#modeling composition

[compmachines]

"Cognition, Convention, and Certainty"

Bizzell starts with an acknowledgement that we all know now that thinking needs to be taught, or at least controlled for, in addition to style and grammar. But there are two camps regarding what that thinking looks like:

- Inner directed (cognition is primarily within the individual)

- Outer directed (social epistemic/discourse community stuff)

"Inner-directed theorists further claim, in a similar paradox, that the universal, fundamental structures of thought and language can be taught... Once we find these models [for modeling appropriate cognitive structures], we can guide students through the process until the students' own thought-forming mechanisms 'kick on' and they can make concepts on their own" (390).

"In contrast, outer-directed theorists believe that universal, fundamental structures can't be taught; thinking and language use can never occur free of a social context that conditions them... What we should do is to teach students that there are such things as discourse conventions" (390).

It's not that students can't learn, "It's just that they can't think or use language in the ways we want them to" (392). Very Bartholomae-ish/Bruffeean. In other words: very social epistemic.

The social epistemic theorists--like Bizzell--lay emphasis on exercises that interrogate the discourse communities in order to learn about the discourse conventions.

But one or the other of these isn't wrong--they can be a "kind of fruitful exchange that enlarges knowledge" (392). So Bizzell is going to demonstrate that by combining one of each model into a new, fuller-focused model. She's going to combine Flower and Hayes' cognitive model with sociolinguistics.

She summarizes Flower and Hayes' model, but also makes some criticisms. The Monitor, for instance, is another way of saying "the writer making choices," and that places a massive emphasis on how something is decided--without actually revealing that process. It's a black box, in other words (my term) (395). She also points out that "translation" is hypersimplified--just pouring thoughts into English containers. She also points out that they separate language out of planning--language can help record it, but it's not super necessary. Bizzell counters that language "doesn't exactly teach us to think, [but it] teaches us what thoughts matter" (395). Also, where do the new goals come from? In the model, they just kinda appear. Bizzell thinks that strange.

She uses sociolinguistics to fill in that weirdly blank "translation" space. Everything depends on the discourse context, or, "educational problems associated with language use should be understood as difficulties with joining an unfamiliar discourse community" (397). Bizzell suggests what Bartholomae eventually will--that we need to explain to students more clearly what our terms are. What does it mean to be specific?

Toward the end, Bizzell notes that it's a serious problem with inner-directed models that they're so easy to develop into heuristics and textbooks, foreclosing on further thought. I can see where that would be a problem. Instead, she advocates for discourse analysis.

[compmachines]

"A Cognitive Process Theory of Writing"

Their question: “What guides the decisions writers make as they write?” (273)

They conceive of writing as a series of hierarchical, embedded processes, guided by a network of goals and sub-goals generated by the writer during the writing process. (274-75)

The stage model is insufficient, as Nancy Sommers and others have demonstrated (275).

Their model is a “hypothesis about a dynamic system” that can never completely capture something “which refuses to sit still for a portrait” (276).

Their study is built on “protocol analysis,” basically an analysis of a writer thinking aloud while writing and thinking.

Elements of their model:

The task environment

The writer’s long-term memory

The writing processes

That last one includes:

Planning

Translating

Reviewing

All under the control of a Monitor.

These processes run essentially as on-command subroutines embedded within one another (284) People create networks of goals (process or content goals) as they write--not just in prewriting.

[compmachines]

"The Basic Aims of Discourse"

"By aim of discourse is meant the effect that the discourse is oriented to achieve in the average listener or reader for whom it is intended.... Is the work intended to delight or to persuade or to inform or to demonstrate the logical proof of a position? These would be typical aims" (130).

Kinneavy's trying to chart a course between the Scylla of intentional fallacy and the Charybdis of the affective fallacy in talking about the author's aim of discourse.

Kinneavy walks us through historical periods of rhetorical/dialectical discussion and figures out the principles of division, and the divisions themselves, that these theorists used. So, Aristotle and Aquinas divided their categories by level of probability, and came up with: 

Scientific (certain)

Dialectical (probable)

Rhetorical (seemingly probable)

Poetic (internally probable)

Other theorists had other classifications, and you can find the chart on 132.

Discourse seems to be dominated by any of the four parts of the rhetorical triangle: Encoder, Decoder, Signal, and Reality (it's interesting that Kinneavy places Signal at the center and Reality at the apex).

I like Kinneavy's metaphor for language: it's like a window. You can throw a brick through it to express yourself, use a chunk of it to chase someone (persuasion), use it as a mirror to reflect things (referentially), or use stained glass to make a thing of beauty (aesthetically). So language can be expressive, persuasive, referential, or aesthetic, and those coordinate to the four parts of the rhetorical triangle.

"each aim of discourse has its own logic, its own kind of references, its own communication framework, its own patterns of organization, and its own stylistic norms" (138), so we can't ignore any of them.

[compmachines]

"William Perry and Liberal Education"

1984: Patricia Bizzell thinks that William Perry (60s Harvard developmental psychologist) was a pretty cool guy.  He thought that a college liberal education should lead students through three important stages: Dualism, Relativism, and end in Commitment in Relativism.  But slow down, you overenthusiastic composition person you, says Bizzell, what Perry considers desirable and what we want to teach our students aren't necessarily the same thing.  It's enough for her article that Perry raises interesting points.

High Point: "Perry provides us with a useful picture of the kind of 'cultural literacy' required in a liberal arts college" (326).

[compmachines]

"Audience Addressed/Audience Invoked"

1984: Lisa Ede and Andrea Lunsford helpfully complicate Walter Ong's take on audience by not only providing the two titular types of interaction with audience, but by highlighting the changes in audience brought on by genre (specifically, in composition versus fiction).  The audience has more ways of coloring the text than simply reimagining themselves, and Ede and Lunsford do their best to sketch out some of the complexity inherent in the notion of "audience."  Plus, they have diagrams!

High Point: "But Ong fails adequately to recognize the constraints placed on the writer, in certain situations, by the audience.  He fails, in other words, to acknowledge that readers' own experiences, expectations, and beliefs do play a central role in the reading of a text, and that the writer who does not consider the needs and interests of his audience risks losing that audience" (88).

[compmachines]

"Contemporary Composition: The Major Pedagogical Theories"

1982: James Berlin is biased, and he knows it.  There are four pedagogical theories he says--Neo-Aristotelians, Current Traditionalists, Neo-Platonists, and New Rhetoricians--but the last is best.  His definitions may be reductive, and he may be a little unclear on Aristotelian rhetoric, but it's a useful and clearly written starting place.

High Point: "To teach writing is to argue for a version of reality, and the best way of knowing and communicating it" (256).

[compmachines]

"Coherence, Cohesion, and Writing Quality"

1981: Stephen P. Witte and Lester Faigley take a very, very close look at the notion of "coherence" in student writing (expressed primarily through connections between student sentences).  In essence, they find that coherent student writing is better student writing.  Those unfamilair with T-units had better start elsewhere.

High Point: "To some extent the types and frequencies of cohesive ties seem to reflect the invention skills of student writers and to influence the stylistic and organizational properties of the texts they write" (250).

[compmachines]

"A Cognitive Process Theory of Writing"

1981: Linda Flower and John Hayes wonder what guides a writer's process: "'purposes,' 'relationships,' 'exigencies,' 'language'" (274)?  They propose a theoretically interesting (but not immediately pedagogically applicable) compositional model, based on the computer system.  The model is nonlinear, and focused on the writer's goals.  Other authors (particularly Patricia Bizzell) point out flaws in the model, but it's useful as just that: a model.

High Point: "A cognitive process theory of writing, such as the one presented here, represents a major departure from the traditional paradigm of stages" (276).