Re: [xmca] Antirecapitulationism and the Logical Impossiblity of Social Progress

From: David Kellogg <vaughndogblack who-is-at>
Date: Mon Apr 07 2008 - 16:10:51 PDT

Ed and Martin:
  I read Ed's link with a frisson of horror. I'd read BOTH articles in Applied Linguistics (the anniversary article that Professor Widdowson wrote and de Beaugrande's captious reply) but I'd never imagined how snotty, snide, snarky and completely off the point de Beaugrande "unbowdlerized" original was--it reminded me of....well, myself, at my hysterical polemical worst. Hence the horror.
  But if you read de Beaugrande very carefully, ignoring the gratuitous insults, appeals to "scholarly authority" and irrelevancies (e.g. the mini-corpus schtick), you can glean the following points:
  a) Halliday really DOES say that meanings get accepted from different metafunctional inputs and are spliced together. This is the opposite of a Volosinovian understanding, whereby meanings are created through communicative interaction and only then can they be prized apart for analytical purposes (and resynthesized as new meaning potentials).
  b) Halliday really DOES say that a grammar of the system can be simultaneously a grammar of the text. This is the opposite of a developmental understanding, which has to explain how the system arises from the texts, and how texts arise from discourses. Of course, not all understandings have to be developmental, but all Vygotskyan ones do.
  c) Halliday and Sinclair assume that the appropriate can be derived from the attested (that is, everything that can be said is actually said somewhere) and de Beaugrande goes even further: "I cannot imagine a linguistic usage being appropriate if it had NEVER been attested". That means that de Beaugrande cannot imagine linguistic creativity.
  d) Halliday does assume that an analysis of the text is an analysis of discourse, and de Beaugrande justifies this by saying: "I have never seen a convincing instance of theoretical 'text' isolated from all discourse."
  Nobody is saying that a text can be isolated from all discourse; if a text is the recorded trace of a discourse than of course at some point it was not isolated from that discourse. But what happens next is a very different matter. As you read these words, you are isolated from a very large number of discourse facts, and some of these are undoubtedly relevant to the text you are reading (the expression on my face, the state of my health, the amount of Widdowson I've read versus the amount of Halliday, the fact that Widdowson was the external examiner for my MA, etc.).
  I think d) is the key problem for us. It's the extent to which communication can be seen as an ongoing, never-quite-completed, simultaneously linguistic and nonlinguistic PROCESS (discourse) or as a finished, finalized, recorded PRODUCT (text)? In other words: Is it the case that people communicate using texts, or do texts somehow communicate through the mouths of people? That's the main problem Widdowson addresses (and Volosinov before him).
  And that's exactly the problem that de Beaugrande REFUSES to address. Look at his title: "Interpreting the DISCOURSE of Widdowson". All he has done is to interpret Widdowson's TEXT. Needless to say, using the text-centred and discourse-free techniques that Widdowson criticizes to criticize Widdowson seems a clever bit of reflexiveness, but it leaves the argument untouched; if Widdowson's criticisms of systemic functional grammar, corpus linguistics and critical discourse analysis are correct (and I think they are) then de Beaugrande's criticism of Widdowson is ipso facto invalid (and I think it is).
  Widdowson's arguments are based on what Andy would call IMMANENT critique; that is, he takes Halliday at his word and goes where the analysis appears to follow. Halliday does indeed assume that the clause is the powerhouse of the language, when, as Tomasello points out (and Volosinov before him) very little of what we say is actually fully formed clauses. Sinclair really does insist that CORPORA, accessed through a CONCORDANCE PROGRAM provide enough context to understand his texts as discourse, but concordance programs only provide (usually) seven words to the right or left of the given input word (and as Hoey points out (and Bakhtin before him) the reach of a speech genre is far greader than fourteen words. Fairclough really does claim that a textual analysis yields an ideological critique in a quite straightforward way, and in fact de Beaugrande himself proves it, by letting his own article degenerate quite straightforwardly from a textual analysis of Widdowson to a
 string of (mostly teflon) epithets.
  So in the end, de Beaugrande simply appeals to authority: Widdowson goes against the Applied LInguistics editorial and goes against Halliday, who has spent a lifetime studying language (according to de Beaugrande, Widdowson's career was a 25 year hiatus). This is not exactly immanent critique; I wonder if it qualifies as a critique at all.
  Alas, on the internet, we often get exactly what we pay for. But Widdowson is now retired, and he has somehow prevailed on OUP to make a lot of the books he wrote during his 25 year hiatus available for nothing. Have a look:
  David Kellogg
  Seoul National University of Education

You rock. That's why Blockbuster's offering you one month of Blockbuster Total Access, No Cost.
xmca mailing list
Received on Mon Apr 7 16:12 PDT 2008

This archive was generated by hypermail 2.1.8 : Thu May 01 2008 - 17:14:13 PDT