Preliminary to a note on Derrida and Rule-Following

There’s an intriguing piece on Robert Brandom and Jacques Derrida by Duncan Law here. The upshot of his piece is that there is a parallel between Brandom’s deontic account of semantic rules and Derrida’s notion of iterability. Thus Brandom claims that objectivity in rule following is constituted by our capacity to evaluate certain performance as inferentially correct rather than by the existence of objective semantic rules. Meanwhile Derrida’s iterability argument shows that objective reference (the relationship between ‘signifier and signified’ we insist on speaking Structuralese) is constituted by the structural possibility of repetition by different time-bound subjects, rather than by the existence of semantic relations between signs and referents. Law thinks, however, that Brandom’s position is superior because ‘vulnerable to empirically-based contestation in a way that the concepts of differance or messianicity aren’t’.

My qualifications for commenting on Brandom are few, being based mainly on acquaintance with the secondary literature. However, Derrida I can claim to know pretty well, so I’ll confine this post to a brief demurral from this transcendentalist reading of deconstruction. It’s fairly tempting to read Derridean deconstruction as a kind of ‘re-tooled’ phenomenology. But Derrida’s arguments undermine phenomenology’s claim to uncover the constitutive conditions of objectivity. They do this, roughly, by showing that the structures which phenomenology takes to constitute the relationship between subject and object (temporality, say) can’t be phenomenologizable at all if they are to do their constituting.

From this, we might want to infer that ‘differance’, ‘iterability’ and the like are just deeper transcendental conditions, churning away in some abscess beyond the reach of theory, objectivity, etc… This reading might be warranted by passages where Derrida apes the rhetoric of transcendental philosophy. But the idea of iterability as an ultra-transcendental wheel only works if we attach credence to phenomenological givenness being what determines objectivity in the first place. Derrida (if his arguments go through successfully) has already undermined this claim: subjectivity (surprise surprise) ain’t up to the job! So what we are left with are  topic-neutral structures such as iterability and supplementarity with no special appurtenance to phenomenology, or to anything else. Their main philosophical interest lies in their capacity to make theoretical openings in areas like ethics and semantics, but these can be appropriated in a realist, metaphysical spirit, as I’ve attempted to show here , for example.

Derrida’s iterability argument may have an advantage over Brandom’s due to its extreme abstraction. As I’ve argued elsewhere, whatever its transcendental inspiration, the logic of the iterability argument is absolutely general. Derrida puts the first premise of the IA with uncharacteristic pith in Speech and Phenomena (50):

‘A sign is never an event, if by event we mean an irreplaceable and irreversible empirical particular. A sign which would take place but “once” would not be a sign; a purely idiomatic sign would not be a sign.’

Metaphysically this is plain and simple fare: all it presupposes is the ideal repeatability of meaningful states (and not just linguistic items).

The argument applies with strict generality to any kind of ‘text’ (e.g. mental representations). Moreover, whereas Brandom, on my limited understanding of him, requires the semantic content of a term to supervene on its normative status within a given community, Derrida’s argument is equally compatible with normativist and dispositionalist theories of rule following. If the meaning of a three place open sentence like ‘…+…=…’ is conferred by the disposition to apply it to the triples {1, 1, 2}, {1,2, 3}…, etc. it still remains true that the open sentence in question must be have an identity transcending each event of its application. This is true, even if, as C.B. Martin and John Heil have argued, the fact of being disposed to apply the predicate  to this infinite set of triples and not some other is ontologically primitive and irreducible to any list of its actual or counter-factual manifestations.

Tracing the implications of the full-dress iterability argument for the rule following debate, though, would take a much longer post than this. I’ll try to return to this topic later in the year.

4 thoughts on “Preliminary to a note on Derrida and Rule-Following

  1. Hi David – thanks for the link and these thoughtful comments. I’m still running behind in trying to catch up on your work, so this is very much a first pass and I’ll try to respond more adequately when I’m better informed. Very quickly though, I definitely agree with this –

    Derrida’s arguments undermine phenomenology’s claim to uncover the constitutive conditions of objectivity

    – and broadly speaking, I think, with the exposition of Derrida that follows. As a first pass at articulating my take on the structure of Derrida’s argument, I like the phrase in Geoffrey Bennington’s book on/with Derrida, something like for Derrida “the empirical is the transcendental of the transcendental (of the empirical)”. I.e. (crassly) I see Derrida as trying to undermine the idea of a constituting transcendental subjectivity, by showing the ways in which that subjectivity’s phenomenal self-relation is itself dependent upon / constituted by a non-subjective something else. This definitely puts Derrida at some distance from phenomenology, even a radicalised phenomenology, at least as I understand it.

    My objection to Derrida’s argument therefore isn’t really that he’s a straightforward (or indeed hyper-) transcendental philosopher – but that Derrida’s critique of phenomenology relies on a concept of constitution that itself is dependent on aspects of the transcendental apparatus that Derrida’s argument also rejects. I think this results, in Derrida’s work, in a vacillation between two alternative theoretical directions: on the one hand a thoroughgoing critique of the transcendental concept of constitution – a critique that I think pushes Derrida in a pretty naturalistic direction (which I like) (and this is the space that a lot of his work on animality and so on is coming from, I think). On the other hand I think that, given the premises of his argument, and if the deconstructive operation is to function, Derrida is unable fully to give up a concept of constitution that relies on transcendental resources, and this results (as you put it in your post) in Derrida at times writing as if differance etc. are really deeper transcendental conditions (which takes him into a much more theological space). I don’t think Derrida ever fullly endorses that move – he’s consistently, if ambivalently, critical of it. But I claim that there’s a double-bind in the structure of Derrida’s argument that makes his theoretical apparatus unable fully to reject this theoretical pole: there’s an unresolvable-in-Derrida’s-terms movement between the anti- and the hyper-transcendental in Derrida’s work. I think we need to start from a different theoretical place from Derrida, basically, if we’re going to break that double-bind.

    I put this about as coherently as I ever have in this post & comment thread (especially my last comment there) from my old blog –

    http://praxisblog.wordpress.com/2008/05/25/matter-sign/

    – although I wrote that post quite some time ago, and I’m not altogether happy with everything I say there. (Also, that post was part of an ongoing debate about Specters of Marx, which makes quite a lot of it pretty opaque & irrelevant for our purposes.)

    I should also emphasise that my reading of Derrida is vulnerable to the fact that my knowledge of Husserl is very slight & superficial. It’s true that all I strictly need to know to talk about this stuff is what Derrida thinks Husserl’s doing :-P. But if I want to discuss this stuff with much authority I really ought to read Husserl himself properly, and I don’t see myself making time in my reading schedule to do that any time soon. So…

    Btw, I don’t suppose you happen to know the work of Henry Staten? I mention it just because his comparative study of Derrida and Wittgenstein was really helpful & important to me when I was working through Derrida, and also because Staten pushes a naturalistic read of Derrida – he’s written more recently on synthesising Derrida & Dennett, though I can’t seem to access the paper here.

    Anyway, many thanks for these thoughts, and I’ll try to say more once I’ve actually read your work on all this… Apologies if I’ve missed the real issues through hastiness…

  2. ‘My objection to Derrida’s argument therefore isn’t really that he’s a straightforward (or indeed hyper-) transcendental philosopher – but that Derrida’s critique of phenomenology relies on a concept of constitution that itself is dependent on aspects of the transcendental apparatus that Derrida’s argument also rejects.’

    Yeah, I’d agree with this and with the charges of vacillation. Nonetheless, what I always liked about JD is that is best work is very intricately through-argued. The iterability arguments, for example, and the stuff on the trace can be applied outside the field of subject-centered transcendental philosophy – as Paul Cilliers has shown with regard to the the interpretation of artificial neural networks.

    I really like Henry Staten’s work on Derrida. He’s an avowed naturalist. I found Wittgenstein and Derrida immensely helpful and there’s a terrific piece by him on Dennett – ‘Derrida, Dennett and the Ethico-Political Project of Naturalism’ that appropriates him along the lines you have explored here:

    http://www.deepdyve.com/lp/edinburgh-university-press/derrida-dennett-and-the-ethico-political-project-of-naturalism-xVzu4KxlsY

  3. Ooh yes, that’s the paper I had in mind, I should reread it. I’m chuffed that you like Staten – he was really important to me in working through a lot of stuff – Wittgenstein and Derrida, and also his Nietzsche book, were the texts that opened the continental theory space for me, having started in more analytic stuff – but I always feel that he’s not a well-known enough figure for me to use as a reference point.

    Anyway, I think we’re largely in agreement about Derrida? I read your Naturalising Deconstruction paper last night – very clever stuff, thank you – though it’ll take me some time to process it. But I think your endnote 4 puts it well. This certainly isn’t the only way one could appropriate / develop Derrida, but it’s likely to be one of the most productive, I think.

    Thanks…

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s