I have no doubt about the materiality of symbolic artifacts and agree
with those who have pointed out that, in order to perceive/utilize them,
one needs a corporeal body, with one or more functioning sense organs and
other physical apparatus for producing them.
The question at issue seems to me to be not whether speech is material
but whether it can be "captured" for further interrogation, etc. With
electronic recording devices it is possible to capture many features of
speech (but by no means all) and so to produce a version that can be
revisited. Prior to the invention of these devices, written language
allowed "what was said" to be captured, although (as I wrote in a previous
message, citing Olson's work) it is doubtful whether that was the purpose
for which writing was invented. Although in alphabetic scripts writing
appears to capture "the very sounds" of speech, a written text is a
rather diferent kind of artifact from an audio or video recording. For
the latter has to be processed in real time at a speed equivalent to that
at which the speech itself was produced, whereas a written text, because
it is inscribed in a visual medium, can be dwelt on, reviewed in any
order, etc. It is also produced by means of different processes, which
allow the author to review and, if necessary radically revise, the text
before "handing" it to a reader.
Nevertheless, both spoken and written texts (or 'utterances', as Bakhtin
calls them) are linguistic artifacts and are similar in that they are
systematically related in functional ways to the culturally constructed
situation types in which they play their roles as mediational means -
that is to say, they are constructed against a "backdrop" of conventional
patterns of expected cooccurrence (intertextuality) that linguists describe
in terms of 'register' and 'genre'.
In his recent message, Peter wrote:
> Now, speech genres, while I
> think theoretically compelling, are very difficult to document. A number of
> us have conducted research looking at patterns of discourse in various kinds
> of discussions in school, yet really *establishing* the presence of a speech
> genre is, I think, very difficult. And, from the perspective of a speech
> genre being a cultural mediator, they remain intangible to those who
> participate in them--that is, researchers can tape record discussions and
> conduct some sort of discourse analysis on them, yet by the time the work's
> done and the data analyzed and the conclusions drawn, the classroom under
> study has long been graduated--in other words, while the patterns of
> discourse have become material to the researcher, they are ephemeral to the
> participants. Yet I agree with Jim that they exist, that they are cultural
> in origin, and that they work semiotically as mediators. They become
> cultural tools, although tools of which the users are often unaware--people
> socialized to them tend to regard them as the natural way that people
> communicate in general.
The difficulty in 'establishing' the occurrence of speech genres does not
seem surprising, given the disagreements among academics about the status
of written genres. If it's problematic for written texts, which are
permanent objects that are much more consciously and deliberately
constructed by writers and readers, then we should not be surprised
that conversational participants are not consciously aware of
"speaking a genre" as they co-construct the text in real time. But that
does not mean that situation/activity-related expectations do not shape and
constrain what is said in the same sort of way as in writing.
In thinking about this, I find the activity-theoretical distinction
between action and operation helpful. Speech/conversation is rarely if
ever the action in which participants are consciously engaged; instead
they are collaboratively trying to do something else (move a heavy object,
win at cards, make a new friend, help somebody to do or understand
something) and the speech is (one of) the operational (mediational) means
that they deploy to achieve their goals. But the point about operations
is that they are deployed more or less automatically - they are
well-practised and do not require conscious attention. Speech genres are
like tools in a toolkit: one selects from those available the most
appropriate for the task in hand and uses it with one's attention on what
one is trying to do with it. Generically, one's behavior (operation)
is functionally appropriate for the task; at the same time, it is
uniquely and personally adapted to the specific conditions in which it is
produced.
I think the same is also true of writing to a considerable extent. When
writing, I don't consciously think all the time (or at all) about the
genre I am using; but if the text I produce is appropriate to my purpose
in the situation, it will - post-hoc - be seen to display many of the
features of the genre that is functionally/conventionally appropriate.
I think the same applies - only more so - in the case of speech.
The other point about operations, as Leont'ev describes them, is that
they start as 'actions' and become operations through repeated use; but,
when a difficulty occurs, or their smooth deployment is impossible in the
situation, they can be stepped up to the action level again and once more
become the focus of attention.
I think there are valuable implications - for analysis and for education -
in using activity theory as a tool for thinking with in this context. But
I don't have time to develop them now.
Incidentally, Peter, how about theories as material artifacts?! :>
Gordon Wells, gwells who-is-at oise.on.ca
OISE, Toronto.