The last two quotes Derrida provides in "The Inside and the Outside" won me over to his side a bit. These are the ones where Saussure complains about the way the writing system is damaging the pronunciation of French, splitting the surname Lefèvre into Lefèvre and Lefèbvre and leading people to pronounce the last two letters of vingt. Saussure depicts this as a decline from pre-existing correct system of French pronunciation, instead of just another instance of language evolution. Any contemporary linguist will tell you that there is no "correct" pronunciation of a language beyond whatever the current consensus among its speakers calls correct, and that consensus is always changing. (Remember that French is nothing but the noble Latin of the Roman emperors after a thousand years of abuse in the mouths of a bunch of Gallic barbarians.) By assuming that there is some objective standard of correctness (which, I'll bet you dollars to donuts, just happens to coincide with the way Saussure says Lefèvre and vingt) Saussure in engaging in what linguists call prescriptivism, which is a mistake when you're trying to be empirical. Furthermore, the excerpts make it clear that this isn't just a matter of tone, but a well-definied assertion. When Derrida shrugs at this and says "Where is the evil?" he's doing exactly what I would do in front of a Linguistics 101 class.
From a modern linguist's perspective it seems odd that Derrida would treat Saussure's prescriptivism as one element in a larger defense of writing, rather than seeing the business about writing as just one aspect of pernicious prescriptivism, but whatever. Derrida's interest is writing here, and it does appear that Saussure focused on writing as well.
Showing posts with label linguistics lesson. Show all posts
Showing posts with label linguistics lesson. Show all posts
Saturday, March 7, 2009
Tuesday, March 3, 2009
The Outside and the Inside of the Lexicon
Derrida makes a good point about words not being the sole locus of meaning. He paraphrases André Martinet approvingly:
So, good point about words, but us linguists are already all over it.
[It's a mitzvah] to make [the usage of the concept of the word] more flexible, to associate it with the concepts of smaller or greater units (monemes or syntagams).Non-linguists often find it surprising that the meaning-bearing elements of language can be smaller or larger than words. To linguists, however, this is old hat. In a Linguistics 101 class you'll be taught that morphemes (what Martinet appears to be calling "monemes"), not words, are the smallest unit of meaning. Going in the other direction, many linguists will identify certain syntactic constructions (Martinet's "syntagams" I think) as meaning-bearing units. For example, in Foundations of Language Ray Jackendoff argues that structures like "X-ed his way across the room" and "X-ed her heart out" are atomic units of English meaning. Construction Grammar (about which I know little) moves this idea front and center. Linguists often frame this question in terms of what kinds of things belong "in the lexicon".
–pg. 31
So, good point about words, but us linguists are already all over it.
Saturday, February 21, 2009
The Phonetic Nature of Writing

Since Derrida seems to be calling attention to phonetic writing systems in particular I just want to take this opportunity to to point out that there are plenty of non-phonetic writing systems in the world. Writing systems vary greatly in their degree of phonetic accuracy, from the non-phonetic (Chinese), to ones that omit large amounts of phonetic information (Arabic), to ones that are basically phonetic but have tricky spelling rules (English, French), to ones that are pretty faithful to the acoustic signal (Spanish, Korean). But even those at the far end of the spectrum aren't completely phonetic. Even the International Phonetic Alphabet devised by linguists for the purposes of transcribing speech sounds can't capture every subtle variation in human speech. Not sure where he's going with the phonetic stuff, but I'm going to keep an eye out in case Derrida makes any sweeping statements about human language which really only apply to French and English. (He wouldn't be the first.)
The discussion of the limitations of the IPA does remind me of something I've said to introductory linguistics students: the only perfect transcription is an audio recording. Or, joking aside, any writing system has to be somewhat faithless to the audio signal, it has to omit some acoustic information, in order to qualify as a writing system. This seems similar to some of the presence vs. absence stuff Derrida discusses, but I can't tell if this is a true analogy or just a surface similarity because both are counterintuitive statements about language.
An Out of Band X
In the discussion of Heidegger's technique of putting an X through "Being" (Translator's Preface pg. xv) the idea seems to be this: at some point in philosophy we're going to encounter concepts that are so abstruse that we'll never be able to adequately talk about them. The usual strategy is to invent technical terms for these concepts, but that makes them look like things we can talk about. It's more honest to just draw a big X through the word, to draw attention to the fact that such-and-such a concept is a tough nut to crack. Apparently Derrida is going to embrace and extend this idea in his own notion of erasure (sous rature).
Heidegger's X is what computer protocol people would call out-of-band information. The standard channel is conventional printing, while the out-of-band channel is created by the typographical novelty of a strikethrough, though here Heidegger crosses out the word "Being", whose odd capitalization is itself already a typographical novelty. (At least it is in English; I'm not sure how this comes across in German.) Note that this only works because strikethrough is relatively rare. If everybody started drawing X's through words willy-nilly, erasure would become just another part of the writing system. Arguably this is what has already happened to shudder quotes.
("Arguably" this "is" what has already "happened" to shudder quotes.)
Spivak shows examples of Derrida putting "is" under erasure. Derrida must be aware that copula be gets expressed in many different ways in the world's languages. Many people just don't say "is" so there's nothing to erase.
Heidegger's X is what computer protocol people would call out-of-band information. The standard channel is conventional printing, while the out-of-band channel is created by the typographical novelty of a strikethrough, though here Heidegger crosses out the word "Being", whose odd capitalization is itself already a typographical novelty. (At least it is in English; I'm not sure how this comes across in German.) Note that this only works because strikethrough is relatively rare. If everybody started drawing X's through words willy-nilly, erasure would become just another part of the writing system. Arguably this is what has already happened to shudder quotes.
("Arguably" this "is" what has already "happened" to shudder quotes.)
Spivak shows examples of Derrida putting "is" under erasure. Derrida must be aware that copula be gets expressed in many different ways in the world's languages. Many people just don't say "is" so there's nothing to erase.
Subscribe to:
Posts (Atom)