses’ seeks to effect the dissolution of Discourse. This is why I call it catachretic.

But the basic ‘discursive’ shifts within the discourse of history itself will, of course, involve figures other than this: metaphor, metonymy, synecdoche, irony will usually be the tropes through which a ‘discourse’ defines for itself the relation perceived (in any given age) between a ‘sign’ and the ‘reality’ or ‘thing’ to which the ‘sign’ is taken to correspond. This involves a fascinating mapping of conceptual space in any given age, according to whether the ‘sign’ and the ‘thing’ are perceived as distant, close, contained within each other, excluded from each other, parodying each other, and so on. And there will be a trope for each. But all these tropes, for Foucault himself, are part of a vast historical catachresis, a bacchanal of misnaming, a naivety without parallel.

Once again, the study of the rhetoric of an investigation is shown to be the proper instrument of decipherment. Similarly, Malcolm Bowie carries out the almost incredible feat of making Lacan comprehensible through an analysis of the rhetorical armature of his prose. The debts to Saussure (the Signifier is privileged over the signified in no uncertain way, S/s means what it says!), and the debts to Jakobson, are presented in terms of the rhetorical use which Lacan can make of them. From Saussure he derives an enormous capital to invest in his idea of the arbitrariness of language; from Jakobson, the twin tropes of metaphor and metonymy. The Freudian triad, Id-Ego-Superego, allows Lacan to vary rhetorically (almost to infinity) the possible collusions within his own substitute triad, the Symbolic, the Imaginary and the Real. Bowie also shows how the manufacture of puns, quibbles and double-entendres in Lacan’s own prose, as well as the enormous importance given to home-made symbols, is itself a rhetorical praxis which seeks to quell for ever the naive belief that, under all this language, there is someone who is trying to say something. Once again, as with Foucault, it is the list of tropes that Lacan finds in the rhetoric of the unconscious which is so striking: ‘Periphrasis, hyperbaton, ellipsis, suspension, anticipation, retraction, negation, digression, irony, these are the figures of style (Quintilian’s figurae sententiarum); as catachresis, litotes, antonomasia, hypotyposis are the tropes whose terms suggest themselves as most proper for the labelling of these mechanisms. Can one really see these as mere figures of speech when it is the figures themselves that are the active principle of the rhetoric of the discourse that the analysand in fact utters?

Jonathan Culler also, when he comes to treat of the arch-deconstructivist Jacques Derrida, finds himself constrained to adopt a ‘double strategy’ in order to describe, without misrepresenting, Derrida’s work. ‘Even the most subtle and scrupulous formulations would not avoid the problem of “misrepresentation”,’ he writes, and readers of Derrida will sympathise with him, knowing what he implies. He decides that a technique of alternating theoretical claims and close examination of texts will best bring out the kind of work that Derrida does. He starts, though, by giving a brief and excellent introduction to what the various published books are actually about, and listing them in order of difficulty. But then he has to plunge in: ‘What is the nature of the project to which all these writings contribute? Derrida’s readings of various texts and the constructions of his own texts are explorations of Western “logocentrism”. The “metaphysics of presence”, which these texts can be shown simultaneously to affirm and to undermine, is the only metaphysics we know, and underlies all our thinking; but it can be shown to give rise to paradoxes that challenge its coherence and consistency and therefore challenge the possibility of determining or defining being as presence.’

Having started on this theme, which is the central one in Derrida, he is necessarily committed to examining the various textual procedures which Derrida adopts to rid texts of spurious ‘presence’. One of the most striking of these is the invented term différance – a term for which Derrida’s name has become famous. ‘Signifying events depend on differences, but these differences are themselves the products of events. When one focuses on events one is led to affirm the priority of differences, but when one focuses on differences one sees their dependence upon prior events. One can shift back and forth between these two perspectives which never give rise to a synthesis. Each perspective shows the error of the other in an irresolvable dialectic. This alternation Derrida terms différance.’ Likewise, Derrida’s reading of Saussure brings to light the interesting feature that Saussure thought that the spoken form of words takes ontological precedence over writing, which is simply the equivalent of an aid to memory. This view, as Culler reminds us, goes back as far as Plato’s Phaedrus, but it again harbours the doctrine of meaning as presence (spoken meaning anyway) and has to be deconstructed. Derrida ‘upsets’ Saussure and Plato by suggesting that ‘speech may not be independent of writing after all and that writing may affect and infect speech.’ Hence the coming into play of Derrida’s lever, the famous doctrine of the supplément, the idea of ‘supplementarity’ in a text. When the machinery of the ‘différance’ is set to work within an examination of the various ‘suppléments’ of a text, the supposed or alleged plenitude of meaning, and presence, is quickly discovered to have been a sham.

Derrida now comes to assert that ‘speech itself is already a form of writing.’ Instead of privileging language as idealised speech (most ‘present’ in spoken utterance and only slightly attenuated in written form), Derrida suggests we conceive of language as itself ‘a play of differences, a proliferation of traces and repetitions which, under conditions that can be described but never exhaustively specified, give rise to effects of meaning.’

Culler’s phrase there, ‘effects of meaning’, refers us again to the central metaphysical concern of Derrida: illusoriness of ‘meaning’ isolable or isolated in a written text. Derrida’s readings of his own contemporaries are harsh and unforgiving when any sort of ‘presence’ is affirmed. His first major deconstruction was of no less a man than Husserl, whose account of the meaningfulness of signs in his essay ‘The Origin of Geometry’ gets short shrift from Derrida. Similarly, Lévi-Strauss, whom one might have taken as a man acutely aware of the necessity of methodological hygiene, is taken to pieces in an essay, in Of Grammatology, for sheer naivety: ‘Lévi-Strauss, whose blind spots,’ observes Culler dryly, ‘are noted in a language that is often disparaging.’ This is the final Derridan manoeuvre, the showing up of ‘blind spots’ in a text, or the ‘blindness’ of an author to certain textual ‘différances’ or ‘suppléments’ in his own text, operative though not observed. ‘Derrida’s critique of his four contemporaries bears on their failure to scrutinise with sufficient rigour the status of their own discourse.’ Mr Culler then takes the four contemporaries in order. It emerges in the final phase of Jonathan Culler’s essay that Derrida’s aim is consistent and negative: ‘Derrida’s deconstructive reversals are strategic interventions. They do not lay the groundwork for a new discipline – grammatology, he says, is the name of a question – but apply pressure to a system of concepts, upset it so as to make its presuppositions and limitations more apparent.’

I have left to the last the consideration of the rhetorical achievements of Roland Barthes, because it is with him that the underlying ‘ideology’ of structuralism is most clearly apparent. In his chapter on Barthes, John Sturrock runs the literary praxis of Barthes in double harness with a pointing-up of his political aims. And in his introduction, Mr Sturrock made it plain that, from the beginning, structuralism was a kind of ‘ideological’ choice, yet a further renewal of the French Left’s disgust with the culture of the bourgeoisie. ‘By the late Sixties, existentialism as a movement was long defunct,’ writes Mr Sturrock: ‘there was no ideological movement to which it seemed possible to subscribe.’ But from the beginning, structuralism was opposed to the concept of the individual as such as well:

In consequence, structuralism has come to stand for a way of thinking opposed to individualism, or even to humanism, for intentional human agency is given a reduced role in its interpretations of culture ... Certainly there is much hostility to all philosophies of individualism in these thinkers’ writings ... So there is a common ideology at work here: of dissolution, of disbelief in the ego. The self, in the traditional sense, would appear to such as Foucault as a ‘theological’ notion, a false transcendence.

Thus, in his chapter on Barthes, Mr Sturrock starts from Barthes’s ‘extreme distaste’ for ‘essentialism’ in all its forms. Sartre had allowed the human person a certain degree of integrity, ‘but Barthes professes a philosophy of disintegration, whereby the presumed unity of any individual is dissolved into a plurality and we each of us turn out to be many instead of one.’ Hence the four-pronged attack, which Mr Sturrock so expertly describes, on the prevailing literary orthodoxies. From Writing Degree Zero (1953), which was a frontal attack on a sacred cow of French culture, the period of high French Classicism, Barthes develops his critique through the essays we now know as Mythologies (1957): ‘The bourgeoisie is the villain of Mythologies’ The development of the central speculative instrument he needs comes in 1960, with his distinction between the écrivain and the écrivant. The former produces texts which do not ‘lead’ anywhere nor have any one precise meaning, and require readers who are themselves ‘producers’, because the text of the écrivain is scriptible; the latter produce Works, which lead all too predictably in one direction and require readers who are little better than ‘consumers’. By thus eliminating most ordinary people (on the grounds of IQ or something of the kind, one must assume) from the proper re ‘scripting’ of Texts written by écrivains, Barthes has packed the deck so that only he and his friends can play. As a further insult to bourgeois academic reading habits, Barthes shows, in S/Z (1970), that the great Realist Balzac was not a Realist at all, but a mere transcriber of the literary commonplaces of the age. Mr Sturrock, who is always judicious, regards S/Z as both unfair and unjust, but shows how the underlying aim (to break up the idea of a unified ‘author’ transmitting ‘reality’ through ‘words’ in a ‘book’) is served immensely well by breaking up Balzac’s story into hundreds of tiny lexies, and sorting these out into five immense ‘codes’.

In The Pleasure of the Text (1973), Barthes speaks of the Text as ‘undoing nomination’ and the battle against the ‘author’ is nearly over. Even in A Lover’s Discourse, which is written in the first person, the violent urge to destroy the originating ‘subject’ is in evidence:

The writer is thus no more than the grammatical subject, real or implied, of a piece of writing: the explicit or the implicit ‘I’. He is not a substantive presence to be located, as in the past, ‘behind’ the text. He has undergone a dissolution because he is to be found everywhere in what he writes ... When he turns to writing about himself, as in Roland Barthes by Roland Barthes, Barthes lives strictly up to his own rules and appears there in the third person, either as ‘he’ or as ‘RB’.

I am glad to see that Mr Sturrock mentions the book on Japan, L’Empire des Signes, which is not often discussed as a part of Barthes’ oeuvre, but has always seemed to me essential to it. In that book, he can indulge himself in a vision of a country which has no hidden signifieds, no depths, no authorial profundities. Everything is flat, vacant and empty. ‘Japan is a country full of rich and intriguing signifiers whose charm is that they have no signifieds.’

One must be grateful to Mr Sturrock on two counts. It is extremely useful to have these five essays to hand, as we turn into the Eighties, with resources of every kind running out, so that we may evaluate both the present and the possible future status of post-structuralism and deconstruction. But we should also be grateful to him for pointing up the political implications of post-structuralism as such. True, he restricts himself to pointing out that the movement has been, in a traditional sort of way, anti-bourgeois, and that, as a whole, it has shown itself to be anti-humanistic. But at least he has raised the matter of the political status of post-structuralism, which is usually not made a talking point at all. It is indeed such a fascinating subject, and treated at such vertiginously high levels of competence, that there has been a tendency to assume that its intrinsic interest exempts it from political responsibility.

Meanwhile there are more worrying questions to be asked. If individual discourse has now been made a function of language (Lacan), if there is no authorial subject (Barthes), if there is no individual speaker in the ‘discourse’ of history (Foucault), and if there is no ‘meaning’ or ‘presence’ as such in the utterance of any concrete individual (Derrida), what status then has the spoken or written activity of a political ‘dissident’ like Academician Sakharov, at present being held in captivity, and probably maltreated, by the Russian Secret Police in Gorky?

It has already become a commonplace to refer to deconstruction as a ‘nihilism’, but the ending of Jonathan Culler’s account of Derrida makes one wonder if that is a sufficient description of it. Derrida is quoted as follows: ‘At this point, through this movement, both faithful and violent, back and forth between the inside and the outside of philosophy – the philosophy, that is to say, of the West – there results a certain textual activity and product which gives great pleasure.’ And Jonathan Culler comments: ‘Great pleasure to all those who have the interest and the patience to follow the argument of texts which displace or undo the most fundamental categories of our intellectual life.’

But to endorse a philosophy, as Mr Culler does here, which ‘displaces or undoes the most fundamental categories of our intellectual life’ would seem to carry with it some responsibility for replacing Western philosophy’s losses with something of at least equal value. But Mr Culler makes no such offer. Where, in our present dereliction, would we find values to replace ‘the most fundamental categories of our intellectual life’? Among those that might count is surely the idea of Kant and Hegel (‘deconstructed’ by Derrida) the idea of the individual as legal and ethical subject within the State and before the Law?

It seems to me that, in getting rid of the authorial subject, or the originating consciousness, post-structuralism has also (probably inadvertently) made impossible the idea of the ethical subject. And that is a matter which does have grave political consequences. It would not be unfair to Derrida’s watchfulness, for example, to point out a ‘blind spot’ in Glas (1974), which implies that Derrida either does not care about the status of the ethical subject before the Law, or else that he does not concede that such an ethical subject exists at all. For what is the deconstructive procedure of Glas? All down the left-hand column runs the text of Hegel’s account of the Family in The Philosophy of Right. All down the right-hand column runs a text from the writings of Jean Genet. The ploy – to show that all meaning ‘deconstructs’ other meaning if sufficiently cannily placed is brilliantly effective. For what more obvious source, in ‘the philosophy of the West’, for the sanctity of the family, and the status of the individual as legal and ethical subject within the State, than Hegel? And what more diabolically clever way of making Hegel look like an idiot than placing over against the Hegel text, block lot block, passages of Genet? Genet, writing in The Thief’s Journal, eulogises the Gestapo:

The French Gestapo contained two fascinating elements: treason and theft. With homosexuality added, it would be sparkling, unassailable ... It betrayed (betraying means breaking the laws of love). It engaged in pillage. And lastly, it banned itself from the world by homosexuality. It thus established itself as an indestructible solitude.

Later, Genet adds: ‘Betrayal, theft and homosexuality are the basic subjects of this book.’ So be it. But are we to infer from this brilliant compositional technique of Derrida’s (borrowed probably from Joyce’s lay-out in Finnegans Wake, supreme ‘Text’, we remember, for Barthes and for Lacan, where Kev and Dolph abuse each other in glosses over the diagram of the geo-mater) that Genet is simply right in his dispute with Hegel? Philosophically, of course, the answer must be no: Hegel and Genet emerge, like the all-in wrestlers of Barthes’s essay, equally successful, equally beaten according to the tropes of their match. But it doesn’t read that way. The ‘blind spot’ of Glas is that it does endorse (unconsciously?) the discourse of the man who believes in nothing and nobody, the non person Jean Genet. If deconstructive reading implies being ceaselessly watchful, ceaselessly critical, of splits and fissures, gaps and hiatuses, in the writings of others, then it should be aware of certain recurrent suppléments in its own writings.

The argument has already moved on, however. In the new collection of essays by the so-called ‘Yale School’ (five essays under the title Deconstruction and Criticism) one can see that this worry about implicit nihilism is very near to coming to the surface. In his preface, Geoffrey Hartman writes:

Caveat lector. Derrida, de Man and Miller are certainly boa-deconstructors, merciless and consequent, though each enjoys his own style of disclosing again and again the ‘abysm’ of words. But Bloom and Hartman are barely deconstructionists. They even write against it on occasion.

The reason why they do, is that they wish to preserve what is called ‘pathos’, in Yale parlance, from the total victory of ‘Nietzschean aesthetic play’. And Harold Bloom, writing in the first essay, dissociates himself from ‘de Man’s serene linguistic nihilism’.

Deconstruction and Criticism is really a companion-piece to Structuralism and Since, though it takes the arguments one stage further. What seems to be happening is that the sheer proliferation of possible meanings released by the ecstatic ‘jouissance’ of Derridan dissemination has split the Yale thinkers into two groups. They may (as Bloom suggests) actually meet at some extreme outer point, but for the moment they are split. Derrida, de Man and J. Hillis Miller belong to the world of Nietzschean plurality, the whirl of meaning, the endless slip of catachreses into each other, missing presence, lost identity. But Bloom himself (and possibly Geoffrey Hartman, it is implied) seems to be taking up a kind of scepticism about the absence of meaning in words which relates more to the classical perspectivism of Protagoras. This debate is becoming more and more explicitly about whether words don’t refer to the world because they just can’t, or whether words don’t refer to the same world because we have different perspectival (or tropological) histories. ‘The impossibility of reading should not be taken too lightly,’ says Paul de Man. Bloom sourly replies: ‘This most advanced version of Deconstruction cheerfully accepts the risk warned against by de Man’s truest precursor, Friedrich Schlegel: “The irony of irony is the fact that one becomes weary of it if one is offered it everywhere and all the time.”’

It is frustrating to see this debate being carried on between Paris and the United States with such verve and panache, and to realise that, as usual, we in England are not playing any significant part in it. John Sturrock’s book does at least now make it possible to join in.

The only conceptually adequate rejoinder from the British side so far has been Frank Kermode’s brilliant 1977-78 Charles Eliot Norton Lectures, now published under the title The Genesis of Secrecy. Although fully aware of the subtleties of the deconstructivist case, he has yet entered a firm non placet to any form of that theory which tends towards a nihilism of meaning. Taking a variety of narratives, and of narrative devices, he manages to show that there is a whole class of narrative which is not susceptible of deconstruction at all, and so far from containing an infinity of equally possible meanings, cannot even securely be said to contain one.

Whether it be the Gospel of St Mark (the hina/hoti crux of Chapter Four, vv. 11-12), Ulysses, or The Crying of Lot 49 (on which he is very funny as well, see pp. 107-9 and footnote 26 on page 160), he takes a certain pleasure in pointing out that the combined efforts of philologists, grammarians, historians and even hermeneuts will never achieve anything other than the maddening and disappointing rediscovery that interpretative difficulties exist and cannot be resolved. Some narratives just insist on remaining obscure, for their own very good reasons, and Kermode regards this as a source of hope and meaning, rather than a cause for despair. In this he seems to be siding with the Bloom-Hartman-Protagoras view of meaning – that it is shifty rather than with the deconstructivist view of meaning that it doesn’t exist.