Re: [xmca] Instantiation Cline, or, Firewalls and Fires

From: David H Kirshner (
Date: Wed Aug 30 2006 - 02:36:15 PDT


I followed you up until the last paragraph:
"It seems to me that ALL languages, in order to remain living languages,
MUST evolve texts and even whole registers which can be understood without
understanding the language system: that is what child language, and
pidgins, and teaching languages are."

Somehow, you seem to be distinguishing dead languages from living languages
from nonsystematic registers, as if a language had to be currently
spoken/evolving to have living texts, and as if child language, pidgins,
and teaching languages aren't also systematic. (I know I must be missing
your intent, here.) I wonder if Lotman's distinction between monologic and
dialogic poles of language use might not be closer to what you're after in
trying to link meaning and interpretation.

David Kirshner
Louisiana State University

PS. I'll be in Seoul to talk at a couple of universities in early October
(right after Chusok). Perhaps we could meet for a meal (or more
realistically, given the challenges of transportation, have a phone
conversation) while I'm there.

Lotman, Y. M. (1988). Text within a text. Soviet Psychology, 26, 32-51.
Lotman, Y. M. (1988). The semiotics of culture and the concept of text.
      Soviet Psychology, 26, 52-58.

                      <kellogg who-is-at To: "eXtended Mind, Culture, Activity"
                      kr> <>
                      Sent by: cc: (bcc: David H Kirshner/dkirsh/LSU)
                      xmca-bounces who-is-at webe Subject: [xmca] Instantiation Cline, or, Firewalls and
                      08/30/2006 03:07
                      Please respond to
                      "eXtended Mind,

First of all, I must apologize for a bit of muddleheadedness. Previously, I
said that I thought that context of situation (which I associate with
PRAGMATICS) and context of culture (which I associate with SEMANTICS) are
not commensurable, because one is in some sense causative of the other.

This is a very silly way to put it indeed; it sounds like I think that
causes and effects are never commensurable. Obviously, when vocal cord
vibrations cause sound waves, the two can both be measured in Hz.

But the situation IS different when we talk about how pragmatic phenomena
like use causes (perhaps "precipitates" is a better word) semantic
phenomena like usage. I'm not sure the second category of phenomena are
really measurable at all. Words, as contextualized, pragmatic events are
measurable; I spend a good deal of my day counting them. But abstract word

When I talked to Halliday, I remember we had a big argument about whether
"meaning" was a type of "information" or "information" a type of "meaning".
At the time I was very thoroughly over-awed and I didn't really understand
what I was saying, much less what he was saying. But now when I think back
on it, I think I understand it perfectly.

Halliday thinks that information is a type of meaning, viz. it is the type
of meaning which is measurable in bytes. But I think that meaning is a kind
of information, specifically, it is the type of information that has been
represented in some way in human consciousness. Sometimes the way it is
represented is susceptible of measurement, but sometimes it is not (I think
this is not unrelated to the extent to which information is

I mention this at the outset because I really feel that what is at stake
here is precisely whether or not we are going to allow Cartesian dualism
back into our discussions by saying that "ideal" phenomena are distinct
from, and in no way linked to, material ones.

Yes, of course, we may build a little firewall by saying that instantiation
is only true of the descriptions we create of language, not true of
language itself. That is what I did, when I said "everything else is
description", and it's also what bb does when he writes:

"I interpret the features in the cline of instantiation figure to be
relationships between theoretical categories. Being purely theoretical,
everything is ideal. Even text is ideal. It is not the actual text created
in the situation -- for example, it is not the discourse as it unfolds, but
it is the recorded, transcribed, analyzed, theorized text. In his verbal
analysis paper, (a copy is on his web) Jay Lemke has written that as we
record, transcribe, and examine a discussion it is transposed -- it becomes
a different thing than the discussion itself. The latter is the use of
language for communication, contextualized with the situation, but as we
analyze it, it enters into a different activity, research, in which it
becomes part of a different meaning making activity, contextualized
differently with theory and other interpreted data."

But this transposition is not a transfiguration; it is not an apotheosis.
Volosinov's whole reason choosing reported speech as the final, "ascent to
the concrete" portion of Marxism and the Philosophy of Language is that
taking a text and quoting it just creates another kind of text, which may
be grammatically subordinated (or not, as the case may be) but which still
exists on the same plane, as text. Bakhtin says "The text (written and
oral) is the primary given of all (linguistic, philological, literary, and
other) disciplines and of all thought in the human sciences and philosophy
in general (including theolotical and philosophical thought at their
sources. The text is the unmediated (!) reality (reality of thought and
experience, the only one from which these disciplines and this thought can
emerge." (Bakhtin, "The problem of the text", Speech Genres and Other Late
Essays, 1986: 103). Makes it VERY clear which he thinks is primary and
which is derivative!

Sometimes the firewall is more dangerous than the fire. My feeling is that
we may say that pragmatics and semantics are distinct, but we cannot say
that they are in no way linked. When Valsiner and van der Veer talk (2000:
397) about how it is quite impossible to tell that inter-mental processes
are transformed into intra-mental ones, and therefore the whole ZPD
construct is a bit of speculative moonshine, my blood runs cold. I hear the
clanking of Descartes' automatons.

So if we say that there is something like an abstract text (or an abstract
word meaning) which does not form part of some superordinate social whole
(that is, for me, pragmatics) we are either, a) reintroducing Cartesian
dualism, or b) declaring the independence from and equivalence of
theoretical descriptions and pragmatic facts. Or both. I can't do that.

But I think Halliday can. Here's what Halliday says, on the very first page
of the Introduction to Functional Grammar:

"To a grammarian, text is a rich, many faceted phenomenon that 'means' in
many different ways. It can be explored from many different points of view.
But we can distinguish two main angles of vision: one, focus on the text as
an object in its own right; two, focus on the text as an instrument for
finding out about something else. Focusing on text as an object, a
grammarian will be asking questions such as "Why does the text mean what it
does (to me, or to anyone else)? Focusing on text as instrument, the
grammarian will be asking what the text reveals about the system of the
language in which it is spoken or written. These two perspectives are
clearly complementary; we cannot explain why a text means what it does,
with all the various readings and values that may be given to it, except by
relating it to the linguistic system as a whole; and equally, we cannot use
it as a window on the system unless we understand what it means and why.
But the text has a different status in each case: either viewed as
artefact, or else as specimen."

I completely agree that there are MANY different points of view possible
(I've always felt unhappy at the limitation of the metafunctions to only
three). But that's about it. As soon as we selected two main angles of
vision, I got lost.

First of all, I don't know at all what it means to focus on a text "as an
object in its own right", except that it sounds like we are focusing on it
out of context, and in that case I don't think we can answer the first set
of questions about why a text means something. Texts mean things in
contexts; a text abstracted from a context is not a text at all, but as you
said, a kind of idealized description, a textoid, whose relationship to an
actual text is very doubtful to me.

Secondly, I can see how it is impossible to understand a linguistic system
without understanding at least one text written in that system; Halliday's
grammar wouldn't make much sense to me if I'd never read any texts in
English. But I really DON'T think the opposite is true; it seems to me it's
quite possible to understand texts without understanding the linguistic
system, else language would not be learnable, by children or by anybody

It seems to me that ALL languages, in order to remain living languages,
MUST evolve texts and even whole registers which can be understood without
understanding the language system: that is what child language, and
pidgins, and teaching languages are.

David Kellogg

Seoul National University of Education

xmca mailing list

xmca mailing list

This archive was generated by hypermail 2b29 : Tue Sep 05 2006 - 08:14:31 PDT