5 min read

Broaching and breaching

This is an email braindump that I've edited a bit for the notebook so that it can go in one place with all my other derridaposting. It's about Derrida's idea of iterability, his fight with Searle over Austin's speech acts, and similarities with Brian Cantwell Smith's idea of the 'middle distance'.

I wrote it fast and it's kind of sloppy, the details might be off.

So... a couple of months ago I came across this paper by Gordon Bearn, Derrida Dry: Iterating Iterability Analytically. This paper dates from 1995 and the context is this big fight Derrida had with Searle. I knew about that already but didn't realise it dragged on for so long:

Timeline of Derrida and Searle's argument, from 1971 to 1994

I think 1977 marks the real start of the argument, but wow, even counting from there it's almost old enough to vote and they're still having it. Bearn is sympathetic to Derrida but trying to write something more palatable to analytic types, by recasting Derrida's original arguments from Signature Event Context in a sort of string-of-propositions logical form. On the way he ends up saying some interesting things about iterability, and I found that part useful.

So, to back up a bit, what's the whole argument about? I only have a rough understanding, but let's have a try at explaining it anyway. In Signature Event Context Derrida is responding to Austin's How To Do Things With Words. I haven't read Austin, but my rough impression is that he was engaged in a project to own the positivists by pointing out that not all statements are things that can have a 'true' or 'false' label slapped on them. His main set of examples were 'performative utterances' like "The court is now in session' that bring about the action they refer to. Instead of 'true' and 'false', he talked about these utterances being 'happy' (did the thing) or 'unhappy' (didn't do the thing), and he was coming up with some taxonomy of unhappy utterances.

So, in the process he decided to exclude 'nonserious' unhappy utterances, where people say 'The court is now in session' without intending any such act at all... maybe as part of a joke, or because it's their line in a play they're rehearsing, or because they're reading a boring book out loud in class while zoning out.

Austin describes these kinds of nonserious utterances as 'parasitic' upon the usual kind. Derrida absolutely lives for pulling apart these kinds of dichotomies, and this is where he brings in his idea of iterability.

Iterability is the ability for some piece of speech or writing to be taken out of its original context and repurposed for something else. It's not inextricably locked to the original speaker/writer and the intended listener/reader, and can function without them. This iterability allows 'The court is now in session' to be reused in many situations without the animating breath of the original intended meaning needing to be there, and prevents the possibility of its meaning being restricted to 'non-parasitic' utterances. Derrida sometimes refers to this as the 'free play' of language.

The fight with Searle was something like... Searle interprets Derrida's talk of 'free play' as a kind of extreme relativism, and points out how language is actually pretty stable in practice, and thinks he's just spouting obscurantist nonsense that a reading of Wittgenstein would clear up. And then Derrida responds with his most trollish and annoying prose, and then it just keeps going. Or something? I don't currently care much about the fight, so I don't know the details. I'm much more interested in the idea of iterability itself.

The thing I really liked from the Bearn paper is his discussion of broaching and breaching. This comes from an inspired translation by Samuel Weber of a sentence in Derrida's Limited, Inc (one of the responses in the Derrida-Searle fight):

Iterability is at once the condition and the limit of mastery [of our mastery of language]: it broaches and breaches it [elle l'entame]

Entamer seems to mean a lot of things, but they include 'to start' (broach) and 'to cut into' (breach), and Derrida wants both of these shades of meaning:

  • Iterability broaches linguistic action because it provides recognisable, reusable lumps of text. This is particularly important for performative utterances because they have to be identifiable as the correct thing. (I don't know how courts work, but I'm guessing that 'I reckon we should start about now' doesn't have the same effect.) Signatures are another good example that Derrida uses: they have to be recognisably 'the same' each time to accomplish their purpose of e.g. declaring a contract valid.
  • Iterability also breaches linguistic action. Because the text is reusable, it can escape its original context and be used for other things. A performative utterance can't always be guaranteed to work every time it's used - it might just be someone reading the sentence out of a book. Derrida is mostly concerned about breaching in Limited Inc, because it disrupts Austin's attempts to avoid 'parasitic' uses of the text.

There's another interesting connection that jumped out at me reading his discussion of breaching:

... even in the ideal case considered by the strategy [of excluding the nonserious]; there must be a certain element of play, a certain remove, a certain degree of independence with regard to the origin, to production, or to intention in all of its "vital, " "simple," "actuality," or "determinateness, " etc. For if this were not so, the "break". . . would be impossible. And if a certain "break" is always possible, that with which it breaks must necessarily bear the mark of this possibility inscribed in its structure.

I've been reading Brian Cantwell Smith again recently, and this is very similar to his idea of the middle distance (particularly the bit I bolded). He talks about representation as involving a kind of 'flex or slop': detached enough from the thing it represents to not just be causally locked to it, but still attached enough to stay in some kind of relation to it. (If that sounds completely abstract, I talked through some concrete examples here).

As a thought experiment he imagines two possible worlds where representation can't get off the ground: a world of non-interacting randomly drifting particles where nothing is stable enough to broach a connection, and a world of precisely interlocked cogs where nothing has the slack to breach a connection. Neither world has the resources required for useful representation. (It's kind of obvious why particle world is useless. For cogs world, you're e.g. missing the ability to manipulate your representations in a lightweight way. In our world, you can update a row in House table in your database without having to drag the actual house around.)

So the analogy between Smith's ideas and Derrida's is a close one which matches in detail. I wasn't hallucinating things when I wondered about this a year ago. (I was trying to match on différance back then, but I think iterability is better now. Need to get clearer on how they're related.)

Is the analogy interesting, though? When something's this precise of a match, my instinct is that there's something worthwhile there, but I'm not sure. Maybe this is a pretty general train of thought and a bunch of other people have similar ideas about representation, and Derrida and Smith are just part of this overall current, in which case it wouldn't be surprising that they sound similar. I guess that would still be interesting to me, because then I could read the other people too, but I wouldn't expect investigating this particular link to be very productive.
I've tried searching, and haven't found anyone else making this connection. Though Smith is not an easy surname to search for, so if it appears without the Cantwell bit I might well have missed it. If you have any leads, let me know.