Bakhtin is certainly wonderful, but I have a concern. It is good to point out
that a genre like the novel is polyvocalic, and generally that communicative
forms frequently embody multiplicity and get interpreted in different ways
by different parties. It just seems to me that it's a mistake to stop there.
Is there anything to say about *which* voices come together in a genre?
About how the *relationships* among those voices are embodied in the genre?
About how, if at all, the relationships among the voices reflect the material
relationships among the social groups whose voices they are (or were)?
Likewise, I am wary of stopping with the observation that different groups
might interpret a text differently, or that they might systematically
interpret a whole genre differently. Differently how? And how do those
differences in interpretation reflect differences in background, strategy,
intragroup dialogue, intergroup relations, etc? And how, if at all, do
those material differences feed back into the construction of the genres?
I'm thinking, for example, of Helene Keyssar's study of black theater, in
which the demands of intelligibility to both black and white audiences, and
certain strategies of address in relation to each audience, are exhibited in
the play text. Or, on a less lofty level, of cartoons that were produced as
short subjects for movie theaters and thus had to entertain both children and
adults.
I don't mean to say that Paul rules out this kind of inquiry, but I do think
that many people have been satisfied with an aspect of Bakhtin that I regard
as essentially a negative point (if a valid one), as opposed to a positive
material analysis.
A comment, too, about Chuck Bazerman's analysis of genre as a psychosocial
ascription category. This may be fine for his purposes, but as he points out
it might not be fine for other purposes. I persist in thinking that genre has
some objective reality. That reality is certainly not wholly formal; Chuck is
right that attempts to define formal grammars, a la Chomsky or structuralism,
for generic categories have not worked. So genre is not a formal category.
But other analyses are certainly possible. For example, in my note I sketched
the tension between formal and functional aspects of genre. This is certainly
not good enough either, but it heads in a different direction.
Here's another approach. Think of genre as sort of a tunnel (or, in more of a
mathematical metaphor, an envelope). No author or speaker ever has a complete
formal sense of genre requirements. They have a concrete situation in which
to write/speak; they have an internalized addressee (which might be thought
of as a generalized other in Mead's sense, only not completely generalized
but generalized to a certain category of addressee within certain social
relationships and typified situations); they have an inductive sense of the
models provided by other examples of the genre; they have the learning process
(legitimate peripheral participation, perhaps) by which they learned to
produce such things; they have others' comments on their previous efforts;
and so on. In writing, they are steered by their sense of "what feels right"
in terms of these internalized others. They might well also be steered by
certain mediating symbols like "methods section" and "state the main idea
up front". These forms of feedback tend to ensure that they produce stuff
that bears a certain family relationship to other stuff within the genre.
A genre can display quite a lot of internal diversity as individuals come
up with a variety of ways of responding to the many influences that act upon
them as they write/speak. The "tunnel" or "envelope" is the bounded -- that
is, constrained -- trajectory they pass through as they produce, hopefully,
socially acceptable and efficacious texts.
Phil Agre