[xmca] Grammar Out of Discourse

From: David Kellogg <vaughndogblack who-is-at yahoo.com>
Date: Mon May 21 2007 - 13:40:19 PDT

Dear Paul:
  Thanks for your extremely interesting notes on Quechua, a language with which I'm completely unfamiliar.
  Korean does not have free-floating particles of any kind either (that is, there are no purely structural elements, with no clear real world referent, that serve only to bind other elements together, which are orthographically separate words). In Korean, and in all SOV languages I know, the particles are always "suffixes".
  The Universal Grammar (Principles and parameters) model explanation for this is that the "parameter" of the language is set to "head last". The head of a sentence is a verb, which is why the verb is last, and the head of a noun phrase is (according to the theory!) an article or determiner, which is why "a" and "the" have to come first in an English noun phrase and similar particles come last in Korean and Quechua.
  The UG model even extends this to syllable structure, so for example the standard model of an English syllable is an onset and a rime, where the "onset" is the "head" of the syllable. So for example the syllable "man" in English is in some vague sense "about" the consonant "m" rather than the ending "an" (this is what distinguishes it from "pa", for example!)
  But in Korean the syllable structure is the onset and the "coda", because what would be a rime in English is the head of the syllable in Korean (so Koreans tend to think of "man" as "ma" + "n" rather than "m" + "an" and the orthography does in fact reflect this, with "ma" written together at the top of the character and "n" written alone at the bottom.)
  One of my colleagues has argued that Korean poetry does not employ rhyme as English poetry does and that Korean poems are more concerned with alliteration. I have discovered that this is not strictly true. It is certainly true that rhyme does not play as important a role in Korean as it does in English, but concern with rhyme has waxed and waned in English as well, and Old English literature is, like Korean, much more concerned with alliteration.
  And since this Britain was build by this baron great
  Bold boys bred there, in broils delighting
  That did in their day many a deed most dire
  More marvels have happened in this merry land!
  (Gawain and the Green Knight)
  This is, of course, an example of fractal structure: the head-firstedness is reflected at the syllable level, and again at the clause level, and presumably again at the exchange level, so that in English we have:
  "Do you like apples?"
  "Yes (I do)."
  The "head" of the exchange contains the topic and the semantics and even most of the grammar of the exchange. In Korean, the corresponding answer has to be grammatically fuller and stands alone, so there is more semantic and topical weight in the tail of the exchange:
  "Sa-gwa-reul Choh-a-hamnikka?" (Are apples pleasing to you?)
  "Ne (choh-a-yo)!" (Yes, [they please me])
  Now, obviously, the UG model and a grammatical "gene" mutation with a "head-first" "head-last" switch doesn't really explain the DISCOURSE situation very well; it's just doesn't seem likely that INTER-personal behavior is determined by a language faculty in the brain.
  But if we accept that grammar is precipitated out of discourse rather than having discourse percolating up from grammar, this makes perfect sense: interpersonal exchanges are head first, and utterances evolved out of them, so utterances still have the semantically head-heavy marks of their interpersonal origins.
  My student Yongho Kim and I describe a contemporary example of how this happens (when Korean children play) in our article in the current issue of Applied Linguistics:
  This idea of "grammar out of discourse rather than vice-versa" is not mine at all. It's very clearly set out in Tomasello's brilliant "The Cultural Origins of Human Cognition" (see especially page 43 ).
  It's not his either, though! It is of course a simple application of the idea that every higher psychological function occurs first BETWEEN minds (discourse) and only subsequently WITHIN them (grammar).
  So it is quintessential Vygotsky. It is also Volosinov (it's the reason why his example of an application is INDIRECT SPEECH, an almost pure example of how a grammatical structure is derived from a discourse.
  It is also Bakhtin's, and it is one reason why he rejects (in "Problems of Speech Genres") the sentence as a fiction and instead adopts interpersonal utterances as the "real unit" of linguistics.
  I think that Vygotsky's ideas about language from from Volosinov and Bakhtin rather than from (as has been argued on this list) Frederic Paulhan. I just struggled through Paulhan's 1929 volume "The Two Functions of Language", and I find absolutely NOTHING in it about "znachenie" and "smysl" (or "meaning" and "sense")
  Paulhan just divides language into two functions, "signifying" (that is, communicative) and "suggestive", and then discusses the psychological aspect of the one and then the other and the social aspect of the one and then the other. He is quite agnostic about the relationship of the psychological and the social; if anything, he seems to imply that the former is primary and the latter derivative.
  So I need to find out more--in particular:
  a) Is Vygotsky's concept of "znachenie" and "smysl" really derived from Volosinov's distinction between "theme" and "meaning"?
  b) Was he using Paulhan as a cover? Why?
  c) What is the relationship between Vygotsky's ideas of "senses evoked in the mind" and PRAGMATIC (negotiable) meaning? What is the relationship between Vygotsky's idea of "the most stable zone of sense" and SEMANTIC (abstract, dictionary, conceptual) meaning?
  David Kellogg
  Seoul National University of Education

Be a better Globetrotter. Get better travel answers from someone who knows.
Yahoo! Answers - Check it out.
xmca mailing list
Received on Mon May 21 14:42 PDT 2007

This archive was generated by hypermail 2.1.8 : Fri Mar 21 2008 - 16:43:26 PDT