At 11:42 AM 5/30/02 -0400, Jay wrote:
>
>I wonder, Keith, if you could share an example, or at least a sketch of how
>both top-down and bottom-up analyses of improvisation in conversation
>together help us understand what is going on?
Here's an example close to the concerns of XMCAers: in a paper titled
"Improvisation and narrative" that is currently under review. In the
"emergent literacy" tradition, children's unstructured pretend play has
been shown to contribute to literacy skills. But an unresolved issue is:
how do you get from the improvisations of play to a structure that can be
defined as a "narrative"? (A variant of the classic structure/practice
tension.) I propose that narratives emerge from play improvisations; a
narrative structure (that can be coded and analyzed by researchers) is the
emergent result of unstructured discursive interaction. To analyze how
this emergent process occurs, I look closely at the turn-by-turn
interactional dynamics in the play group, analyzing transcripts of play
conversation.
Documenting evidence of top-down downward causation is more involved. My
favorite example is in the forthcoming book "Improvised dialogues"; I'll
sketch it briefly here.
I contrast two different improv theater groups that both improvise a
60-minute play, without using any predetermined scripts, characters, or
plots. One group allows their actors to step out of character at any
moment and comment on the ongoing action, proposing new ideas for the
drama; the other group requires their characters to stay in character at
all times.
Using both qualitative and quantitative analyses, I show that (1) these
interactional differences result in the emergence of unintended effects in
the types of narrative structures that are generated. The in-character
group ends up with more complex characters and relationships; the in- and
out-of-character group ends up with more complex plot lines but relatively
simple characters and relationships. I use a theory of the metapragmatics
of discourse to explain why this happens. (2) Once these structures have
emerged, I then show that they have different causal effects on actor's
discourse moves in individual turns, and that these effects increase
throughout the performance, as the dramatic frames become more elaborated.
R. Keith Sawyer
http://www.keithsawyer.com/
Assistant Professor
Department of Education
Washington University
Campus Box 1183
St. Louis, MO 63130
314-935-8724
This archive was generated by hypermail 2b29 : Thu Jun 27 2002 - 08:02:50 PDT