Ushering "Audience" Out:
From Oration to Conversation

[As published in Textual Studies in Canada, No. 1 (1992): 45-64].

Writing implies a reader. The act seems straightforward and unremarkable: the writer writes, the reader reads. However, the idea of an "audience" -- and the writer's relationship to that entity -- has been an important and difficult concept in composition and rhetoric since Plato, at least.¹ The problem is fundamental: where and what is the "audience," and how does one come to "know" it?

The current social constructionist perspective on writing, with its attendant focus on discourse communities, has further complicated the writer-reader relationship.² How does "audience" fit with "community"? Are the terms complementary? Synonymous? Contradictory? The idea of "audience," with its roots deep in the rhetorical tradition, implies a group at some distance from an author who controls the discourse and manipulates the readers. The idea of "community," with its postmodern and poststructural connotations, reduces the importance of the author and locates power (and the writer) within the group. As Petraglia puts it, "constructionists focus on the ways in which the audience (that is, the community) shapes the discourse of its members" (40).

In this article, I propose a transformation of "audience" into "community," and I argue that this shift in conception permits a view of discourse as conversation rather than oration, as interaction rather than transmission. As an example of the conversational nature of community discourse, I present a brief profile of social workers writing reports for the juvenile court in Quebec. Finally, I suggest some of the pedagogical implications which I believe follow once the "audience" has been ushered out.

"All those folks out there in chairs"

Douglas Park explains that "the basic image from which the concept of audience derives is that of a speaker addressing a group of people in some fairly well defined political, legal, or ceremonial situation" (249). In other words, "audience" is literally correct in reference to those present during formal speeches; however, when applied to written discourse, "audience" becomes a metaphor.³ Somewhere in the transfer of rhetorical concepts from the classical to the modern tradition, the audience changed from fact to figure of speech, but much composition theory and practice seems unaware of that altered reality. Walter Ong: "when orality was in the ascendancy, rhetoric was oral-focused; as orality yielded to writing, the focus of rhetoric was slowly shifted, unreflectively for the most part, and without notice" (9).

Like the central metaphors of every discipline, "audience" carries considerable meaning. Park says it "might almost be said to mean too much, to block thought by making us think we know what we are talking about when we often do not" (248). Elbow identifies five meanings of "audience"; Spilka and others write of "multiple audiences" in the workplace. Ede and Lunsford suggest a limitless number of meanings on a continuum from actual ("addressed") to imagined ("invoked") audiences, and Selzer offers farther shades of meaning by extending Ede and Lunsford's discussion and setting "audience" in the context of community. Such multiple meanings are a great improvement over the vague "general reader" identified or implied in much writing research and pedagogy. However, continued use of the term "audience" sets the writer apart from readers and leads to a restricted, literal interpretation of the metaphor: "all those folks out there in chairs" (Park 249).

With the reader as audience "out there," the writer becomes orator or actor. In either case, the text operates as performance rather than interaction. The orator/writer declaims, attempts to win the case, to sway opinion, to fulfill the forms of public ritual; the relationship is one to many, but many as collective. The audience here becomes an aggregate -- the opposition, the jury, the voters, the congregation, the mob -- the motivation becomes manipulative: the orator seeks to move the crowd. Instances of discourse are completed statements, not moments in a dialogue.

The actor/writer image has a certain romantic richness and may complement the notion of "persona" in literary discourse, but between actor and audience is the proscenium. arch and the imaginary "fourth wall" of the theatre. Actors on stage stare into the bright front-of-house lights and see only the dark, indistinct shape of a faceless crowd. It is true that good actors sense the audience, adjusting their performance by heeding the subtle cues of sound and silence, but the audience only responds, never initiates, and the relationship begins with curtain up and ends with curtain down.

This image of readers as "folks out there in chairs" leads to serious misconceptions about the writer-reader relationship:

• The writer-reader relationship is largely one way: the writer acts on passive readers.
• It is monologic: the writer speaks, the readers listen.
• It is singular (one-to-one): multiple readers of the same text form a single, monolithic entity called "the audience."
• It is temporary: the writer's relationship to readers begins and ends with the text; the relationship is embodied (given form) or created by the text.

there are enduring intellectual and cultural traditions that support a Platonic view of invention: traditions in literary studies emphasizing the individual unit as a focus for study; romantic notions of the isolated creator, inspired from within; and a strong regard for individualism in capitalistic, patriarchal societies. The Platonic view fits smoothly as well with other prevalent . . . traditions in Western thought: the Cartesian ego, the cognitive 'mind,' and the psychoanalytic concentration on the psyche of the individual. (22)

Writing textbooks and research influenced by this Platonic view usually describe the audience as an aggregate, knowable as a pre-writing variable, and static during the composing process. Subjects and students are asked to write for generic readerships, vaguely defined by demographic or psychological characteristics: age, education, lifestyle, religion, knowledge, attitude, opinion. Knowledge of the audience is assumed to exist in the black box of the mind, similar to, but discrete from, knowledge about topic, genre, rhetorical goals, and other composing variables. The writer's audience is seen as a mental construct, an internal or cognitive representation of an external reality.

Out of this misconception (both of audience and mind) comes cause for real concern: research does lead to generalizations about writing, and textbooks do set classroom agendas. However, as composition has expanded its focus outward from the individual writer to the contexts for composing, a new understanding of the writer-reader relationship has begun to develop.

From Audience to Community

Throughout the 70s and into the 80s, writing theory and practice was dominated almost exclusively by a cognitive perspective. More recently, however, many different disciplines have begun to influence the way we think about discourse. Cognitive theories of writing have been put into context: writing has come out of the mind and into the world -- or, more accurately, writing and the mind have come into the world. Feminists, anthropologists, sociologists, literary critics, philosophers, and many others have demonstrated the extent to which writing and reading are social acts.

Simultaneous with that discovery (or rediscovery), researchers have come out of the lab and into the writing world. Not surprisingly, studies of writers and readers in their natural habitats have shown the complexity and importance of the environment, of the interaction and interdependence found in rhetorical contexts. Writers, even those who seem to work alone, are clearly linked to other writers and texts. Rather than standing apart, addressing the crowd, the writer can only exist within the group. Moreover, readers are often not part of any collective that could accurately be called an "audience"; instead, they are individual members of complex social structures created by the variety of explicit and implicit roles, relationships, interactions, and levels of power within a given community. In their explicit manifestations, these social structures may look something like a distribution list, an organizational chart, the sequence of readers in publication procedures, a membership list, or the people cited in an article. They are not fixed and immutable structures, they live and change, animated by the relationships, the alliances and oppositions, of the writers and readers within them.

This dynamic interdependence allows us to speak of an "ecology" of writing. In Marilyn Cooper's words: "language and texts are not simply the means by which individuals communicate information, but are essentially social activities, dependent on social structures and processes not only in their interpretive but also in their constructive phases" (366). Central to that ecology is the intricate network of social relationships for which texts serve as foci. When considered as 'a social act, writing can be seen to happen between and among people in ways not captured by the audience metaphor: "The social perspective . . . moves beyond the traditional rhetorical concern for audience, forcing researchers to consider issues such as social roles, group purposes, communal organization, ideology, and finally theories of culture" (Faigley 235-236). Not just researchers, but theorists and teachers too, must consider the place of writing in the life of communities. A social perspective on writing creates the need for a new metaphor, one which can suggest the rich social dynamics that surround and support texts, one which can account for the reciprocal relationships among writers and readers described by recent research. That metaphor has been provided by the image of writing as part of the human conversation. As Clark puts it, "The conversational model . . . makes us aware that when we write or read we join an ongoing social exchange" (48). In an attempt to capture the social nature of writing, a number of theorists have evoked the image of face-to-face talk in the form of discussion, dialogue, or conversation (e.g., Bakhtin, Bruffee, Burke, Oakeshott, Rorty). Now in widespread use, the conversation metaphor has allowed us to think imaginatively and creatively of the dialogic nature of discourse communities. For Burke, the conversation is the whole universe of discourse:

Imagine you enter a parlor. You come late. When you arrive, others have long preceded you, and they are engaged in a heated discussion, a discussion too heated for them to pause and tell you exactly what it is about. In fact, the discussion had already begun long before any of them had got there, so that no one present is qualified to retrace for you all the steps that had gone before. You listen for a while, until you decide that you have caught the tenor of the argument; then you put in your oar. Someone answers; you answer him; another comes to your defense; another aligns himself against you . . . . However, the discussion is interminable. The hour grows late, you must depart. And you do depart, with the discussion still vigorously in progress. (110-11)

When writing is conceived as social action within communities, we may also begin to reconceive the writer-reader relationship. The following beliefs are offered as correctives to those listed above:

• The relationship implied by texts is not simply one way, from the writer to the reader, but between the writer and other writers, between readers and other writers, and between readers and other readers.
• The relationship is not monologic, but dialogic; the text is the writer's "turn" in a conversation.
• The relationship is not singular, but plural; through the text, the writer is involved in a variety of different relationships with different people, whether the text is for the self or a mass, anonymous readership.
• The relationship may not be created, defined, or bound by the text; it often exists prior to the text, changes during composing, and may be continued, altered, or ended by the text.

Writing as Conversation: Discourse in Community

The focus of this profile is the writing done by a group of social workers associated with the juvenile court system in Montreal. The excerpts I have selected come from write-aloud protocols and interviews collected before, during, and after the writing of a document called a Predisposition Report (PDR).⁴ The PDR is a social worker's advisory report to a judge on the sentencing (disposition) of a minor found guilty of an offence under the Criminal Code. Offences range from shoplifting to murder, and "consequences" (i.e., sentences) range from acquittal to up to three years' custody, but normally include some form of community service, victim restitution, and/or probation. Naturally, with such consequences, the PDR is a critical and difficult report; one worker described it as "the rope that can hang the kid."⁵

The PDR is usually 8-12 pages long and contains versions of the offence supplied by the adolescent ("client"), the client's parents, the police, and the victim (if there is a victim); these versions are collected by the social workers from face-to-face interviews, telephone calls, and a variety of documents. As well, following home visits and further interviews, the workers write detailed assessments of the adolescent and his or her family, and recommendations for sentencing. The report is 'entered as evidence at the adolescent's sentencing hearing and is read by the adolescent, the family, the judge, two lawyers -- the defense and the prosecution -- and court clerks and secretaries. In addition, a large number of potential readers exists, from social workers at other points of service to psychologists and psychiatrists. The PDR is an example of Bakhtin's "heteroglossia": the text is full of the words of others; it grows out of and enters back into an intricate network of voices, roles, and relationships.

In addition, the PDR does not exist in just one discourse community, but, rather, must overlap a number of different communities, obeying the rules and conventions of each. Quite literally, a number of conversations are going on at once. A worker captured this multiplicity of readers nicely when she said, "I write with one pen and twenty hats." For example, since the PDR is a court document, the workers are constrained by the inadmissibility of certain kinds of information (hearsay evidence, for example). Workers may be called to the stand to justify their reports and may be censured or even found in contempt of court for anything from unfounded speculation to inclusion of inadmissible evidence. These constraints are part of a whole structure of formal, legal relations among judges, lawyers, and social workers. But the document goes to the family as well, and the workers, serve as probation officers for the adolescents they report on, so the text also acts as the focus for a set of other relationships, less formal but just as constraining. In an interview, a worker describes this task:

. . . you must be able also to give information to the judges and lawyers in a clinical, professional way, [and] it has to be understood by a parent who is very emotionally caught up into the situation about their child. You are describing them, you have to be careful not to hurt them also.

. . . in spite of the fact that this is [not] his first offence, (Mother] . . . let's add something . . . told this worker that [Client] should be returned home. Mother is going to be happy because I put her point of view and the judge will see how unrealistic and protective she is. Isn't it nice! You break . . . you kill two birds with one stone.

With regards to his sister, he continues to dislike her. . . . That's mildly put, but I'd rather say "dislike" than "hate." With regards to his sister, he continues . . . . You know, I have to be sensitive as well. They're going to be reading that in court and, uh, I don't want to make it more than it is.

Should I keep it vague or should I mention it? . . . Maybe I shouldn't even mention it . . .. Oh, what the heck, I shouldn't do that. No, this kid has the right to be considered innocent until he's undergoing trial, so I'm going to keep it . . .. I have no right of mentioning anything. Here we go . . .. Unless it would serve a purpose but, no, municipal by-laws are not criminal offences anyhow . . .. Alright, that's nice enough for me if I'm working with the kid but . . . I shouldn't put too much or else I won't get what I want.

The use of quotation to preserve a whole network of relations can be seen in this next protocol excerpt, spoken while a worker drafted the final part of a report section:

Um, last, how can I leave? I don't want to end like this, talkMother also stated that [Client] has a tendency to "tarnish the truth" when he is in some kind of troubleing about he's a compulsive liar. Um . . . how can I say that? . 'Who's the Judge for that? Is it an English judge, at least? Yeah, [Judge], so he's going to be able to read between the lines, which will not be too offensive for the kid, not too harsh, and does not risk to create a conflict in the family.

The use of other people's words or ideas in text is often a way of establishing alliances or oppositions with individual readers or groups. Attribution, use of quotations, and referencing are all methods of establishing, altering, and maintaining relationships within discourse communities. When, in academic discourse, we use another's criticism of a community member, rather than being critical ourselves, we often do so to preserve our relationships in much the same way that the social worker uses the mother's quote in the example above. Sometimes, however, quotation may jeopardize rather than maintain relationships:

. . . when this worker asked what he thought he might receive as a sentence for this offence [Client] answered: "A lot." He thought he deserved, like, I'm sure the defense lawyer's going to blast me if I leave that. This looks like the kid is approving of whatever will be given.

Although the description of texts as "products" suggests the ends or results of processes (in much the same way that rehearsal leads to performance), most texts operate as part of ongoing social processes. As Dorothy Smith explains, "In many formal organization contexts the making of factual records of various kinds is a continuous part of the enterprise" (260). Also continuous are the changing contexts for the reading of those records. The writer juggles relationships in the past, present, and future. Some of those relationships are set in advance, formally defined by the roles imposed by context. For example, how the social workers relate, through the text, to lawyers and judges is to some extent constrained by the roles and rules of legal contexts. However, those roles are played by individuals, and the text enters into the ongoing process of their relations. An audience metaphor suggests that relationships are created by the text and end with the text. But, as Karen Burke LeFevre has argued, "The beginnings and endings of rhetorical acts are . . . not clearly obvious or absolute" (41). In the excerpt below, from a discourse-based interview, the worker discusses a judge she refers to as "the iron lady" of the juvenile Court system:

I don't know if I mentioned that (the Judge] gives me sweat just to see the name, just to have to go and present a report. She is dry, mean, and she despises social workers and the social field in general. So, to produce a report that I personally am not satisfied of, in front of her, is a major source of concern. Ah, I'm going to try to have the two lawyers on my side, because I'm sure she's going to tear me into pieces.

. . . um, which Judge is this? God, I've got so many files I get mixed up in my judges. Smith. The Honourable, they like to be flattered, even if they are not all honourable. The Honourable . . . ah, what's his first name? Humphrey. Just put H., or else the kid's gonna have cramps laughing at the judge's name. The Honourable H. Smith.

A similar concern for interactions between and among readers is often expressed when writers attach c.c. (carbon copy) or distribution lists to their documents. Such lists alert all readers to all other readers, a handy ploy which exploits the influence texts have on the whole structure of relations within which they operate.⁶

There are many documents like the PDR: texts that serve as the focus for shifting sets of relations among writers and readers; texts that are part of ongoing social activities -- not products, but points in a process; texts which involve writers and readers in a variety of roles -- some public and socially prescribed, some personal and idiosyncratic.

Even writing which seems most like oration or performance for a collective and anonymous audience -- journal articles, for example -- may be the result of a considerable amount of conversation-like discourse. Ede and Lunsford and Reither and Vipond have demonstrated the degree to which spoken and written dialogue, and a variety of readers, may shape an academic article. Work by people such as Charles Bazerman, Greg Myers, and others has shown that academic and scientific documents are themselves part of an ongoing conversation, a conversation that contributes to the social construction of knowledge.

Unfortunately, school discourse rarely involves students in this type of interaction. In the final section of this article, I will describe some of the pedagogical problems associated with the audience metaphor, and suggest some classroom procedures that I believe can help students and teachers join in conversation.

Conversation in the Classroom

Much school writing, in all disciplines, is done for a vague person or group "out there in chairs." That "audience" might be implicit in school writing assignments, but it is there nonetheless, waiting to be persuaded, amused, informed, corrected, or otherwise acted on according to the rhetorical aims set or assumed by the teacher or student. Often, there is an audience of one, the teacher -- who expects some form of performance or oration, but rarely interaction.

Unacknowledged and unexamined, this underlying audience metaphor has serious pedagogical consequences. It perpetuates the myth that knowledge is produced in the individual mind and transmitted to others through discourse. It leads to an alienating sense of aloneness in student writers and a receptive passivity in student readers. And it places manipulative power in the hands of the writer, ignoring the community's role in controlling discourse and knowledge. These misconceptions can be corrected in classrooms across the curriculum, but for that to happen, students must see that disciplinary discourse is an ongoing debate; they must become actively involved in the construction of knowledge through discourse, and conscious of the ways in which communities use their power to shape the conversation of their members.

We might start by helping students discover that the individual essays, articles, experiments, or books they study are only pauses in a community's conversation; though they are frozen in print, they do not begin or end a discussion. Within any discipline, it is possible to trace controversies, note conflicts and debates, identify different schools of thought or shifts in opinion. Myers' analysis of a corpus of scientific texts, using methods associated with the linguistic analysis of conversation, uncovers a remarkable variety of tactics designed to challenge, maintain, and improve interpersonal relations. In other words, he describes interactions, not pronouncements.

Another step would involve offering our students readers, rather than an "audience," for their written work. If we are to capture the dynamics of conversation in our writing assignments, we must stop providing our students with a steady diet of ostensibly single-minded, "general" readers. In their place, we should offer a great variety of quite specific readers in many different relationships to the writer. When possible, the texts produced for those readers should be part of some larger discussion or project. We needn't invent or simulate writer-reader relationships; our classrooms are full of, and surrounded by, writers and readers.

For example, the engineering students in my writing class frequently write papers on technical or scientific topics for students at a local high school. A recent variation on that assignment had them working in teams of three to prepare a paper on physics for a senior (Grade 11) class, at the request of the physics teacher. They began by reading the high school physics text and selecting a topic they believed was poorly explained in the book. In their teams they discussed their topic (often quite heatedly), explained, argued, drew diagrams, and went off to do some reading. Next, each team collaborated on a first draft. Naturally, they read and responded to each other's work as they drafted. That draft was then reviewed in my classroom by a "committee of peers": each team received comments from at least one other team. Once a second draft was prepared, each team met with some of the high school students for a discussion of physics in general and their paper in particular; a major (voluntary) revision of the paper followed this meeting. Finally, the papers were given to the physics class, and we received feedback from the high school students and their teacher.

A second example concerns a course for graduate students in education. Modelled. roughly after the exemplary work done at Saint Thomas University (Reither and Vipond; Hunt, Parkhill, Reither, and Vipond), this course allowed students, individually or in small groups, to choose areas of special interest or concern within the broad topic of language across the curriculum (e.g., language and science education; use of journals; language and gender in education). There was no course text or required readings; instead, students found their own readings and other resources and recommended them to the class when they felt they were relevant. Typically, class time was taken up with brief presentations on the progress of individual research and thinking, small group discussion, and frequent freewriting on the many issues and controversies produced in class discussion or discovered in reading. Freewriting was circulated and, on occasion, edited by the group for classroom publication. Rather than citing "experts," students in the course began to refer to each other. Moreover, their writing became an extension of the classroom conversation: they acknowledged the different positions expressed in class, supported some, criticized others, and advanced their own arguments. The final, formal product of the course was a text of over 400 pages, which included individual and co-authored papers, freewriting, annotated bibliographies, small group position statements, and hundreds of teaching ideas.

In both these examples, students collaboratively constructed knowledge; they did not simply collect or consume, they created. Their writing joined a larger discourse, a conversation with many voices. Rather than writing for some imaginary or dimly conceived "audience," the students wrote for a variety of readers and reasons. The texts they produced were used, responded to, discussed; writing was a social activity and knowledge was its product. Such activities help dispel the myth that knowledge is produced in the individual mind and transmitted to others through discourse, a myth supported by the audience metaphor.

The metaphor also supports the myth of the autonomous, inventing author, and does not easily permit a critique of institutional power and its hold on the individual. After all, the audience does not tell the actor or orator what to say. To counter this myth, students need to discover that the communities within which they read and write influence, even determine, the meanings they can make. In order to become critical members of those communities, students need to see how groups both authorize and restrict discourse, how they permit some conversations while prohibiting or discouraging others. Foucault explains:

. . . in every society the production of discourse is at once controlled, selected, organized and redistributed according to a certain number of procedures, whose role is to avert its powers and its dangers, to cope with chance events, to evade its ponderous, awesome materiality. (216)⁷

Once students begin to experience discourse as conversation within communities, they can relocate themselves in the world. They can see themselves as talking with, rather than talking at. From this new perspective, they can more easily understand and critique discourse and community.

Although the concept of discourse communities has altered forever our sense of the writer-reader relationship, it will be difficult to remove the term "audience" from our conversations. The metaphor is rich, evocative, and ancient. But I believe we might do well to heed Frank Smith: "Metaphors are the legs of language, on which thought steadily advances or makes its more daring leaps. Without metaphor thought is inert, and with the wrong metaphor, it is hobbled" (117). How does "audience" make us think? What relationships are implied by the use of that term in our classrooms and fields of study? We are constrained in thinking and talking about these things by the ways we have thought and talked in the past. The knowledge we have constructed, the relationships we have imagined, make it difficult for us to see anew. One way of improving our vision is to consider critically our central metaphors and the structures, relationships, and dynamics they imply. Audience is one such metaphor. When Barthes announced the death of "the author," he was attempting to transfer attention and power to the reader. The "audience" may have to suffer a similar demise if we are ever to recognize the power of community.

Acknowledgements

I would like to thank Mark Aulls, Ann Beer, Rick Coe, Patrick Dias, Lester Faigley, Russ Hunt, Jim Reither, Graham Smart, Rachel Spilka, Doug Vipond, and others too numerous to mention for conversations about earlier versions of this article. I would also like to thank Nan Johnson and John Dyck, who reviewed the article for TSC.

Notes

1. Commentators (e.g., Ede; Kroll) usually trace the origin of the problem of "audience" to Plato and Aristotle. Plato argued that rhetoricians must classify human souls and speeches in order to 'show why one soul is persuaded by a particular form of argument, and another not" (271 a-d). Aristotle answered this challenge in Book II of his Rhetoric, where he attempted to describe "the various types of human character in relation to the emotions and moral states, to the several periods of fife and the varieties of fortune" (131-32). For contemporary discussions of  "audience" see, for example, Ede and Lunsford, Kroll, Long, Minot, Ong, Park.

2. Neither "social constructionist" nor "discourse community" are unproblematic terms. Petraglia's recent critique indicates that no cohesive social constructionist theory exists in composition and rhetoric. And, although there has been plenty of discussion about discourse communities for some years now, no widely accepted definition has emerged. In fact, like "audience," the term is likely to resist precise definition and thus gain connotative power. In addition, a number of other concepts have contributed to the term's rich etymology, for example, "interpretive community" (Fish), "intertextuality" (Porter), "heteroglossia" (Bakhtin), and 'speech community" (e.g., Gumperz) have all been used in efforts to describe what a discourse community is and does. By itself, this last concept has a lengthy tradition in sociolinguistics that can be traced back to 1923 and Malinowski's use of the terms "context of situation" and "context of culture." The effect of all this cross-fertilization is that a rich discussion of "discourse community" his developed in a relatively short period of time.

3. The metaphoric implications of the term may be dependent on factors of context and genre. Perhaps some groups of readers could accurately be called an "audience," although I find it difficult to conceive of such a collective. Are the people who subscribe to a newspaper or magazine an "audience"? Are the differences between readers of literary and non-literary texts sufficient to call one or the other an "audience"? Walter Ong says that the audience is always a fiction created by the writer: "A reader has to play the role in which the author has cast him, which seldom coincides with his role in the rest of actual life" (12). This may be true for literary texts but, as I argue in this essay, it is clearly not true for many non-literary texts. I am concerned here mostly with what Dorothy Smith calls "documentary texts," for which readers' and writers' roles are often set less by text than by context. In the report described here, the social worker may "cast" the judge as a sympathetic or outraged reader, but she has no say in his predetermined, "actual life" role as judge, a socially sanctioned and defined role.

4. The currently preferred methodology for research of this type is ethnography, however, concern for confidentiality prevented me from using that method in this situation. The mix of methods I used was designed to balance the advantages and disadvantages of each method. Discussions of protocol analysis by Flower and Hayes and of discourse-based interviews by Odell, Goswaml, and Herrington can be found in the same collection of essays. I use the term "write-aloud protocol" because I believe it makes a more supportable claim than the term "think-aloud protocol."

5. The social workers are all bilingual (English/French), but write most often in English. Errors in the excerpts are attributable to the fact that French is the first language for some of the workers.

6. See Schwartz for a writing assignment that exploits the often conflicting interests of the primary and carbon copy readers.

7. Foucault's essay, "The Discourse on Language," describes three sets of rules governing discourse. At one level, all of Foucault's work might be read as an analysis of the relationships among discourse, knowledge, and power.

Works Cited

Aristotle. The Rhetoric of Aristotle. Ed. and trans. Lane Cooper. New York: D. Appleton Century, 1932.

Bakhtin, Mikhail. "Discourse in the Novel." The Dialogic Imagination. Trans. Caryl Emerson and Michael Holquist, Austin: U of Texas P, 1981. 259-422.

Barthes, Roland. Image - Music - Text. New York: Hill, 1977.

Bazerman, Charles. "Scientific Writing as a Social Act." New Essays in Technical Writing and Communication. Eds. P.V. Anderson, R. J. Brockmann, and C. R. Miller. Farmingdale, NY: Baywood, 1983. 156-184.

Bruffee, Kenneth. "Social Construction, Language, and the Authority of Knowledge: A Bibliographic Essay." College English 48 (December 1986): 773-90.

Burke, Kenneth. The Philosophy of Literary Form: Studies in Symbolic Action. Berkeley: U of California P, 1973.

Clark, Gregory. Dialogue, Dialectic, and Conversation: A Social Perspective on the Function of Writing. Carbondale, IL: Southern Illinois UP, 1990.

Cooper, Marilyn. "The Ecology of Writing." College English 48 (April 1986): 364-375.

Ede, Lisa. "Audience: An Introduction to Research." College Composition and Communication 35 (May 1984):140-154.

Ede, Lisa, and Andrea Lunsford. "Audience Addressed/Audience Invoked: The Role of Audience in Composition Theory and Pedagogy." College Composition and Communication 35 (May 1984): 155-171.

Elbow, Peter. "Closing My Eyes As I Speak. An Argument for Ignoring Audience." College English 49 (January 1987): 50-69.

Faigley, Lester. "Nonacademic Writing: 'The Social Perspective." Writing in Nonacademic Settings. Eds. Lee Odell and Dixie Goswami. New York: The Guilford Press, 1985. 231-248.

Fish, Stanley. Is There a Text in This Class? Cambridge: Harvard University Press, 1980.

Foucault, Michel. The Archaeology of Knowledge and The Discourse on Language. New York: Pantheon Books, 1972.

Gumperz, J. Language in Social Groups. Stanford, CA: Stanford UP, 1971.

Harris, Joseph. "The Idea of Community in the Study of Writing." College Composition and Communication 40 (February 1989): 11-22.

Hayes, John and Linda Flower. "Uncovering Cognitive Processes in Writing: An Introduction to Protocol Analysis." Research on Writing: Principles and Methods. Eds. P. Mosenthal, L. Tamor, and S. Wa1msley. New York: Longman, 1983. 206-220.

Hunt, Russell, Thom Parkhill, James Reither, and Douglas Vipond. "Writing Under the Curriculum: Learning to Write By Using Writing to Teach."Panel discussion. Conference on College Composition and Communication. St. Louis, March, 1988.

-----. "Finding Identity in Community: Crafting Knowledge through Collaborative Inquiry and Writing." Panel discussion. Conference on College Composition and Communication. Chicago, March, 1990.

Kroll, Barry. "Writing for Readers: Three Perspectives on Audience." College Composition and Communication 36 (May 1984):172-185.

LeFevre, Karen Burke. Invention as a Social Act. Carbondale, IL: Southern Illinois UP, 1987.

Long, Russell. "Writer-Audience Relationships: Analysis or Invention?" College Composition and Communication 31 (May 1980): 221-226.

Malinowski, Bronislaw. "The Problem of Meaning in Primitive Languages." In C. K. Ogden and I. A. Richards, The Meaning of Meaning. London: Routledge and Kegan Paul, 1923. 451-510.

Minot, Walter. "Response to Russell C. Long." College Composition and Communication 32 (October 1982): 335-337.

Myers, Greg. "The Pragmatics of Politeness in Scientific Articles." Applied Linguistics 10 (March 1989):1-35.

Oakeshott, Michael. Rationalism in Politics. New York: Basic Books, 1962.

Odell, Lee, Dixie Goswami, and Anne Herrington. 'The Discourse-Based Interview: A Procedure for Exploring the Tacit Knowledge of Writers in Nonacademic Settings." Research on Writing. Principles and Methods. Eds. P. Mosenthal, L. Timor, and S. Walmsley. New York: Longman, 1983. 221-236.

Ong, Walter. "The Writer's Audience is Always a Fiction." PMLA 90 (January 1975):9-21.

Park, Douglas. "The Meanings of 'Audience'." College English 44 (March 1982): 247-257.

Petraglia, Joseph. "Interrupting the Conversation: The Constructionist Dialogue in Composition." Journal of Advanced Composition 11 (Winter 1991): 37-55.

Plato. "Phaedrus." The Works of Plato. Trans. Benjamin Jowett. Ed. Irwin Edmin. New York: The Modern Library, 1928. 263-329.

Porter, James. "Intertextuality and the Discourse Community." Rhetoric Review, 5 (Fall 1986): 34-47.

Reither, James A. and Douglas Vipond. "Writing as Collaboration." College English 51 (December 1989): 855-867.

Rorty, Richard. Philosophy and the Mirror of Nature. Princeton: Princeton UP, 1979.

Russell, David. "Writing Across the Curriculum in Historical Perspective: Toward a Social Interpretation." College English 52 January 1990): 52-73.

Schwartz, H. J. "Writing with the Carbon Copy Audience in Mind." College Composition and Communication 39 (February 1988): 63-65.

Selzer, Jack. "More Meanings of 'Audience'." The Rhetoric of Doing: Essays Honoring James Kinneavy. Eds. R. Cherry, N. Makadate, and S. Witte. Carbondale: Southern Illinois UP, (In press, 1992)

Smith, Dorothy. "The Social Construction of Documentary Reality." Sociological Inquiry 44 (1974): 257-268.

Smith, Frank. Essays into Literacy. London: Heinemann, 1983.

Spilka, Rachel. Adapting Discourse to Multiple Audiences. Invention Strategies of Seven Corporate Engineers. Unpublished doctoral dissertation. Carnegie Mellon University, 1988.

Trimbur, John. "Consensus and Difference in Collaborative Learning." College English 51 (October 1989): 602-616.