Over at Larval Subjects Levi Bryant has reasserted his contention that Derrida is some kind of linguistic idealist. Here’s a representative quote:
To me it seems that Derrida’s core thesis is that reality is structured by the signifier for humans. At the heart of this claim are the two central theses of structuralist linguistics that 1) signifiers are differentially constituted (hence all the stuff about differance and the trace), and 2) that signifiers only refer to other signifiers. As Lacan articulated the latter point, “the signifier represents the subject for another signifier”. Within this framework, then, the referential function is undermined as language only refers to language, never to world. Derrida constantly explores the limits of language– hence his interest in Levinas –but only, it seems to me, as opening on to an Other that exceeds all possibility of being articulated in language.
Methodologically, the conclusion that follows from this is that talk about the world can only ever be talk about talk or language. This can take the form of the analysis of texts, of speech, of the semiotics of clothing, etc. But working in the background is always the thesis that reality is a signifying construction. To talk about an object such as the telephone on my desk is thus not to talk about my telephone, but about a signifying construction (here reference should be made to Derrida’s discussions of manifestation and phenomena vis a vis Peircian semiotics in Of Grammatology). So sure, Derrida talks about hospitality, cosmopolitanism, friendship/fraternity, etc. (whoever suggested otherwise), but isn’t this always on the horizon of a nominalism where reality is linguistically constructed and where we are inextricably trapped within language? Put a little bit differently, of course Derrida talks about objects all the time (coins, cats, dogs, weapons, etc), but he can do this because all of these objects are texts. His dialectic thus unfolds perpetually around a play between the trace of an Other perpetually withdrawing from language and unreachable and the text of being that’s a signifying construction.
Aping high Derridean style, I should declare the absence of a position before considering these claims in detail. I began losing interest in any systematic take on Derrida’s work around the time Spectres of Marx was published in English. Here was a text purporting to be about Marx which seemed unable to address Capitalist social structure and dynamics. Bien pensant left-liberal sentiments and utopian longing are no substitute for thought-through philosophy or empirically responsive social theory (Writing a PhD on Derrida and Metaphysics also had a marvelous prophylactic effect).
I still think early Derrida is an interesting philosopher, but in the way that John Cage is an interesting composer. Derrida’s published work contains a functional assortment of tools, arguments and strategies – a kind of philosophical entrencher. There may also be a systematic philosophy in there as well. But where it emerges, it tends to be extirpated Terminator-style by the engine before it can grow up and mount the anti-machine resistance.
And that’s a good thing!
Let’s take Derrida’s putative linguistic idealism as a case in point. What form might it take? Levi claims that Derrida is a nominalist:
Derrida talks about hospitality, cosmopolitanism, friendship/fraternity, etc. (whoever suggested otherwise), but isn’t this always on the horizon of a nominalism where reality is linguistically constructed and where we are inextricably trapped within language? Put a little bit differently, of course Derrida talks about objects all the time (coins, cats, dogs, weapons, etc), but he can do this because all of these objects are texts.
Nominalism is qualified anti-realism. Nominalists deny the existence of certain classes of abstract objects or universals: e.g. essences, propositions or sets. The relevant form of nominalism here is presumably the denial that there are essences, kinds or other such abstract structures in nature. The only real distinctions are those between particulars.
It follows that insofar as the world is arranged into kinds this organization must come from without: e.g. by God, a transcendental subject or language. All kinds are ‘unnatural’ (see Frank Farrell’s excellent Subjectivity, Realism and Postmodernism: The Recovery of the World in Recent Philosophy for a fascinating analysis of nominalism in modern philosophy, from Ockham to Foucault).
Note that this form of linguistic idealism does not amount to a denial that the world exists. For nominalism the world is an unstructured bundle of individuals. So if Derrida is a linguistic idealist qua nominalist it is because he claims that all distinctions of kind are imposed by a demiurgic ‘language’.
Let’s, for a moment, waive the obvious comeback to this: namely, that languages – according to the nominalist account – ought to be just further collections of particulars and in no condition to impose form on the formless.
Is Derrida committed to this position?
Well, Derrida can’t consistently assert that reality is structured by the signifier because, like Donald Davidson in ‘A Nice Derangement of Epitaphs’, he denies there is such a thing as language (where by ‘language’ here is meant something like Saussure’s langue: a shared structure which constitutes the semantic identity of individual utterances). This idea of language runs afoul of Derrida’s principle of iterability, an idea to which he adverts throughout his work:
Every sign, linguistic or nonlinguistic, spoken or written (in the usual sense of this opposition), in a small or large unit, can be cited, put between quotation marks: in so doing it can break with every given context, engendering an infinity of new contexts in a manner that is absolutely illimitable . . . This citationality, this duplication or duplicity, this iterability of the mark is neither an accident nor an anomaly, it is that (normal/abnormal) without which a mark could not even have a function called ‘normal’ (Derrida 1988, p. 12).
If any mark is iterable, then no mark belongs to a language strictly speaking. Take any set of semantic rules S which we can take to constitute the semantic nature of a mark M. Iterability states that M can be employed contrary to S. For example, suppose an anti-nominalist about language claim that S for ‘If’ consists in the grammatical and logical rules governing its employment in English conditional constructions (“If….then…” clauses). So no tokening which violates these principles is really a token of ‘If’ (note that this implies obvious problems for quotational contexts of the kind used here). Derrida points out that all words or signifiers can be used in deviant contexts – indeed if they couldn’t be used deviantly, they could be used period. For example, “If’ is used without a conditional construction in the title of Lindsay Anderson’s eponymous film.
The same goes for inferential roles. We might think that ‘mass’ is just fixed by its use within a particular physical theory. But ‘mass’ crops up in both Newtonian Mechanics and Relativity but occupies discrepant inferential roles in each (there is no distinction between rest mass and relativistic mass in Newton’s physics). Thus far, I think, Derrida’s position is not only well argued. It is plausible. But it tells against any reification of language of the kind made by structuralists. Of course, we can always expand S to include the deviant cases. But iterability just applies to those rule sets recursively.
Like Donald Davidson, then, Derrida denies there is such a thing as language qua langue. If any mark is iterable, then no mark belongs to a language strictly speaking. Languages are reifications which, for Donald Davidson, allow us to construct theories of meaning to engage with consistent or idiomatic speech behaviour. There may, of course, be some loose semantic conventions and habits, but our ability to engage with the likes of Joyce and Mrs Malaprop show that we can revise our theory of what language our interlocutor is speaking in line with the Principle of Charity (Davidon 1986).
So the initial rejoinder is spot on for reasons that go to the heart of Derrida’s philosophy. If nominalism is true, then we won’t get our missing structure from language because we can’t.
Could Derrida’s linguistic idealism take some other form? Well, Levi also accuses Derrida of a kind of pantextualism. As he puts it: Derrida talks about objects (cats, coins, etc.) because these objects are in the end just texts.
It should be obvious now that pantextualism and nominalism are different doctrines, so moving indifferently between the two accusations, as Levi does, amounts to a conflation. However, there are grounds for accusing Derrida of pantextualism. By ‘text’, of course, Derrida doesn’t mean ‘writing’ or linguistic texts in any empirical sense. He means something that instances the nuclear traits of all generalized writing (Derrida 1988, pp. 8-9). Words are instances of general textuality, but so are any kind of representation or mark. Indeed, the totality of experience is textual in this sense insofar as the content of any mental state is also constituted by its ideal repeatability, hence its iterability (Derrida 1988, p. 10).
What about cats and coins? Well, Derrida, like Davidson, is committed to some version of the inscrutability of reference. This is a consequence of iterability. If the identity of the language one speaks is fixed by interpretative convenience, then equally good references schemes can, in principle, be constructed for the same utterances – even if this is notoriously hard to achieve in practice (ditto of course, for experiences, mental representations, etc.). For example, semantic theory ST1 could assign my neighbors’ cat as the referent of the singular term ‘Marmite’ while another, ST2, might treat Marmite as a set of undetached cat parts. From the perspective of radical interpretation, there is nothing to commend one ontology over the other if both cope equally well with predicting global speech behaviour. Where does this leave reality? Here’s another quote from Limited Inc, which seems to back Levi up:
What I call ‘text’ implies all the structures called ‘real’, ‘economic’, ‘historical’, socio-institutional, in short: all possible referents. Another way of recalling once again that ‘there is nothing outside the text’. That does not mean that all referents are suspended, denied, or enclosed in a book as people have claimed, or have been naive enough to believe and to have accused me of believing. But it does mean that every referent, all reality, has the structure of a differential trace, and that one cannot refer to this ‘real’ except in an interpretative experience. The latter neither yields meaning no assumes it except in a movement of differential referring. That’s all (Derrida 1988, p. 148).
This sounds like pantextualism to me too! But at this point there are (at least) two realist rejoinders:
1) Derrida is being confused and confusing. The fact that reference or satisfaction (for predicate terms) is interpretation-relative or (more strictly) language-relative does not entail that referents are (Roden 2004). The only reason anyone might think is that they been raised in a philosophical tradition which treats reference as akin to transcendental constitution (ie. phenomenology). But Derrida’s engine is corrosive of phenomenology in its transcendental guise, so if this is what is being suggested, he’s just being inconsistent with his best insights (Roden 2006).
2) Another way through is what I understand to be Martin Hagglund’s route (his book is still on my shelf of shame, sadly) which is to take the trace structure as a kind of formal ontology of temporality. If this proves to be the right way to read Derrida, then pantextualism need be no less anti-realist than object oriented ontology. They are just competing theories of the real.
Davidson, Donald, (1986). ‘A Nice Derangement of Epitaphs’, in Ernest LePore (ed.) Truth and Interpretation: Perspectives on the Philosophy of Donald Davidson (Oxford: Blackwell).
Derrida, J. (1988). Limited inc. Samuel Weber (trans.). Northwestern University Press.
Roden, D. (2006). Naturalising deconstruction, Continental Philosophy Review 38(1-2): 71-88.
Roden, David. (2004), ‘Radical Quotation and Real Repetition’, Ratio: An international journal of analytic philosophy, XVII/2 (2004), pp. 191–206.