Two Can Play at this Game

An uncharitable reading of J.D.'s refusal to define his terms is that he's bluffing. A charitable reading is to say that he's doing it on purpose because he's thinks making precise philosophical definitions is a fool's game. No one can actually do it, so why pretend–it's more honest to just start using a technical term like "trace" and let it accrue whatever meaning it's going to accrue. (More honest in the way that writing Being is more honest.) Maybe this is a valid point about philosophical nomenclature, except the definition-by-ostension strategy seems parasitic on someone else's definition-by-describing-what-you-mean strategy. For example, the whole hullabaloo about "traces" falls flat unless we already have the clearly delineated Saussurian distinction between /b/ and /p/ in mind.

Hey, can I push this a little further and deconstruct Derrida? Let's try.

1. Tacit dichotomy...J.D.'s definition-by-ostension differs from other philosophers' definition-by-description.
2. Privileged member of dichotomous pair...Definition-by-description is for chumps. The only way to introduce a philosophical term is to just start using it.
3. Self-undermining nature of dichotomous privileging...Definition-by-ostention only works if there are definition-by-description terms to play off of.

And...gotcha! That was kinda fun, but ultimately unsatisfying.

Of course another possibility is that J.D. isn't doing anything particularly tricky here and I'm simply failing to follow his argument.

Bluffing?

A thought coalesced while I was looking back over the last few posts: Derrida introduces philosophical concepts like "transcendental signified" or "trace" that he clearly intends to be extensions of or rebuttals to more familiar concepts like "referent" or "structural distinction". J.D.'s new terms bear a family resemblance to the ones they supplant and so partake of those terms' long-established significance. But because J.D. keeps his definitions murky, it's not clear where his contribution lies. It's like he says, "You know that old idea of 'referent', well I've got this thing called 'transcendental signified' which is like referent, except better!" And then he's off on a tear and never gets around to telling you why it's better. The uncharitable reading of this technique is that he's bluffing.

Without a [[Trace]]

Derrida introduces the crucial and elusive technical notion of a "trace" in passing. Not only does he fail to define it, but he doesn't even give any indication that he's introduced a technical term. It's just all of a sudden there in a who-invited-this-guy-and-why-is-he-hogging-all-the-snacks kinda way. Sigh.

Piecing together the meaning of "trace" from context, it's something like a gap in a semiotic system, a telling patch of negative space. Because I can't figure out what J.D. is getting at beyond this sketchy surmise, I find myself falling back on structuralist metaphors: a trace is an element of a symbolic system that takes on meaning by virtue of being different from the elements around it. What matters about a /d/ is that it's not a /t/ or a /ɖ/, and so on. Basically, my definition comes straight from Saussure, which is probably not what was intended, but since certain 20th-century Franco-Algerian philosophers I could name think defining one's terms is just so square I guess it'll have to do.