- On Deconstruction: Theory and Criticism after Structuralism by Jonathan Culler
Routledge, 307 pp, £16.95, February 1983, ISBN 0 7100 9502 3
Deconstruction, the subject of six new books reviewed in a recent issue of the American journal the New Republic, must be judged, simply by virtue of the commentary it has generated, an important cultural phenomenon. Although it originates in the philosophical writings of the French philosopher Jacques Derrida, deconstruction has exercised its main influence upon the teaching of literature in American universities. Just a few years ago, Derrida’s work was introduced into the American academy by Professor Paul de Man; it was then taken up by his students and colleagues; and for the past five years it has been at the centre of academic literary debate. Intellectual culture thrives upon debate. Although opponents of deconstruction may accuse it of nihilism and anti-humanism, nothing could be more humanistic than vigorous arguments about the nature and aim of literature. Deconstruction has forced traditionalists to look to their assumptions and protect their theoretical flanks. Defensive critics have responded to its challenge by denying the importance of literary theory altogether. That manoeuvre will not work, for anti-theory is itself a theoretical position, and a particularly vulnerable one at that.
But deconstruction has itself benefited from cultural impulses that are anything but theoretical, and has served as an outlet for emotional and institutional needs that have no logical connection with Derrida’s philosophy. Indeed Jonathan Culler rightly says in his workmanlike book that Derrida has not dealt with the ‘task of literary criticism’ and that ‘the implications of deconstruction for the study of literature are far from clear.’ In fact, Derrida’s philosophy has no special implications for literary study or any other subject. As a general philosophy, it entails no specific program in politics, literature, or anything else – though by accident of history it did imply for Paul de Man a scepticism that happened to suit his temperament as a literary critic. But deconstruction as a philosophy holds no more implications for reading books than does, say, the philosophy of Bishop Berkeley. Nonetheless, deconstruction has been applied to literary study, and because of its elusiveness and difficulty, graduate students and others interested in literary theory will wish to have a reliable guide to Derrida’s philosophy from a literary point of view. This Jonathan Culler has supplied in On Deconstruction with his customary lucidity and care. He does not address the non-literary, cultural question as to why Derrida should have caught on in the American academic scene. (This baffled even Derrida, as he told me some years back.) Nor does Culler place Derrida in a wider philosophical context. Culler sticks to the literary applications of deconstruction and he speaks as a disciple and advocate.
In this review I shall pay rather less attention to Culler than to his master. For Culler is mainly an accurate transcriber of Derrida’s views and an acute observer of their uses in de Man and others. Moreover, it is easy to get lost in the details of Culler’s account, despite its lucidity, and I shall use material in his book as a starting-point for rather general observations about Derrida’s philosophy. The sanction that Derrida gives to deconstructive literary criticism must in the end derive from his adequacy as a philosopher. And we will not get very far in gauging his philosophy if we approach deconstruction either as acolytes who accept Derrida at his own (high) estimation or as antagonists who demonise him as a nihilist and anti-humanist. Derrida deserves to be taken seriously – but perhaps not as seriously as either his epigones or his opponents have taken him.
He belongs to a school of modern philosophy that has representatives in both the Anglo-American and Continental camps and includes such diverse names as Wittgenstein, Heidegger, Quine and Sellars – all of whom, despite their diversity, are united in their criticism of the idea that knowledge can have a firm foundation in anything. Not in sense data, nor intuition nor divine revelation. Everything we know is already theory-laden, imprinted with foreknowledge, already an interpretation rather than a given. (The best description of this theme in modern thought is Richard Rorty’s Philosophy and the Mirror of Nature.) Derrida, in criticising ‘presence’ and ‘Western Metaphysics’, is, along with Wittgenstein, Heidegger, Sellars and Quine, criticising the ‘myth of the given’ – the myth that knowledge can be based on something to which we could have direct access. I believe that this attack on the given has succeeded, and that it marks a genuine advance in the history of philosophy. But I don’t by any means accept the idea that it therefore puts an end to ‘truth’ or ‘Western Philosophy’, or does anything as portentous as Derrida and others claim. (It simply marks the end of the myth of the given.)
Derrida’s version of this modern theme makes claims that are open to challenge, but one must concede him both the basic seriousness of his effort and the basic correctness of his attack on ‘presence’ and the given. What raises doubts about the adequacy of his philosophy is its reduction of thought and experience to ‘textuality’. (‘Il n’y a pas de hors texte,’ there is nothing outside text.) This, the most distinctive element in Derrida, is of course the element that has appealed to some of the experts about texts – literary critics. It is also a theme in his philosophy which deserves careful scrutiny.
1. Axioms of Deconstruction
Only the central section of Jonathan Culler’s work is devoted to Derrida’s philosophy as such, the rest being concerned with literary criticism. And even the philosophical section of Culler’s book refers constantly to Derrida’s relevance for the activities of professional critics. This weighting of Culler’s exposition towards the literary domain makes perfectly good sense for the audience he has in mind. But it also creates a certain haziness of focus for those interested in understanding and evaluating Derrida’s thought. Culler’s emphases on ‘iterability’, ‘marginality’ and ‘hierarchical oppositions’ identify points of contact with literary criticism, but these deconstructive fruits have roots that lie elsewhere. If, in seeking those roots, I were to avoid Derrida’s lingo and were to describe his underlying ideas in ordinary terms, something like the following axioms would emerge:
Axiom 1. Everything can be given at least two equally cogent explanations.
Axiom 2. In the temporal process of thinking about anything, one explanation collapses into its contrary.
Axiom 3. This entire process occurs within a linguistic-semiotic structure of thought. From these three axioms and the critique of the given mentioned above can be derived all of the chief doctrines of Derrida’s writings.
1. The Antinomies of Thought. ‘Everything can be given at least two equally cogent explanations.’ Derrida does not argue that everything has at least two equally cogent explanations: he assumes it, and makes it the basis of his second axiom, which is the central principle of his philosophy. But this first assumption should be brought into the light, not only because it is true, as Hume demonstrated in his Treatise of Human Nature, but also because it exposes the hidden connections between Derrida and the traditions of Western philosophy he rejects. Here I refer not only to such traditions as the Cretan liar paradox and Kant’s antinomies (which disclose irreducible bafflements of understanding) but, more particularly, to Hume, the deconstructionist par excellence, who bluntly stated his version of deconstruction as follows: ‘The understanding, when it acts alone, and according to its most general principles, entirely subverts itself and leaves not the lowest degree of evidence in any proposition.’
2. The Instabilities of Thought. ‘In the temporal process of thinking about anything, one explanation collapses into its contrary.’ This collapse into the contrary is the characteristic movement of deconstruction. What we thought to be present turns out to be absent; what we thought to be marginal we discover to be central. This movement is the hallmark of Derridean criticism. Culler states the critical implications of the principle: ‘to deconstruct a discourse is to show how it undermines the philosophy it asserts.’ We know in advance that this interpretative manoeuvre will succeed, because it is founded upon Hume’s inviolable principle that ‘the understanding entirely subverts itself.’
This collapse into the other has its antecedents in other pre-Derridean philosophers, particularly Hegel. Hegel explored how the here and now, the given, is subverted by the passage of time: ‘The Now is pointed out; this Now. “Now”: it has already ceased to be when it is pointed out. The Now that is, is other than the one indicated, and we see that the Now is just this – to be no longer the time when it is.’ Hegel observed that it is the same with any ‘This’: ‘A This is set up; it is however rather an other that is set up; the This is superseded: and this otherness, this cancelling of the former, is itself again annulled’ (The Phenomenology of Mind, Chapter One). A brilliant development of this Hegelian insight is to be found in Heidegger’s introduction to Being and Time, where he meditates on the concept of ‘phenomenon’ – the given that is not given. In still other writers – Blake, William James – this collapse into the contrary is conceived as a cyclical process within intellectual history. Certainly, in this central feature of his philosophy, Derrida has not broken with ‘Western Metaphysics’.
3. The Textuality of Thought. ‘The collapse into the contrary occurs within a linguistic-semiotic structure of thought.’ Like the structuralists his predecessors, Derrida accepts as a starting-point the idea that thought is language in some sense of the term ‘language’. Both structuralists and post-structuralists hold that thought is dependent upon language, and that the structure of thought is like the semiotic structure of a language. Derrida’s originality lies in his further development of this idea. The normal view had been that speech is the basis of language, and thus of thought. Derrida reverses this. He argues that ‘writing’ (in a special sense) is prior to speech. Derrida reasons that since nothing in speech is truly present we must interpret speech as a ‘trace’, an iterable ‘engram’ in memory, which is just what writing is, an engram. Hence writing founds speech, not vice versa. But having made this point (which properly understood is less paradoxical and significant than first appears), Derrida goes on to treat writing as the structuralists treated speech – that is, as a ‘system of differences’.
This notion of language as a ‘system of differences’ started with Saussure, whose original account – from Part One, Chapter Four of the Course in General Linguistics – is worth quoting for its clarity:
Psychologically, our thought, apart from its expression in words – is only a shapeless and indistinct mass. Philosophers and linguists have always agreed in recognising that without the help of signs we would be unable to make a clear-cut, consistent distinction between two ideas. Without language, thought is a vague, uncharted nebula. There are no pre-existing ideas, and nothing is distinct before the appearance of language ... In language there are only differences. Even more important: a difference generally implies positive terms between which the difference is set up: but in language there are only differences without positive terms.
Since thought is a language-like system dependent upon language, and since language is a structure of differences without positive terms, it follows that thought will also exhibit this structure of differences. But from Axiom Two (the collapse into the contrary) we know that when a thought arises from a momentary play of differences it will never be available as a stable present. ‘Now’ constantly becomes ‘Then’, and is constantly deferred. The meaning that arises from the play of differences is therefore never present, but is always being deferred. By combining Axiom Two (deferment) with Axiom Three (Saussure’s ‘difference’), we join deferment with difference, yielding the punning neologism ‘differment’, or in Derrida’s original French, Différance. This neologism and the metaphor of ‘writing’ are twin features of Derrida’s philosophy.
2. One-Sidedness of Deconstruction
Unfortunately for the coherence of that philosophy, deferment and difference do not fit together harmoniously. The principle of difference as enunciated by Saussure requires a stable system of oppositions: Saussure is very clear that the system must be momentarily stable in order to give rise to meaning and the play of differences. That is the basis for his discrimination between ‘synchronic’ (stable) states of language and ‘diachronic’ changes of language over time. The principle of deferment, however, is a principle of constant instability for the system as a whole. Deferment creates a system in which nothing stands still, in which nothing is synchronic. Hegel memorably describes such a system as a ‘bacchanalian revel, where not a member is sober’. One is here compelled to choose between Hegel and Saussure.
That choice ought to be in favour of Hegel. For one thing, it was empirically wrong of Saussure to claim that meaning in language arises exclusively from the systematic play of differences. Although Saussure rightly stressed the autonomous character of language systems, and rightly opposed the view that language is just a set of names for extra-linguistic realities, he was wrong to state his point so absolutely. Language is partly an autonomous system and is partly a set of names that derive their meaning from outside the system. Saussure’s purely internal conception of a language system encouraged him to state flatly that ideas cannot exist before language, but the truth is the other way round. First we have ideas (object concepts), and then we name them. For a recent account of empirical work on names and the function of language in the development of concepts and vice versa, see Language Acquisition.[*] Saussure was a great and original linguistic theorist, but his idea of language as purely a system of differences is incorrect, and is a very weak foundation on which to erect the whole edifice of modern French thought.
But even if the concept of ‘difference’ were not based on an overstated linguistic theory, it would still consort badly with the concept of ‘deferment’. Difference is a kind of pan-lingualism (in Derrida, it is a pan-textuality –Il n’y a pas de hors texte). Difference is thus monistic, even idealistic, in flavour. But deferment – the collapse of one thought into its contrary – is dualistic in flavour. Hegel overcame this inherent dualism by positing an Absolute at the end of the process – an end to deferment. Derrida does not end in an Absolute, not even an Absolute Text. Deconstruction, by coming to a stop in a monistic conception of difference à la Saussure, is at odds with its own genuine insights.
Derrida’s literary followers are even less careful than Derrida on this score. Here is a statement by Culler (the italics are mine):
When one attempts to formulate the distinction between reading and misreading, one inevitably relies on some notion of identity and difference. Reading and understanding preserve or reproduce a content or meaning, maintain its identity, while misunderstanding and misreading distort it; they produce or introduce a difference. But one can argue that in fact the transformation or modification of meaning that characterises misunderstanding is also at work in what we call understanding ... We can thus say, in a formulation more valid than its converse, that understanding is a special case of misunderstanding.
In a similar vein, Culler argues that for the opposition literal-v.-metaphorical, the latter is foundational: a literal expression is a ‘metaphor whose figurality has been forgotten’. Such tendencies to monism are a persistent danger for deconstructionists, and a danger that they rarely avoid in practice. Yet to be a monist is precisely not to be a deconstructionist! One ought therefore to distinguish between authentic deconstruction and capital-D Deconstruction, which in its monistic forms is a very inconsistent philosophy indeed.
As an example of the one-sidedness of Big-D Deconstruction, we may consider how it treats the following list of contraries:
1. part .................................... whole
2. percept .............................. object
3. signifier ........................... signified
4. temporal ......... spatial (non-temporal)
5. difference ........................ sameness
Big-D Deconstruction characteristically chooses the left-hand side of this list. It reduces the right-hand side to an illusion whose reality is on the left. The collapse into the contrary seems to go just one way and come to a halt. Of those contraries listed above, perhaps the fourth, the non-temporal v. the temporal, could be viewed as the basis for Deconstruction’s other leftward-tending preferences. Temporality, after all, is the ground for ‘Deferment’. Derrida holds that mental life is purely temporal, is just one-thing-after-another; one moment is always different from another moment of mental life. Husserl’s profound argument against this temporal conception of mental life led Derrida to devote a whole book (Speech and Phenomena) to attacking Husserl. But Derrida never touched Husserl’s key argument favouring a dualistic, i.e. a temporal-nontemporal, conception of mind. Derrida concentrated instead on Husserl’s admittedly vulnerable conception of Presence, as though by thrashing Husserl on that peripheral issue he could also defeat his other ideas. But to the extent that empirical psychology has any say in the matter, Husserl’s dualism is a correct, and Derrida’s monistic temporality an incorrect, account of mind. Even if that were not so, Deconstruction would be inconsistent in accepting temporality as an adequate description of mental life. On this point, as on so many others, Hume showed himself to be the more authentic deconstructionist when he admitted that the persistence of self-identical objects over time cannot be either confidently asserted or denied. Hume also said in similar vein that ‘a true sceptic will be diffident of his philosophical doubts, as well as of his philosophical conviction’ (Treatise, VII). That is the authentic principle of deconstruction – not Derrida’s ‘différance’ but Hume’s ‘diffidence’.
3. Deconstruction and Formalism
Derrida’s weaknesses as a philosopher are somewhat beside the point, however, when we enter the realm of literary deconstruction as Culler describes it. Culler seems to admire the success of Deconstruction in sanctioning and continuing the professional occupation of writing about writing. His account suggests that Deconstruction has a self-sustaining effect on university publication. An academic institution, like any other, adopts an ideology that preserves the institution as it is. This is the powerful principle of institutional homeostasis.
No harm in that. But the cultural question that needs to be asked is whether we want to sustain the institutions of textual analysis that have dominated academic literary criticism in the past forty years. The trouble with keeping that tradition going under a new deconstructive guise is not that it is wrong or radical, or inhumane, but that the tradition of academic literary analysis is uncommitted to any cultural values at all. Literary Deconstruction is another version of formalism. It is quite unconcerned, for example, with choosing a new canon. (The old canon will do fine, Culler informs us.) Deconstruction stresses the how of criticism rather than the what. And like the New Criticism before it, Deconstruction claims that the how is the what of literature. Similarly, just as the New Criticism tended to find that the subject of literature was literature, Deconstruction finds that the subject of literature is Deconstruction. In exposing this feature of Deconstruction, Culler’s account exhibits the twin virtues of clarity and explicitness: ‘When considered at the first level, literature is remarkable for the diversity of its themes ... At the second level, a powerful theory with literary implications seeks to analyse those structures which it takes to be most fundamental or characteristic, and thus emphasises repetition ... Although deconstructive readings work to reveal how a given text elucidates or allegorically thematises this ubiquitous structure, they are not thereby promoting one theme and denying others but attempting at another level to describe the logic of texts.’
‘To describe the logic of texts’ is to describe their form, logic being the study of form par excellence. Such preoccupation with form in the American academy is part of a general tendency in American education to inculcate reading and writing skills without committing one to any preference for particular cultural contents. Recently we have discovered that this educational formalism will not work even in teaching elementary reading skills. To think that formalism could suffice in teaching a literary tradition is an even more obvious mistake. Deconstruction as practised in America is part of a pervasive educational formalism that avoids advocating specific values and contents. But in literary education such formalism is an evasion.
Nothing could be more illustrative of this evasive, American use of Deconstruction than Culler’s treatment of feminist criticism. Culler deserves praise for treating that subject at all, and he is right to say that feminism is ‘one of the most significant and broadly based critical movements of recent years’. But after spending twenty pages in analysing recent work on the subject, he summarises feminist criticism as follows:
From these varied writings a general structure emerges. In the first moment or mode, where woman’s experience is treated as a firm ground for interpretation, one swiftly discovers that this experience is not the sequence of thoughts present to the reader’s consciousness ... In the second mode, the problem is how to make it possible to read as a woman ... In the third mode, the appeal to experience is still there ... But experience always has this divided, duplicated character; it has always already occurred and yet is still to be produced.
In short, the logic of feminism follows the general logic of Deconstruction. Whether or not that is so, this abstracting of feminism to its Derridean ‘logic’ or ‘structure’ seems to me to express no significant truth at all about the feminist movement in criticism, and provides no basis for calling it ‘one of the most significant’ critical movements of recent years. That it certainly is, because of its content, not its form, and because it has encouraged a change in our canon, and in our estimate and use of particular works.
In my view, the most glaring weakness of American Deconstruction is not its intellectual incoherence but its cultural evasiveness. ‘English’ in American schools and universities has always been a cultural, not a progressive, intellectual subject. Although ‘English’ does have connections with the genuine disciplines of history and philosophy it came into being for cultural rather than disciplinary reasons. Every attempt to show that ‘English’ is a discipline with a logic and method of its own has so far proved specious and unenduring. Such narrow approaches to literature do not butter any intellectual or cultural bread. The function of ‘English’ is to help sustain or change traditions, to help provide the myths and values we live by, and to help create a culture that is worth living in. Formalism has seduced American literary study away from these authentic and original cultural purposes. Is it too much to hope that Deconstruction, the reductio ad absurdum of formalism (and also a very inconsiderable philosophy), may be the last gasp of this evasive tradition?
[*] Edited by Eric Wanner and Lila Gleitman. Cambridge, £30 and £10.95, 10 February, 0 521 23817 X.