Stanford Presidential Lectures and Symposia in the Humanities and Arts

Lecturers || Jacques Derrida Home

Deconstructionist Theory

The following text was extracted from The Cambridge History of Literary Criticism -- vol.8 From Formalism to Poststructuralism. Cambridge University Press, 1995.

Deconstructionist Theory

Most of Derrida's work continues a line of thought which begins with Friedrich Nietzsche and runs through Martin Heidegger. This line of thought is characterized by an ever more radical repudiation of Platonismoof the apparatus of philosophical distinctions which the West inherited from Plato and which has dominated European thought. In a memorable passage in The Twilight of the Idols, Nietzsche describes 'how the "true world" became a fable.' There he sketches an account of the gradual dissolution of the other-worldy way of thinking common to Plato, to Christianity, and to Kant, the way of thinking which contrasts the True World of Reality with the World of Appearance created by the senses, or matter, or Sin, or the structure of the human understanding. The characteristic expressions of this other-worldliness, this attempt to escape from time and history into eternity, are what deconstructionists often call 'the traditional binary oppositions': true--false, original--derivative, unified--diverse, objective--subjective, and so on.


What Heidegger called 'Platonism' or 'metaphyics' or 'onto-theology' Derrida calls 'the metaphysics of presence' or 'logocentrism' (or, occasionally, 'phallogocentrism'). Derrida repeats Heidegger's claim that this metaphysics is utterly pervasive in Western culture. Both see the influence of the traditional binary oppositions as infecting all areas of life and thought, including literature and the criticism of literature. So Derrida entirely agrees with Heidegger that the task of the thinker is to twist free of these oppositions, and of the forms of intellectual and cultural life which they structure. However, Derrida does not think that Heidegger succeeded in twisting free. As he says:

What I have attempted to do would not have been possible without the opening of Heidegger's questions ... But despite this debt to Heidegger's thought, or rather because of it, I attempt to locate in Heidegger's text ... the signs of belonging to metaphysics, or to what he calls onto-theology'.
(Derrida, Positions, pp. 9-10)

In order to distance himself from Heidegger, Derrida proceeds to invent bits of philosophical terminology (trace, différance, archi-écriture, supplement, and many others) designed to mock and displace Heidegger's own terminology (Ereignis, Lichtung and the like).1 Whereas Heidegger's words express his reverence for the ineffable, the silent, and the enduring, Derrida's express his affectionate admiration for the proliferating, the elusive, the allusive, the ever-self-recontextualizing. He sees these features as exemplified in writing better than in speech -- thus reversing Plato's (and Heidegger's) preference for the spoken over the written word. By constructing this terminology, Derrida is trying for the position for which Heidegger had implicitly nominated himself, that of the first post-metaphysical thinker, the prophet of an age in which the reality--appearance distinction has entirely lost its hegemony over our thought.

By abandoning Heideggerian nostalgia, Derrida freed himself from those elements in Heidegger's thought which chimed with Heidegger's own sentimental pastoralism and nationalismotraits which led him to Nazism. Derrida thus helped free Heidegger up for the use of the political left. Further, and more importantly for the purposes of deconstructionist literary critics, he turned from Heidegger's sentimental question 'How can we find traces of the remembrance of Being in the texts of the history of philosophy?' to the quasi-political questions 'How can we subvert the intentions of texts which invoke metaphysical oppositions? How can we expose them as metaphysical?' He turned from Heidegger's preoccupation with the philosophical canon to the development of a technique which could be applied to almost any text, past or contemporary, literary, or philosophical. This was the technique which has come to be called 'deconstruction.'

The word deconstruction plays as small a role in Derrida's writing as Abbau and Destruktion played in Heidegger's. 'Deconstructionism' was, initially, no more Derrida's chosen label for his own thought than 'existentialism' was Heidegger's label for the doctrines of Being and Time. But, because Derrida was made famous (in English-speaking countries) not by his fellow-philosophers but by literary critics (who were looking for new ways of reading texts rather than for a new understanding of intellectual history), this label has (in those countries) become firmly attached to a school of which Derrida is, rather to his own surprise and bemusement, the leading figure.2 As used by members of this school, the term 'deconstruction', refers in the first instance to the way in which the 'accidental' features of a text can be seen as betraying, subverting, its purportedly 'essential' message.3


One can generalize Derrida's comment on Heidegger as follows: anyone who says something like 'I must repudiate the entire language of my culture' is making a statement in the language she repudiates. She will be doing so even if she rephrases her repudiation in the form of a metaphorical, rather than a literal, use of the terms of that language. Alternatively: someone who wants not to talk about beings is compelled to spell out his intentions in -- what else? -- terms used to talk about beings. Any attempt to do anything of the sort which Heidegger wanted to do will trip itself up. So, Derrida concludes, we must try for something very similar to what Heidegger attempted, but also very different.4

Derrida thinks of Heidegger's attempt to express the ineffable as merely the latest and most frantic form of a vain struggle to break out of language by finding words which take their meaning directly from the world, from non-language. This struggle has been going on since the Greeks, but it is doomed because language is, as Saussure says, nothing but differences.5 That is, words have meaning only because of contrast-effects with other words. 'Red' means what it does only by contrast with 'blue', 'green', etc. 'Being' also means nothing except by contrast, not only with 'beings' but with 'Nature', 'God', 'Humanity', and indeed every other word in the language. No word can acquire meaning in the way in which philosophers from Aristotle to Bertrand Russell have hoped it might -- by being the unmediated expression of something non-linguistic (e.g., an emotion, a sense-datum, a physical object, an idea, a Platonic Form).6

Derrida says of the logocentric philosophers who hold out this hope of immediacy: 'Univocity is the essence, or better, the telos of language. No philosophy has ever renounced this Aristotelian ideal. This ideal is philosophy.' (Margins, p. 247) To succeed in twisting free of the logocentric tradition would be to write, and to read, in such a way as to renounce this ideal. To destroy the tradition would be to see all the texts of that tradition as self-delusive, because using language to do what language cannot do. Language itself, so to speak, can be relied upon to betray any attempt to transcend it (see Derrida, Writing, pp. 278-81).

[...] These claims have been the subject of hostile, sometimes bitter, criticism, from Derrida's fellow philosophers. Jacques Bouveresse in France and Jürgen Habermas in Germany have criticized Derrida severely. But the fiercest criticism of him has come from British and American analytic philosophers, members of the philosophical school which has dominated the English-speaking academic world since the Second World War. For many of these philosophers, heirs of a tradition which began with the logical positivists' opposition to metaphysics (not in the wide Heideggerian sense of the term, but in a narrower sense in which 'unverifiable' metaphysical, theological and moral claims are opposed to 'verifiable' scientific claims) Derrida's work seems a deplorable, frivolous, wicked, regression to irrationalism.

There are two main lines of criticism of Derridean philosophy. Those who take the first line see Derrida's doctrines as a sort of reductio ad absurdum of doubts about 'realism' -- about the claim that our language and thought are structured and given content by the world, by non-language. They treat Derrida as a linguistic idealist -- someone whose much-quoted slogan 'There is nothing outside the text'7 is supported by nothing more than the bad old arguments of Berkeley and Kant. One such critic, David Novitz, says that Derrida believes that 'our use of language is never constrained by a non-linguistic world' ('Rage', p.53), and that this conclusion does not follow from the fact that 'we cannot experience an object apart from our mental constructs', for that 'is just another way of saying that we cannot experience an object apart from our experience of it' ('Rage', p.50). As Novitz says, 'We still observe non-linguistic or non-semiotic objects in order to ascertain whether we have described them correctly'. The fact that we do so, he thinks, shows that 'there must be a non-linguistic, non-semiotic, non-constructed world ... one, moreover, which exercises some constraint on what we say, how we organise, differentiate and codify' (ibid. p.51).

The question raised by Novitz' criticism is whether the fact (which Derrida would hardly deny) that there are non-linguistic objects which constrain (in straightforwardly physical, causal ways) both our linguistic and our non-linguistic behaviour refutes the Derridean suggestion that, in Novitz' words, 'our concepts and meanings ... do not represent, convey, or correspond to a non-linguistic reality, a "transcendental signified"' ('Rage', p.49). 8 There is obviously a gap between 'X constrains Y' and 'Y represents, conveys or corresponds to X'. 'Realist' philosophers think that they can cross this gap. They think that the causal influence of the environment upon linguistic behaviour enables us to give a clear sense to the claim that some bits of language 'correspond' to something non-linguistic. Their opponents, both 'anti-realists' and those who try to set aside the realism/anti-realism issue as misconceived9 think that no such sense can be found.

A respectable body of opinion within analytic philosophy holds that the existence of causal relations between language and non-language does not suffice to give a sense to the notion of 'correspondence between language and reality'. This view seems implicit in Wittgenstein, and it is explicit in the work of contextualist philosophers of language such as Donald Davidson.10 So one can argue that Derrida's views are no more scandalous or absurd than those of these latter figures.11 On this argument, Wittgenstein, Davidson and Derrida have all preserved what was true in idealism while eschewing Berkeley's and Kant's suggestion that the material world is the creation of the human mind. Many followers of Wittgenstein and Davidson, however, sympathize with a second, somewhat milder, line of criticism of Derrida. According to this criticism, Derrida starts off from a philosophical position which rightly emphasizes the self-contained character of language, rightly holds that (in Wilfrid Sellars' words) 'all awareness is a linguistic affair' (Science, p. 160), and rightly repudiates what Davidson calls 'the scheme-content dichotomy.'12 However, he sets out this position in such an extravagant, hyperbolic, way that his misled followers pardonably, though fallaciously, derive silly consequences from it. John Searle, for example, sensibly remarks that the fact that language is a system of differences 'does nothing to undermine the distinction between presence and absence' for

I understand the differences between the two sentences 'the cat is on the mat' and 'the dog is on the mat' in precisely the way I do because the word 'cat' is present in the first while absent in the second, and the word 'dog' is present in the second, while absent from the first ... the system of differences is precisely a system of presences and absences.
('World', p.76)
More generally, one can argue, in the spirit of Wittgenstein, that philosophy 'leaves everything as it is' (Investigations, Part I, sect. 124) except for previous philosophy, and that abandoning the specifically metaphysical quest for univocity through confrontation with a non-linguistic referent (what Derrida calls 'the telos of language . . . this Aristotelian ideal') does not mean abandoning the every-day distinction between relatively univocal and relatively ambiguous uses of words.13 Again, to say that the objective-subjective distinction is relative to context and purpose is not to repudiate that distinction, but merely to caution against thinking that 'objective' can mean more than 'intersubjective'. Searle speaks for many analytic philosophers when he says that 'in the twentieth century, mostly under the influence of Wittgenstein and Heidegger, we have come to believe that this general search for these sorts of foundations [ontological, epistemological, or phenomenological foundations of the sort sought by Plato, Descartes or Husserl] is misguided'. But he insists that 'this doesn't threaten science, language, or common sense in the least.' ('World', p.77) For Searle, Derrida's anti-foundationalism is neither new nor particularly interesting. On his view, only the philosophical naivety of Derrida's followers lets them see anti-foundationalism as having startling consequences for literary criticism or for politics.

Searle here raises perhaps the most widespread objection to deconstructionism: why should we think that the abandonment of Platonic ideas and strivings would have important ramifications for the rest of culture?14 Why, for example, should we believe that, as Derrida (following Heidegger) insists, science has been constrained by 'metaphysical bonds that have borne on its definition and movement from its beginning'? Why not say instead (with, for example, Reichenbach, Popper and Dewey) that the natural sciences have done a lot to loosen those bonds, and to make possible a post-metaphysical culture? These questions cannot be pursued in this space, but raising them is useful for understanding Derrida's relation to the philosophical world of his time as well as for understanding the reception of deconstructionism by dubious or angry spectators of the movement.15


1. For an argument that notions like trace and différance come together to make up something rather like a philosophical system, see Gasché, Tain. This very thorough and impressive work argues that Derrida has been misread because of his appropriation by literary theorists, and that he needs to be restored to philosophy proper. (See especially p.3 on this point). For criticism of Gasché, see Rorty, 'Transcendental'.

2. For a good discussion of the difference between Derrida's orginal interests and the interests of his English-speaking followers, see Grumbrecht, 'Deconstruction deconstructed'. For the claim that deconstruction should not have been extended from metaphysics to literature, that it was a mistake to have taken 'a legitimate philosophical practice ... as a model for literary criticism', see Eco, 'Intentio', p. 166.

3. See de Man's reply to a request for a definition of 'deconstruction' by Robert Moynihan, in the latter's A Recent Imagining, p. 156: 'It's possible, within text, to frame a question or to undo assertions made in the text, by means of elements which are in the text, which frequently would be precisely structures that play off the rhetorical against grammatical elements.'

4. For Derrida's discussion of the similarities and differences between Heidegger's project and his own, see Margins, pp. 25-7 and 134-6. See also Megill, Prophets, chapter 7.

5. See Saussure, Course, chapter 4, sect. 4. The same point is made by Wittgenstein at many places in Philosophical Investigations.

6. This is not, of course, to say that there is no such thing as linguistic reference to non-language. But merely to repeat Wittgenstein's point that ostensive definition requires a lot of 'stage-setting'. The common-sense claim that 'There's a rabbit' is typically uttered in the presence of rabbits is undermined neither by Wittgenstein's point, nor by Quine's arguments about the inscrutability of reference, nor by Derrida's about the tendency of the signifier to slip away from the signified. For the impact of such arguments on the notion of meaning, see Stout, 'Meaning', and Wheeler, 'Extension'.

7. This phrase occurs at Derrida, Grammatology, p. 158. In its context it has a more specific and complicated sense than that usually attached to it by hostile commentators.

8. The phrase 'the transcendental signified' is one of Derrida's terms for an entity capable (per impossible) of halting the potential infinite regress of interpretations of signs by other signs. See Derrida, Grammatology, p.49, where he agrees with Pierce that nothing can stop such a regress.

9. For the distinction between these two types of philosopher, see Fine, 'Anti-Realism' and Rorty, 'Pragmatism', pp. 351-5.

10. See Davidson, 'Myth', p.165: 'Beliefs are true or false but they represent nothing. It is good to be rid of representations, and with them the correspondence theory of truth.'

11. On Derrida and Wittgenstein, see Grene, 'Derrida and Wittgenstein'; Staten, Wittgenstein; Rorty, 'Nutshell'. On Derrida and Davidson, see Wheeler, 'Extension' and 'Indeterminacy'.

12. See Davidson, 'Myth', p.163: 'Instead of saying that it is the scheme-content dichotomy that has dominated and defined the problems of modern philosophy Oe one could as well say that it is how the dualism of the objective and the subjective has been conceived ... [T]he most promising and interesting change that is occurring in philosophy today is that these dualisms are being questioned in new ways or are being radically reworked.' Such passages in the writings of Davidson, Putnam and other analytic philosophers parallel the decontructionists' attacks on 'the traditional binary oppositions'.

13. A similar point is made by Robert Scholes at pp. 67-73 of Protocols. Scholes is concerned to distinguish a metaphysical from a pragmatic sense of 'presence', and to argue that scepticism about the former is irrelevant to the latter.

14. See Stout, 'Relativity', pp.109-10, and Rorty, 'Circumvention', pp.20-1 for two ways of posing, and expanding on, this rhetorical question.

15. For dubeity, see Abrams, 'How to do Things'; for angrier reactions, see Hirsch, Aims, p.13, and Bate, 'Crisis'.

©1995, Cambridge University Press


Top of Page || Home Page || Stanford University Libraries || Stanford University