Text Production 
by Michael Riffaterre, translated by Terese Lyons.
Columbia, 341 pp., $32.50, September 1983, 0 231 05334 7
Show More
Writing and the Experience of Limits 
by Philippe Sollers, edited by David Hayman, translated by Philip Barnard.
Columbia, 242 pp., $31.50, September 1983, 0 231 05292 8
Show More
The Reach of Criticism: Method and Perception in Literary Theory 
by Paul Fry.
Yale, 239 pp., £18, October 1984, 0 300 02924 1
Show More
Blindness and Insight: Essays in the Rhetoric of Contemporary Criticism 
by Paul de Man, edited by Wlad Godzich.
Methuen, 308 pp., £7.50, November 1983, 0 416 35860 8
Show More
Displacement: Derrida and After 
edited by Mark Krupnick.
Indiana, 198 pp., £9.75, December 1983, 0 253 31803 3
Show More
Authoritarian Fictions: The Ideological Novel as a Literary Genre 
by Susan Rubin Suleiman.
Columbia, 299 pp., £39, August 1983, 0 231 05492 0
Show More
Show More

One gets the impression from Riffaterre’s book that he enjoys playing single-minded hedgehog to the foxy representatives of Parisian post-structuralist fashion. Despite some fairly arcane terminology, he is basically an old-style formalist whose forays into theory are largely in the service of traditional interpretative ends. The literary text, for Riffaterre, is an object of patient and erudite close-reading, a ‘monument’ whose utterly distinctive character the critic sets out to describe and explain. At bottom, there is not much difference between this kind of ‘structuralist’ activity and the techniques of verbal analysis perfected by ‘old’ New Critics like Wimsatt and Brooks. Of course there is a shift of technical idiom, from the homespun rhetorics of Ambiguity, Irony and Paradox to a post-Saussurian language of signifier and signified, text and intertext. But Riffaterre’s commitment to the structuralist project stops well short of dissolving the poem into a play of circumambient codes and conventions beyond all reach of formal analysis. His readings are squarely opposed to the current mood of ‘textualist’ euphoria which merges poem and commentary in an endless exchange of productive signification. Criticism has its work cut out, he thinks, in explaining what it is about the nature of literary texts that both marks them out clearly as ‘literary’ and preserves their meaning against the ravages of time and cultural change. There is much here that would gladden the heart of a rearguard New Critic, but little – besides the somewhat de rigueur terminology – that a current post-structuralist would want to take on board.

By an odd quirk of circumstance, Columbia have also brought out a translation of Philippe Sollers’s Writing and the Experience of Limits, a collection of essays first published in the heady mid-Sixties. Nothing could demonstrate more plainly the gulf between Riffaterre’s conservative structuralism and the other, more ‘radical’ versions which were then being promoted, mainly in the journal Tel Quel. This was the era of what Barthes called ‘semioclastics’, of a textual theory committed to overthrowing the bourgeois regimes of language, ideology and narrative realism. In the name of ‘semantic materialism’ (or Lacan’s ‘insistence of the letter’), texts were deconstructed to reveal their complicity with an age-old metaphysic which subjugated writing to reading, active production to the passive consumption of sense. The bad old habits of interpretative criticism were seen as products of a mystified endeavour to conceal the sheer disruptiveness of literary texts. Any talk of ‘meaning’ was somehow in league with the process of capitalist accumulation which worked to expropriate labour (= textual ‘production’) in the form of conceptual surplus value. It all added up to a heady rhetoric, a fine iconoclastic mixing of metaphors which grafted Marx onto Freud, the latter (as interpreted by Lacan) providing a link between psychic economy and the workings of materialised textual desire. Bliss was it in that dawn to be reading Tel Quel and watching the signs that led up to les événements of May ’68.

Sollers was among the chief theorists of that ‘moment’, along with Barthes and Derrida. It is not hard to see the connection between these essays and a text like Barthes’s S/Z, where the scientistic dream of a structuralist ‘method’ is likewise forced up against its limits and discarded in favour of a liberated, hedonistic discourse of fragmentary meaning. In 1967 Derrida published the two volumes of essays (Grammatology and Writing and Difference) which proclaimed an end to the repressive era of ‘Western metaphysics’ and, along with it, the concept of ‘structure’ as a last-ditch metaphysical ploy. If Anglophone readers had a long time to wait before those cardinal texts appeared in translation, they will be even more belated in catching up with the ‘moment’ represented in Sollers’s essays. For it is a matter of record that Sollers and most of the Tel Quel editorial collective went on to abandon their Leninist-Maoist stance and to take up a rhetoric not far removed from that of the French New Right. Barthes maintained a certain canny ambivalence with regard to his radicalised colleagues and admirers. Derrida took issue more directly, arguing (in a well-known interview) that much of what passed for ‘materialist’ dialectic still bore the traces of a residual idealism which a deconstructive reading would bring to light. His diagnosis was strikingly borne out. It now seems clear that the Tel Quel brand of ‘textual’ leftism was a token politics easily adaptable to just about any kind of new-found radical creed.

Riffaterre might derive a certain satisfaction from these vagaries of ideological style. His own poetics is conceived as proof against any such temporal visions and revisions. Hedgehog-like, he sticks to the one Big Idea which holds out the promise of a veritable method for unlocking the mysteries of literary texts. He flatly asserts what Sollers just as flatly denies: that there exists a body of texts which deserve the title ‘literature’, and whose workings are amenable to a certain kind of formalist description. For Sollers, this is precisely the root metaphysical prejudice which has hitherto prevented texts from being read – productively, transgressively read – as modes of ‘material’ signifying practice. ‘Literature’ is the locus of all surplus value, the concept which drains off the energies of writing into a timeless realm of plenary meaning. Textuality, he writes, ‘is not to be identified with the historically-determined concept of “literature”. It implies the overturning and complete rearrangement of both the role and effects of this concept.’ His exemplary readings – of Dante, Sade, Bataille and others – all have to do with the textual traversal of limits which unlooses a writing beyond all grasp of traditional notions like ‘literature’, ‘authorship’ and ‘meaning’. Writing becomes the very practice of radical transgression, breaking down the various structured oppositions which separate mind from body, soul from sense, the ‘meaning’ of a text from the physical activity which went into its making. It is all a far cry from the quest for the ‘literary’ structures of signification which Riffaterre sees as the only valid form of critical enquiry.

What this difference comes down to in the end is the question of how far texts can be treated as autonomous, self-sufficient verbal artefacts. Riffaterre is at one with Brooks and Wimsatt – not to mention Aristotle and his formalist descendants – in believing that they must be so treated if poems are to retain any kind of enduring significance. He doesn’t, of course, deny that certain meanings may be lost, and others accrue in their place, by a process of cultural transmission which the text can never anticipate or control. It is always possible, in reading a poem, that we may have lost contact with the ‘descriptive system’ which its words once evoked, so that ‘we read the same sentence as the first readers, but we have lost its echo.’ This is the province of what Riffaterre calls the literary ‘intertext’, including all those shared descriptive topoi (not necessarily conscious allusions) which make up the cultural background of an age. But Riffaterre’s point is to set firm limits to the degree of mutability entailed by this time-bound condition. There are things going on within the text, he argues – processes of complex verbal patterning or ‘overdetermination’ – which save it from a total dependence on transient echoes and allusions. It is here that the structuralist machinery comes into play, describing the various generative systems which ensure that the poem survives more or less intact against the shifting codes of intertextuality.

Riffaterre’s poetics is basically a theory of literate (perhaps ultra-literate) reader-response. The first stage of interpreting a text is to encounter certain problems in the way of a first-order mimetic or ‘referential’ reading. Anomalous details – whether redundant, contradictory or simply obscure – force the reader to abandon this literal interpretation and look for coherence at a higher, textual level. This move ‘from mimesis to semiosis’ – from naive to sophisticated reading – is the mark of a genuinely ‘literary’ response. From there it is a matter of going on to recognise that details which lack ‘motivation’ in straightforward mimetic terms may take on a high charge of meaning as elements in the complex textual weave. This is what Riffaterre has in mind when he refers to the ‘overdetermined’ character of significant detail. The components of the literary sentence, he writes, ‘are tied together by the syntagm [i.e. the unfolding sequence of words], but these relations are repeated by other, formal or semantic, relations. Each word, therefore, appears to be necessary many times over, and its relations with the other words appear to be multiply imperative.’ Once alerted to the presence of these verbal mechanisms, the reader makes sense of them by as far as possible reconstructing the chains of associative detail which produced them in the first place. This involves a sequence of interpretative moves which Riffaterre pursues through a wide variety of well-chosen examples. In each case, the text is shown to yield up structures of closely-wrought verbal patterning which point interpretation away from the vagaries of mere impressionistic reading.

Riffaterre is certainly an adept reader when it comes to backing up these generalised claims with passages of detailed explication. Most often he fastens onto some literary tag or familiar – even cliché’d – expression which the text appears to work into various forms of covert or displaced repetition. To this extent, the reading is necessarily ‘intertextual’. It relies, that is to say, on analogues and a range of cross-reference which demand not only great erudition but a measure of speculative daring on the critic’s part. Yet the end of this activity, Riffaterre insists, is to rescue both literature and criticism from the mere piling-up of random interpretative guesswork. If his method is worth anything, then it is, quite simply, the one proper and answerable way to interpret texts. This readiness to stake his claim upon ultimate truths is a measure of his distance from the current post-structuralist vanguard. ‘The text’s true significance lies in its consistent formal reference to and repetition of what it is about, despite continuous variations in the way it goes about saying it.’ Hence the underlying conservatism which everywhere qualifies Riffaterre’s ingenious seeking-out of plural meanings. Even the baroque extrusions of technical vocabulary – ‘hypogram’, ‘paragram’, ‘matrix’ – amount to little more than a handy scaffolding of argument within which to reaffirm the text’s enduring significance. It may be, as Riffaterre concedes, that ‘the reader who shares the author’s culture will have a richer intertext.’ But the formalist has a ready answer: that even such a privileged reader ‘will be able to draw on that wealth only when semantic anomalies in the text’s linearity force him to look to non-linearity for a solution’. The book comes adorned with admiring comments from Murray Krieger, a defender of ‘old’ New Critical practice. It is a thoroughgoing statement of the formalist case pushed – at times – to a perverse extreme, and all the more usefully debatable for that.

If there is one likely objection to Riffaterre’s method, it is the lack of historical specificity which goes along with his radical formalism. Texts become monuments to their own power of resisting the temporal flux that would otherwise place their sense beyond hope of recovery. Riffaterre admits all kinds of cultural ‘code’ to the deciphering of texts, whether from the visual arts, the clichés of period taste or the repertoire of popular mythology. All this gives his writing great scope and apparent flexibility. What he cannot do, on the terms laid down by his method, is provide an account of historical change that doesn’t collapse straight away into a timeless, unending circulation of meanings and interpretative strategies.

The following passage (from a chapter on Francis Ponge) shows the process at work: ‘Insofar as it is a body of lexical collocations, clichés, and stereotypes established by years of usage, language provides the tertium comparationis, the meeting-point, the verbal association that would be a pun if it appeared in the text. But since it does not appear in the text, it is the matrix, the key to the enigma that the formal strangeness of the verbal sequence would pose, if only the latent association did not reveal the logic behind that sequence.’ This nicely recapitulates the manner of reasoning that Riffaterre employs in most of his worked-out examples. It presupposes an undifferentiated signifying play within language, a stock of ahistorical codes and conventions which offer themselves up to formal analysis as the only means to make ‘logical’ sense of their bewildering multiplicity. But the logic in question is an arbitrary construct everywhere dependent on Riffaterre’s well-oiled descriptive machinery. As with most machines, what comes out in the end is determined by what goes in to begin with, modified only by a certain set of operating instructions. Riffaterre is a shrewd operator, but his readings do seem very often self-condemned to a kind of routine ingenuity.

If Riffaterre holds out for the virtues of a disciplined, methodical criticism, Paul Fry has arguments in plenty for doubting that method can ever be more than a delusive dream. His book is a sustained and elegant meditation on the limits of formalist theory, the ways in which reading confounds or eludes the rules laid down by ‘rigorous’ thinkers. Up to a point it resembles the work of such hardline deconstructionist critics as Paul de Man. Texts are unable to say precisely what they mean, caught up in an alternating rhythm of ‘blindness’ and ‘insight’, their most confident statements always undermined by rhetorical forces beyond their control. Criticism is most deluded – though also, for that reason, most intensely revealing – where it sets up to speak the truth of the text from a standpoint of assured conceptual mastery. Then it becomes clear, through a deconstructive reading, that the critic’s own language is as prone as any other to those twists and contradictions of figural sense which indicate that theory is after all caught up in a common textual predicament. Interpretative essays deserve close reading of the kind traditionally reserved for ‘literary’ texts. What’s more, de Man would argue, our traditional habits of response are naive and self-blinded in their refusal to recognise the perpetual non-coincidence of logic and rhetoric, statement and intent. Interpretation cannot escape the uncanny dialectic of insight and error that structures its every performance. There is no ‘meta-language’ available to critics, no means of getting outside the contaminating influence of text upon text, ‘literary’ language on the discourse that tries to comprehend it.

This was de Man’s leading argument in an essay on Riffaterre published in the American journal Diacritics (Winter, 1981). What the formalist approach was constrained to ignore were those irruptions of uncontrolled figurative sense in its own discourse which threatened the critic’s hard-earned methodical composure. Though there is finally no escape from this predicament, there is at least the preferable option of remaining alert to it, refusing the easy consolations of method and certainty. The point was driven home in the essays which made up de Man’s Blindness and Insight, published in 1971. Methuen have now issued a second (paperback) edition which finds room for supplementary essays. These include ‘The Rhetoric of Temporality’, without doubt the most important and influential text of American deconstruction. It is a wonderful book, compellingly argued, the product of a restless intelligence nowhere content to operate at less than full self-critical stretch. As a matter of record, it demonstrates clearly that de Man had arrived at his own deconstructionist standpoint by a route independent of Derrida’s influence. What he has to say here about Derrida on Rousseau is shrewdly beforehand with the subtlest of Derridean strategies. Certainly philosophers could take a few lessons from de Man in that mode of vigilant close-reading which shows up the slippage between logic and rhetoric, method and actual performance. His untimely death, a few weeks ago, deprived American criticism of its keenest – if to some most perversely scrupulous – philosophical intelligence.

Fry is at one with de Man in thinking that any too-rigorous critical methodology is sure to produce such unlooked-for effects of textual repression and displacement. Where he differs – from de Man as from the deconstructive enterprise at large – is in not trying to substitute his own kind of self-denying rigour for the methods and theories in question. Nor is he simply content, like your typical bluff British empiricist, to reject all ‘theory’ out of hand and trust to sound instinct and plain good sense. In closely-argued chapters on a diverse company of critics he sets out to show how formalism ultimately defeats its own best efforts, but also how productive those defeats may be when one interprets them aright. If his book constantly draws attention to ‘the blunting of interpretation by method’, it is equally ready to acknowledge ‘that method as such is what constitutes the necessary error of representation.’ The essays that result from this tension – especially those on Longinus and Dryden – are among the most subtle and resourceful that recent criticism has to show.

What is it, Fry asks, that a ‘poetics of order’ – like Aristotle’s radical formalism – has to leave out of account? This residue includes ‘the actual, material world, it includes the sublime, and it also includes points of contact and divergence between these horizons of representation. Such factors, which are difficult if not impossible to assimilate into any theory of manifest form, appear interchangeably and with the greatest vividness at the extremes of the minute and the boundless.’ The formalistic bias of Aristotle’s thinking – along with his insistence on the perfect Mean – betrays a defensive, even ‘claustrophobic’ attachment to principles of order. As Fry reads the Poetics, this gives rise to a repression of those disorderly or downright ‘irrational’ motives which would otherwise wreak havoc with a play’s formal unity. Yet the upshot of this repression is a certain problematical nexus of themes – metaphor, recognition and kinship – which challenge the logic of Aristotle’s argument by its very need to reduce them to order. Formalistic notions of ‘unity’, like structural definitions of metaphor, may run up against sharply paradoxical limits. What Aristotle avoids is a ‘scene of recognition’ more disturbing than anything allowed for in his shapely, self-regulating concept of form. Thus when Aristotle says – probably thinking of the ‘well-knit’ Oedipus – that ‘the whole will be disjointed or torn apart if the parts are moved around,’ he is forgetting (in Fry’s words) that ‘this play, and many others too, are finally about the break-up of relations that are too closely established.’ The formalist quest for ‘objective’ structures of meaning fails to take account of those ‘uncertain issues’ and irrational forces which ‘can never be purged or purified by the catharsis of interpretation’.

Fry then turns to Longinus as a critic better able to point a way beyond these inbred limitations of formalist method. In fact he suggests that his book might be read as an attempt to ‘recover nearly the whole field for Longinus’, as opposed to the dominant critical tradition descending from Aristotle to the present-day formalists. The Longinan Sublime becomes a touchstone, for Fry, of that willingness to entertain doubts and perplexities which formalism vainly strives to subdue. In place of an irritable reaching after method, Longinus holds criticism open to an ‘interplay among nature, feeling and language’, the effect of which is to break down the firm categorical distinctions of Aristotelian reasoning. The Sublime is that ‘grace’ beyond the reach of art which eludes and subverts all manner of formal description. It beckons the interpreter to flights of emulative fancy which abolish – if only momentarily – the distance between subject and object, reader and poem. Furthermore, as Fry argues, the Sublime is closely related to that complex of subliminal motives – repression, displacement, repetition – which Freud tried to theorise in his writings on ‘the uncanny’. These properties all have the virtue of chastening the positivist rage for order, the root Aristotelian assumption that form should reflect the tidy categories of logical thought. To extract any kind of ‘theory’ from Longinus is (Fry admits) a swerve into method, and hence a misreading of his fragmentary text. At most, there is a salutary lesson to be drawn: namely, that the interests of criticism are often best served by a hesitant seeking after ‘tenuous symmetries’ which nowhere congeal into certitudes of method. ‘Having arrived at a liminal understanding of form ... by feeling along its edges like someone who is blind, the interpreter begins, at that point of exhaustion, to interpret.’ Thus the Oedipal ‘scene of recognition’ translates from an Aristotelian context of structural necessity to a realm of uncanny blind revelation more akin to the authentic Longinan Sublime.

Paul de Man’s essay on Hegel (in the Krupnick volume) attempts a more rigorous deconstructive account of the Sublime, in this case as related to the discourse of political and ethical reason. De Man typically singles out those moments of rhetorical self-undoing which mark the aesthetic – the realm of art and figural language – as the problematic crux of Hegel’s enterprise. In a long tradition of speculative thought beginning with Kant, the problem of knowledge has been linked to that of practical (or political) reason only by way of an intervening aesthetic critique. It is here, de Man argues, that the ‘usurped authority’of abstract concepts is most effectively challenged and made to reveal its fictive (or figural) basis. It is a measure of this disruptive power that the Sublime plays so crucial yet marginal a role in the Aesthetics, while that work itself occupies an ‘enslaved position’ within Hegel’s production as a whole. Hegel’s theory of art is riven by internal problems and contradictions, as de Man’s brief reading powerfully demonstrates. Yet it is only by making this detour through aesthetic reason that thought can resist the kind of systematic, totalising bent that marks the speculative enterprise. ‘Poets, philosophers and their readers lose their political impact if they become, in turn, usurpers of mastery. One way of doing this is by avoiding, for whatever reason, the critical thrust of aesthetic judgment.’ Thus deconstruction lays claim to a power of political leverage, contrary to the arguments of those (like Terry Eagleton) who deplore its self-occupied textual games.

Several of Krupnick’s contributors make a point of stressing this alignment between deconstruction and a radical critique of existing political discourse. Michael Ryan has an essay on the inbuilt contradictions of liberal reason, arguing that we need a different epistemology – an alternative idea of language, politics and truth – in order to grasp and transcend these antinomies. The trouble with this post-liberal deconstructive theory is that it tends to take on authoritative tonings markedly at odds with its own express argument. Thus, to cite one unfortunate instance: ‘The principle of infallible, decisively singular truth as absolute cognitive certainty in liberal reason is intolerant of the undecidability (of individual and collective, universal and particular, fact and value, subject and object, etc) that is in fact the case in the world.’ The awkwardness of this passage is more than stylistic, involving as it does a flat contradiction between statement and rhetoric, an attack on ‘absolute certainty’ couched in terms of what is ‘in fact the case’. The tension is hardly eased when Ryan goes on to argue that ‘undecidability is “true”, that is existent, but it escapes the rationalist classificatory scheme of liberal Reason.’

Still, there are some cogent arguments at work in Ryan’s essay, as in other contributions to this timely and provocative collection. Gayatri Spivak writes on deconstruction from the standpoint of a radical feminism reflecting on its own marginal status vis-à-vis the mainstream intellectual discourses of Hegel, Marx and Freud. Others choose to focus on Derrida’s deconstructive strategies, his style (or profusion of styles) and the cultural tradition which his texts both challenge and strikingly rejoin. Susan Handelman has intriguing points to make about the affinity between deconstruction and the Jewish (chiefly Kabbalistic) practice of heterodox textual interpretation. What these essays all witness to is the extraordinary energy of thought and style which Derrida’s writings have unleashed among admiring and dissenting commentators alike. No doubt there are powerful institutional reasons why so many critics, philosophers and others have treated deconstruction as, in Paul de Man’s words, either a ‘harmless academic game’ or a kind of ‘terrorist weapon’. At worst, it can be pushed in both directions, as Derrida himself has been quick to point out. But the real cause for anxiety seems to be that ‘literary theory’ – that hybrid, subservient activity – is busily creating such a shake-down of intellectual disciplines that its claim had better be ignored or ridiculed. That this marginal status can be turned to advantage by various forms of conscious textual ‘displacement’ is the message proclaimed throughout these essays.

Susan Suleiman’s Authoritarian Fictions is a patient and well-documented study of one specific genre – the roman à thèse – and the typical narrative structures and devices to be found in it. Suleiman concentrates on French novels written in the first half of the 20th century and representing both left and right-wing political views. Her main exemplars include Sartre, Aragon and Mauriac. The great virtue of her book is its ability to synthesise a range of theoretical ideas – whether formalist, structuralist or ‘reader-response’ – in the service of a clear and compelling critical argument. Thus she explains how it is that didactic novels impose their ideological message by various kinds of ‘redundant’ (or overdetermined) narrative detail. Options are effectively closed off – at least for the compliant reader – through a process of heavily ‘coded’ narration which allows few liberties of response. All this Suleiman demonstrates with pointed economy by drawing on the resources of formalist and structuralist method. Value-judgments are tactfully suspended, along with any form of overt political weighting which might want to distinguish mere right-wing ‘propaganda’ from straightforward left-wing calls to action. Thus Sartre and Aragon are shown to deploy persuasive techniques much akin to those at work in the exemplary fictions of a Catholic writer like Mauriac.

But clearly there are limits to this attitude of studied detachment, and Suleiman does well to acknowledge them. Nor is it a question of simply shifting gear, dropping the pretence of white-coated clinical neutrality and owning up to a frank dislike for novels which (whether crudely or cleverly) box the reader into an ideological corner. For Suleiman there is no such line to be crossed between the interests of ‘theory’ and those of doing justice to our involvement as active, self-aware readers. The objections to narrative propaganda can be stated in what amount to post-structuralist terms. Thus Suleiman writes: ‘What constitutes, for me, the greatest interest of the roman à thèse lies in its hybrid character, generating tension between two opposing tendencies: the simplifying and schematising tendency of the thesis, and the complicating and pluralising tendency of novelistic writing.’ A good part of her book is devoted to explaining just how and where such texts overrun the bounds of programmed doctrinal adherence. Altogether she mounts a most impressive case against the arguments of those who regard post-structuralism in any guise as a species of textual mystification devoid of political relevance.

Send Letters To:

The Editor
London Review of Books,
28 Little Russell Street
London, WC1A 2HN


Please include name, address, and a telephone number.

Read anywhere with the London Review of Books app, available now from the App Store for Apple devices, Google Play for Android devices and Amazon for your Kindle Fire.

Sign up to our newsletter

For highlights from the latest issue, our archive and the blog, as well as news, events and exclusive promotions.

Newsletter Preferences