Jay,
"...that SFL consider more dynamical models, i.e. ones in which linguistic
meaning, in the sense of "texts" (spoken or written, singly or interactively
with others), is made in real time"
In my context of language education, many have taken the notion of genre as
being stable text types - constituency representations of the stages of a
text. This often results in frustrations at the classroom level due to an
almost transmission style pedagogy being implemented. Learners' texts are
compared with the model genre and debited according to their structural
differences as well as their differences at the level of register. In terms
of being able to assess and support language development (I am mainly
interested in spoken texts), it seems to me that using genre as systems of
dynamic social processes that are realised through language choices at the
level of field, tenor and mode in real time allows a way in for the
instructor and the learner to focus on how each move (or discourse unit, or
meaning section as you called it) is working at the lexicogrammatical level
to express desired meaning. As you say, "[how] the probabilities of next
meanings-to-be-made
changed as a function of the prior meanings-already-made..." This was my
initial question to the group - can this logogenesis be likened to
microgenesis, which I think I am now beginning to understand (though I'll
need a lot more time to bring it together).
Reading your ideas about the notion of topology of genre has helped me to
reconcile the problem of using SFL in the above way. I have often had a
problem with the way that SFL classifies grammatical units as either-or's (a
problem that Jim Martin discusses when introducing your notion of topology
complementing typology of genres). This has allowed me to view texts as more
fluid and overlapping rather than simply types of social processes (this is
probably not what you intended, but it works for me). It also seems to fit
with Bakhtin's claim of the heterogeneity of speech genres and the lack of
understanding of their linguistic characteristics.
Thanks so much for the references, and apologies to all for turning this
discussion forum into an SFL forum.
Phil
----- Original Message -----
From: Jay Lemke <jllbc@cunyvm.cuny.edu>
To: <xmca@weber.ucsd.edu>
Sent: Thursday, October 03, 2002 8:33 AM
Subject: Re: Logogenesis and Microgenesis
I am gradually emerging from several weeks of "transitioning" to my new
home, office, and position at the U of Michigan. A big change! ... but a
very welcome one.
Like Gordon, I am another person who lives around the intersection of
AT/Vygotskyan perspectives and Halliday's social-functional linguistics.
Notions of multiple timescales are not very well developed in Halliday's
approach, or in SFL (systemic-functional linguistics, the formal name for
it) generally. This is an area I am very interested in at the moment.
Logogenesis is an idea of considerable interest, but not much developed in
the theory. It is an attempt, partly in response to some questions and
suggestions I made to Halliday and others in the 80s, that SFL consider
more dynamical models, i.e. ones in which linguistic meaning, in the sense
of "texts" (spoken or written, singly or interactively with others), is
made in real time. We imagined a process in which after each meaning
selection, or some group of such selections, say enough to determine or
generate a phrase or clause, the probabilities of next meanings-to-be-made
changed as a function of the prior meanings-already-made. Halliday
developed this idea a bit as an extension of his model of system
probabilities (the likelihood of various alternatives being selected, e.g.
present tense vs. any other tense; active voice vs. passive voice, etc.,
either in the language as such, or in some particular specialized register
or subject area). These were conditional probabilities (depending on other
"simultaneous" meaning choices), and he then added the idea of sequential
or dynamic probabilities as I just described.
About the same time I wrote a paper on dynamic textproduction that also
developed the basic notion. The term logogenesis I think arose in later
conversations Halliday had with his then student, Christian Matthiessen,
around computer implementations of the idea. There was some awareness of
the relations of semogenesis, an older idea in SFL (how new meaning
possibilities evolve in the language), and logogenesis (how a text
"evolves" or develops in conversational or compositional time), as on
longer and shorter timescales. Both of these were seen as on shorter
timescales than longterm language evolution (i.e. over centuries or
millennia). Jim Martin I think has written a bit about these relations.
I think the best exposition is in Halliday & Matthiessen, _Construing
Experience through Meaning_, Continuum Press, 1999/2000. Here there is an
explicit contrast of phylogenetic (evolutionary), ontogenetic
(developmental), and logogenetic (textproduction) timescales, but not too
much about how they actually interact, though there is one interesting
section on this.
My original paper is: Text Production and Dynamic Text Semantics." In E.
Ventola, Ed. Functional and Systemic Linguistics: Approaches and Uses. [pp.
23-38]. Berlin: Mouton/deGruyter (Trends in Linguistics: Studies and
Monographs 55). 1991.
JAY.
---------------------------
JAY L. LEMKE
Educational Studies
University of Michigan
610 East University
Ann Arbor, MI 48109 USA
http://www-personal.umich.edu/~jaylemke
---------------------------
This archive was generated by hypermail 2b29 : Fri Nov 01 2002 - 01:00:06 PST