I've now reached chapter 4, 'Meaning and Representation'. This is the chapter I read during the hot weather a few weeks ago, and my notes are terrible. It also looks like an important chapter, oh dear. Before I start I'll just link this twitter conversation on structuralism that I had with David Chapman and others, for future reference.
There's some more difficult stuff at the start of the chapter. I'm going to skip forward a couple of pages and start with this quote about repeatability that I liked. Signs need to be repeatable rather than one-off unique events, they need to be somewhat abstracted from the flux of experience:
When in fact I effectively use words, and whether or not I do it for communicative ends (let us consider signs in general, prior to this distinction), I must from the outset operate (within) a structure of repetition whose basic element can only be representative. 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. A signifier (in general) must be formally recognizable in spite of, and through, the diversity of empirical characteristics which may modify it.
This is going to be important for understanding his différance concept I think. But also I find this idea of repeatability interesting in general. One of the things I was looking for in the visual programming post was what sort of repeatable 'grammars' of elements were present in non-text-based languages. Rules for connecting lines and boxes, for instance.
It must remain the same, and be able to be repeated as such, despite and across the deformations which the empirical event necessarily makes it undergo. A phoneme or grapheme is necessarily always to some extent different each time that it is presented in an operation or a perception. But, it can function as a sign, and in general as language, only if a formal identity enables it to be issued again and to be recognized.
I'm still muddled on how repeatability connects back to indicative vs expressive signs, and to the monologue argument. And I'm also confused by what Derrida/Husserl mean by 'representation'. I think I need some help from a secondary text at this point... googling around, Derrida's Voice and Phenomenon by Vernon W. Cisney has a couple of recommendations from other people I've found helpful. Got hold of a Kindle copy and I'll have a try at that before tackling any more.