Using the cognitive model to describe writing
The cognitivist theory is something I’ve heard referenced quite a bit through the classes I’ve taken so far. But I had yet to hear the theory from its source, Flower and Hayes. On an emotional level, I’ve been “suspicious” of the cognitive theory. To me, describing writing as a series of discrete cognitive processes felt reductive. And criticism from social constructionists seemed valid: what about the people reading it? what about the social environment in which we compose? and, what about the effects of our writing? Still, particularly in light of last week’s readings, some of the goals of Flower and Hayes seem admirable.
In particular, the value of the work Hayes and Flower accomplished took shape when they discuss the implications of their model, which they call “first and foremost a tool for researchers to think with. By giving a testable shape and definition to our observations, we have tried to pose new questions to be answered” (375). Charney, for instance, seems to believe this testability is important to formalize the study of writing. In particular, she believes that composition theorists need to be able to compare their work and test one another’s theories. This can be difficult if everybody uses a subjective, idiosyncratic methodology. Echoing the worries of a statistician, Pearson, Charney goes so far as to suggest this might implicate composition theorists in falling for unexamined acceptance of an elite few (284). And, we see Charney arguing for objective methods to facilitate the communication and openness that would combat relying on the theories of an elite few (288). She lauds scientific communities for moving toward consensus rather than making fact claims “by classifying and prioritizing the rival claims that must be explored, by drawing on prior knowledge, by establishing criteria for consistency, and by systematically addressing the sources of indeterminancy” (287).
Personally, having a method for composition theorists to compare notes, so to speak, sounds great. And the cognitive model Flower and Hayes argue for seems to be one way of doing so. But, as I mentioned, I still have misgivings about the model. First, though appealing, the “model” Flower and Hayes create is almost certainly lacking. Hart-Davidson, Spinuzzi, and Zachry hint at some shortcomings when they note that theorists have modified the model to accommodate other elements of the writing process (71). But that alone shouldn’t drive me to abandon the model. After all, being able to tweak the model to better represent the writing process seems in line with both Flower’s and Hayes’s goals for the model and with Charney’s ideals for being able to compare and challenge the research of others. But for me, it’s a deeper issue. The model seems to imply we can define different processes, and, through the definition, know them. This smacks of positivism. And I can’t imagine any model truly capturing the work of every person. The question then, is, What is the model useful for? And what are its limitations?
The work of the model capturing the writing process seems more suspect when Flower and Hayes begin to describe the model in action. They argue that “[t]he processes of writing are hierarchically organized, with components embedded within other components” (375). This sounds fine, but I wonder how true it is. After all, while writing, I’m certainly not thinking about hierarchical components. And, while Flower and Hayes derived this idea from the “thoughts” of writers as they composed, I doubt its full accuracy. After all the “thoughts” (yes, with quotes) are really the way the authors were verbalizing their thought process. In this way, the authors themselves were likely trying to organize their thoughts in a way that would be easier to understand verbally. That’s a bit of a let down for a “cognitive” model, because the verbalization of thoughts seems less connected to the writing process and more connected to composition generally.
Finally, as Flower and Hayes move into the details of “goal-directed” writing, their analysis seems to be drifting toward a dated view of writing: learn certain tasks, and you’re good. This is something that reminds me of learning “forms” to be fine, or even of “pre-writing, writing and re-writing,” which Flower and Hayes explicitly reject (367). This takes shape for me in their fourth section, where Flower and Hayes talk about generating and regenerating goals. They break goal generation into three typical patterns: explore and consolidate, state and develop, and write and regenerate (382-386). When mapped onto their cognitive model and fleshed out in their examples, this discussion can be rich in adding to our understanding of writing. But, giving these three distinct “tricks,” while perhaps departing from standard models of writing, does not seem to depart from standard teaching or writing nor does it seem to avoid reductive methods. Flower and Hayes seem to suggest that training or learning these “tricks” will make a beginning writer into a more advanced writer. And training specific tasks in this way seems old hat.
I have not dwelled on the social aspect of writing or on how this cognitive model ignores this aspect, though I certainly think that’s a fair critique. The importance of a social aspect in understanding writing seems well established today. And, I think one could create a model that incorporates this more fully. Instead of worrying about including a social aspect, I worry about using models generally. And this is where my frustration lies in studying writing. On the one hand, I believe we need models. That gives researchers a way to compare notes and further our understanding of writing. But, I think any model needs to be addressed, first and foremost, as what it is: a model. I don’t believe writing researchers will ever be able to truly model the writing process. (And I’m a little afraid of the implications that would follow if we could do so.) Therefore, as we use models to test hypotheses or analyze writers, perhaps we should foreground the ways in which the model is incomplete. And perhaps we should analyze how the model will affect the teaching and practice of writing, for any model creates a framework for how we understand writing and how writing is translated into the classroom or workplace. That is, rather than striving for a perfectly “correct” model or other description of writing, we should use that very model to inform our own goals about how writing should be taught and practiced.
Works cited:
Charney, Davida. “Empiricism is Not a Four-letter Word.” Central Works in Technical Communication. Ed. Johndan Johnson-Eilola and Stuart A. Selber. New York: Oxford University Press, 2004. pp. 282-297. Print.
Hart-Davidson, William, Clay Spinuzzi, and Mark Zachry. “Visualizing Writing Activity as Knowledge Work: Challenges and Opportunities.” (Wasn’t sure where this article came from.)
Flower, Linda and John R. Hayes. “A Cognitive Process Theory of Writing.” College Composition and Communication. 32.2 (1981). 365-387. PDF.