Readers, Writers, and Texts: Writing In the Abyss

Jasper P. Neel

Radical ideas receive better treatment when they are supported by ancient tradition. The pedagogy of the advanced com- position course described below does not seem radical to me, but, judging from the reaction generated when I have spoken about the course, the theories about writing that caused the pedagogy apparently are quite radical. Thus I will begin with ancient tradition.
When all things began, the Word already was. The Word dwelt with God, and what God was, the Word was. The Word, then was with God at the beginning, and through him all things came to be; no single thing was created without him. John 1.1. A case should, in fairness, be fought on the strength of the facts alone, so that all else besides demonstration of fact is superfluous. Nevertheless, external matters do count for much, because of the sorry nature of an audience. Aristotle's Rhetoric 3.1.
Sing, goddess, the anger of Peleus' son Achilles. Iliad I, 1. Tell me, Muse, of the man of many ways . . .. Odyssey I, 1.

The first quotation implies that language and external reality are mutually dependent. No purely interiorized language can exist; no purely exteriorized reality can exist. The second quotation states bluntly that purely factual writing cannot occur. The nature of the audience and the available means of persuading it must always be considered. Once rhetorical strategies occur, pure science, pure knowledge, and pure fact vanish. The third and fourth quotations demonstrate that very early in human history writers credited their compositions to a force other than themselves: a muse, a voice, some unknown speaker both part of and part of themselves.

I recognize that we are long past the age in which an appeal to the ancients works rhetorically. My point is not to justify my case by such an appeal but to show that my ideas about texts, readers, and writers are hardly new. Section IV below and Appendices 1-3 explain my procedure in teaching advanced composition. Sections I-III summarize the theoretical foundation for that procedure. Though the procedure itself is too complex for classes below the sophomore level, the theories described in I-III and the implications summarized in V apply to composition at any level.

I. Texts

"Writing," says Eric Havelock in Preface to Plato, "made possible the separation of the knower and the known."1 This is, of course, self-evident. It is so obvious that saying it seems silly. But it is also profound. Nowadays it is difficult to imagine the wonder this new technology must have engendered. For the first time human knowledge could be used without the knower of the knowledge present. It could even be used after the knower was dead. Human knowledge was no longer limited to what one person could remember, and a powerful technology for describing and altering human circumstances was available. So powerful was this technology that speculation about how it could and should be used has gone on for millennia. The last twenty years have been a particularly fruitful period for such speculation.

Some of the most influential and profound work on the effect of writing on human beings and human civilization has been done by Walter Ong, who argues that the psychological base for writing is alienation. "With writing," Ong says, "the earlier noetic state undergoes a kind of cleavage, separating the knower from the external universe and then from himself." Writing allows for art and for science, "but it does so at the price of splitting up the original unity of consciousness and in this sense alienating man from himself and his original lifeworld." What Ong calls "disengaged, pure thought," is a product of literacy which has as its first consequence alienation.

Next Ong argues that the "real word" is the spoken word, which "is always an event . . . an action, an ongoing part of ongoing existence." The written word, in contrast, is an immobile objectification of this event: "Writing and print, despite their intrinsic value, have obscured the nature of the word and of thought itself, for they have sequestered the essentially participatory word . . . from its natural habitat, sound, and assimilated it to a mark on a surface, where a real word cannot exist at all."2

The cultural effect of writing has been to convince most educated people that real language is the written language of professional writers while spoken language is inconsequential. Thus, we have style manuals,, grammar handbooks, and dictionaries, all purporting to tell us what words really mean and how language should work. "Permanent unreality [meaning print]," Ong surmises, "is more plausible and comforting than reality that is transient [meaning speech]." Chirographic and typographic biases actually redefine the psyche and make themselves a valued norm. They create artificial,, but insidiously powerful, graphocentrism. Ong's startling conclusion is that both the writer and the reader are masks, "For writing is itself an indirection. Direct communication by script is impossible."3

Ong's work is highly theoretical and based on his extensive study both of oral cultures (especially in Africa where he has studied and taught) and of the cultural effects of European rhetorical theory since Ramus. In the last fifteen years,, however, work of a more specialized sort on the nature of texts has been done by two groups—on one hand by psycholinguists and reading theorists, and by post-structuralist literary critics on the other.

The most accessible and influential work on the psycholinguistics of reading has been done by Frank Smith, who argues that most (perhaps ninety percent) of the information generated in the reading process is generated in the brain through a process of forecasting and confirmation. In other words, the reading channel works this way:

assumption in brain --------> activation of visual - neural apparatus ------- > confirmation on page

The brain literally tells the page what it expects to see. "The brain," says Smith, "contributes more information to reading than the visual symbols on the printed page." Reading is an oscillation between prediction and confirmation; in effect, telling the text what it will say is what allows proficient readers to operate at such a tremendous rate.4

Of course, the idea that the transaction of meaning in the reading process is more an imposition on the text than an acquisition from it is not entirely new. In 1908 E. B. Huey's Psychology and Pedagogy of Reading laid the foundation for modem theories focusing on the reader's contribution, and John Dewey's Art as Experience (1938) argues that every reading experience is unique with each reader creating "something new, something previously not existing."5

In 1975, Eleanor Gibson and Harry Levin argued Dewey's case more fully. Mature readers, they show, use highly "economical" strategies, processing very little of the text's vocabulary and punctuation. An adept reader moves across the surface of the printed text relying in a highly idiosyncratic, entirely unconscious way on the barest minimum of the chirographic marks.6

Two collections of essays that appeared at the end of the 1960s (The Psycholinguistic Nature of the Reading Process and Basic Studies in Reading) push the reader-centered argument even further. Some of the conclusions reached by the writers in these books are that fluent readers do not identify letters at all and rarely identify words. Rather, they skip directly from a few distinctive features to comprehension, and this leap occurs right away. Reading depends on separate word identification to about the same degree as listening.7

In the early 1970s Kenneth and Yetta Goodman and others working in the area of reading miscue analysis showed almost beyond question that readers can comprehend texts in dialects and with vocabularies that the readers themselves have not mastered. More recently, reading theorists have argued that texts depend on the pre-existing "schema" in the reader's brain. Texts provide directions by which readers construct meaning based on their prior knowledge of the world. Reading is not the retrieval of information stored in a text; indeed, the text must provide a significant number of directions fitting into the reader's decoding schema or the reader will be unable to read at all.8

The most radical analysis of the relationship between reader and text has been advanced by Marxist revolutionary Paulo Freire, who contends that reading ability depends on the self-conception created by an encounter with a text. Literacy campaigns in capitalist countries are always doomed, says Freire, because it is impossible to be both oppressed and literate. If basic literacy teachers do not wish to change the political status of adult illiterates, nothing happens. Were reading politicized, however, and presented as a way of overthrowing the oppressive existing order, illiterates would learn to read quickly as they have done in Cuba and Latin America.9

My point in surveying all these theories of reading is to show that there is persuasive evidence that texts exist mainly in the minds of readers. They do not exist the same way in any two minds, and if they do not include enough of the appropriate signals or if they force on the reader an undesirable self-image, they do not exist at all. Physically they exist, of course; the marks are there on the page. But practically they are useless. The writer must recognize that the reader, rather than listening to the text, talks to it. If the text cannot account for a significant percentage of what the reader says, the reader stops talking, and the text remains nothing but marks on a page. Such ideas raise interesting questions about the nature of texts. In particular, they suggest that a text may be more an activating device for the self- creation of a reading persona than a repository of new and unimagined information.

Anybody brave enough to read PMLA, Critical Inquiry, New Literary History or any other of the journals focusing on literary theory surely knows that in critical circles a lot is being said about the dubious nature of texts. In 1980 Jane Tompkins collected a group of essays entitled Reader-Response Criticism: From Formalism to Post-Structuralism, which includes essays by such people as Walker Gibson, Gerald Prince, Walter Michaels, Michael Riffaterre, Wolfgang Iser, Stanley Fish, David Bleich, Jonathan Culler, and Norman Holland. The essays, she says in her introduction, destroy any possible belief in the objectivity of a text. While it is true that Tompkins (as well as the other literary critics I discuss below) is writing about "literary" texts, they are texts just the same: consisting of writing and occurring as a transaction between a writer and a reader.10

In effect, there are three general groups of post-structuralist critics. The reader-response group argues that a text without both a reader and a context is no text at all. In the words of Stanley Fish, "there always is a text . . . but what is in it can change, and therefore at no level is it independent of and prior to interpretation." Fish questions the distinction between "literary" and "ordinary" language and argues that definitive readings are "tricks" because the "evidence" on which they depend "is always a function of what it is to be evidence for, and is never independently available . . . the interpretation determines what will count as evidence for it, and the evidence is able to be picked out only because the interpretation has already been assumed."11

Norman Holland proposes a reader-response model where reading is mediation with a text; "not an analysis that 'gets something out of it,' but a replenishment of the words that can be different for every literant and occasion."12 David Bleich goes even further than Fish and Holland, arguing that any attempt to explain how other readers ought to read a text is a falsification of the explainer's own experience while reading.13

Semiologists argue that in reality there is no meaningful distinction between the writer, the text, and the reader. All operate within a predetermined and predetermining sign system. "On the stage of the text," says Roland Barthes, there are "no footlights: there is not, behind the text, someone active (the writer) and out front someone passive (the reader); there is not a subject and an object. The text supersedes grammatical attitudes: it is the undifferentiated eye which an excessive author (Angelus Silesius) describes: 'The eye by which I see God is the same eye by which He sees me.'"14

The deconstructors extend post-structuralist suspicion of the text even further, concluding that because texts have no real authority, authoritative readings are impossible. After adopting the stance advocated by Jacques Derrida, one becomes skeptical, even antagonistic, toward texts, asking not what the text says, but what it hides. Writing, says Derrida, occurs "sous rature," under erasure. Rather than being the embodiment of present meaning it is the hiding of absent meaning. Deconstruction is the strategy whereby an analyst scrutinizes the traces of absence in the text in order to show that the text originated not as the presence of information to be "communicated" but as the hiding of a gap or empty space which is obscured by the unending play of signifiers.15

In varying degrees Harold Bloom, Paul de Man, Hillis Miller, and Geoffrey Hartman are all influenced by Derridean analysis. Bloom, who contends that he is not a deconstructor at all, sees reading as "combat." It is not polite, nor does it fit with the academy's demand of civility. Reading is really a sort of aggressive, transgressive "misreading" through which the reader takes the text away from the writer and masters it.16 De Man's strategy is to undo the "system of tropological transformations" by which the text hides its rhetorical nature. Once the string of tropes has been exposed the reader discovers a pattern of rhetorical play extending metonymically in one direction and metaphorically in another but never coming to rest on the disclosure of truth. The text uses its "rhetoricity" to conceal that its only existence is rhetorical.17

Hillis Miller shows that by their very nature texts cannot be complete. The most pressing necessity for the author is "to draw a line around the material to be treated, to give it an edge or border which appears as a natural stopping place . . . beyond which there is nothing relevant to the subject" and then treat everything within the charmed circle as if it were treated completely with nothing omitted. No text, according to Miller, can be brought to closure; thus, all texts are "unreadable," Miller's term for the "presence in a text of two or more incompatible or contradictory meanings which imply one another or are intertwined with one another, but which may by no means be felt or named as a unified totality."18

Geoffrey Hartman equates reading with theft. To read, the reader must "master" the text, and that requires "stealing" it from the author. The reader creates a fictional version of himself which allows him enough distance from the text to master it. This pretended self is, for Hartman, the Archimedean instrument by which the text can be displaced and stolen from the writer. Like the other deconstructors, Hartman refuses to equate a text with "embodied meaning"; he urges quite the opposite. Written words always carry with them a certain absence, or indeterminacy of meaning. The reader's job is to overthrow the text by showing the original absence or exposing the indeterminate chain of signifiers.19

In sum, there are a lot of questions being asked about the nature of texts. It would certainly be possible to reject these questions as trivial, but two of them seem to have profound implications for the teaching of writing. If in fact the process of reading consists of talking to a text from a pre-existing mental structure, and if texts are less the presence of meaning than the hiding of absence, it seems very likely that the "product" offered by a writer is much more illusory and undefinable than anyone in the composition industry has dared admit.

II. Readers

From the beginning, rhetorical study was the study of audience manipulation. As anyone who has read it knows, Aristotle's Rhetoric "is a searching study of the audience. . . . A speech is to be judged by its effect on someone. Since discourse has its end in persuasion, the speaker or writer must know the nature of the soul he wishes to persuade . . . each single item in the speech is to be determined by its effect upon the soul."20 Early in the Rhetoric Aristotle makes a point something like this: In a world of mere fact, pleaders (lawyers) would have nothing to say. But the world is not mere fact. It is not even largely fact. Indeed, it may not be fact at all, only opinion. Hence the goal of rhetoric is "the production of a certain mental attitude in the judge."21

In an essay first printed in PMLA and then reprinted in Interfaces of the Word, Walter Ong explains how a writer produces a text that will create this "desired mental attitude." The writer "has to make his readers up, fictionalize them." Learning how to give body to an audience is perhaps the most difficult thing for a writer. If he succeeds, "it is generally because he can fictionalize in his imagination an audience he has learned to know not from daily life but from earlier writers." First, "the writer must construct in his imagination . . . an audience cast in some sort of role." Second, the writer must tell the readers how they are to fictionalize themselves and tell them in such a way that they will agree.22

Walker Gibson defines this same idea in a slightly different way by saying the writer must create a "mock reader," "whose existence is entirely a function of the text." Gibson, explains that "a bad book is a book in whose mock reader we find a person we refuse to become." Above all, there must be enough clues in the text for the reader to know who he or she must become in order to process the text.23

This explains why most English teachers cannot read grocery-store romances. English teachers are unwilling, perhaps unable, to become the reader implied by those romances. There are, of course, millions of people who become that implied reader eagerly. This also explains why so much writing by composition students is so intolerable. Poor writers do not know that they "invent" a reader and then give clues in their text so that actual readers know who they are supposed to be. The writing by most composition students implies no mock reader at all. It is difficult enough to read a text that implies a mock reader we are reluctant to become. It is nearly impossible to read a text that implies no mock reader at all. In fact, no one can do it except composition teachers who are paid to do it. But reading for pay is a highly unusual activity, occurring only in the extremely rarefied, otherworldly environments of the composition and publishing industries.

III. Writers

Writers, like readers, must be fictionalized. Every piece of prose that has ever been written was the product of a fictional writer created in the imagination of the author. This fictional writer, who both is and is not the author, lives and dies with the prose through which the authorial voice speaks. Wayne Booth says that, "As he writes, [the real author] creates not simply an ideal, impersonal 'man in general' but an implied version of 'himself' that is different from the implied authors we meet in other men's works. . . . Whether we call this implied author an 'official scribe,' or . . . the author's 'second self'—it is clear that the picture the reader gets of this presence is one of the author's most important effects. However impersonal he may try to be, his reader will inevitably construct a picture of the official scribe who writes in this manner—and of course that official scribe will never be neutral toward all values." "Our sense of the implied author," Booth adds, "includes . . .the intuitive apprehension of a completed artistic whole; the chief value to which this implied author is committed, regardless of what party his creator belongs to in real life, is that which is expressed by the total form."24

Building on Booth's argument, Seymour Chatman concludes that all prose carries with it a fictional author, but he goes on to say that all texts are mediated. That is, no value-free text, no purely factual text can exist because every text exists through the mediation of a not entirely knowable, yet entirely fictional, writer.25

Geoffrey Hartman argues a much more radical position even than Chatman's. Writing, for Hartman, is a kind of "forgery" in which metaphoric and metonymic strings pretend to embody a reality of meaning that could exist only outside the metaphoric and metonymic chains of displacement that constitute language. Writing is also "violation" because anything inscribed is stolen. No one born into language can occupy an original place because all the places available in language have already been occupied by many, many others. The closest one can come to originality is to steal something that has already been thought and said by someone else. Even this original space is compromised, however, because the stolen words must be presented in a sign system that will always be older and more powerful than the writer. In "deconstructing" the mode that he calls "managerial social science" (by which he means something akin to "expository prose"), Hartman traces the growth of faith in the communicative ability of this medium and then attempts to expose the medium as nothing more than illusion.26

Writers, then, are (1) only partially identifiable with the character of the person holding the pen (Booth), (2) a not entirely knowable yet entirely fictional medium through which a text gets transmitted (Chatman), and (3) forgers or thieves who must steal a place in a sign system that largely forms them (Hartman). Were one to base a composition on these three, instead of saying, "Be clear, brief, and sincere,"27 one might well say, "Be clever, deceitful, and covetous."

The idea of implied author, persona, or mask—which in the New Criticism was treated under the rubric point of view—is hardly new in discussions of literary texts, but it is quite new for the composition industry, most of whose textbooks, both basic and advanced, merely assume that the author of "expository" writing is the writer himself. It is radical to argue that the writing process is not a process of self-discovery but rather a process of learning to create a fictional self who is embodied in the prose, who exists only in the prose, and who can be made to say and believe nearly anything. This process of fictionalization goes on in all advanced composition assignments. It even goes on in English 101; indeed, one could argue that the levels of fictionalization are especially intense in 101. The following diagrams show what I mean. The assignment is for a 101 student to write an essay on nuclear energy; it could have been on abortion, ERA, football, Reaganomics, assembling a bicycle, the theme of "Lamia," chemotherapy, or nearly anything else.

Deep Situation
Writer

student in English who wants to pass
Message

give me a good grade
Reader

teacher paid to read essays prove writing ability, and fail those who do not write well enough

Situation Implied by Text
Fictional Writer

Someone with something to write about nuclear energy who has found a motive and opportunity for writing
Fictional Message

something about nuclear energy (e.g., its history, how it differs from other forms of energy, whether it is good or bad, how it works, etc.)
Fictional Reader

someone who has a motive and opportunity for reading what the writer has written

The freshman student must somehow know how to create this fictional self, how to make the fictional message look like a real message, and how to create a mock reader whom the teacher is willing to become. It is indeed hard to overestimate the complexity of this situation. Of course it is not so difficult for advanced composition students, who have already learned the process of embedding one text in a seemingly unrelated one, for if they had not learned this process, they would never have arrived in advanced composition in the first place.

IV. Pedagogy

Before describing the actual procedure of my advanced composition course, I will sum up the three theories that support it: 1) Writing is by its very nature falsification. No comprehensive, definitive reading of a written text is possible, though many partial and fully satisfactory readings are possible. 2) For writing to occur, the writer must make up (fictionalize) an audience, and for reading to occur the audience must fictionalize itself as the reader implied by the clues in the writing. 3) To write, the writer must create an author through whose voice the writing can occur. This author lives only so long as the text embodying him lives.

While it is very difficult to get my students to accept the above premises, once they do, the actual assignments in the course are satisfying for them and a joy for me. Throughout the course all students are required to describe in detail each situation in which they will write. The minimum criteria are these:

1. finding a person whose job involves the kind of writing task assumed
2. visiting that person's place of work, talking about the writing required, and getting several samples of previous work
3. talking with some member of the audience for this work and learning how the work is evaluated
4. preparing a detailed Jakobsonian diagram of the writing situation (see Appendix 1 below)
5. preparing Austinian and transformational analyses of the prose models chosen for imitation (see Appendices 2 and 3 below)
6. writing
7. doing a close analysis of the finished product.

All this effort as a preface to composing does not change the nature of written words: they still constitute an indeterminate, illusory medium, produced by a fictional writer and interpreted by a fictional reader. But the effort does set some fairly clear boundaries about what sort of fictional readers and writers are acceptable, and it makes the students a little more comfortable with a medium that seems absolutely precise yet is absolutely metaphoric. Some students are unwilling to accept the theories supporting the course and drop. Those who stay find themselves fictionalized as everything from literary critics and evangelical preachers to CPAs and tractor salesmen.

I will give one example of the sort of thing that can occur so that the analytical diagrams described in the Appendices will be clear. One of my students studied the writing of the Vice-President of a rental uniform service. She met with him several times, got to know him well, and read almost everything he wrote during a one-month period. One of the documents she analyzed was a memo to the company's account representatives. A typical sentence in the memo read as follows: "The Company's ultimate, long range objective is the attainment of full laundry services market dominance in the citywide area." Analysis of this sentence and the context in which it was written appears below. The analysis was done by the student (with a good deal of help from me).

Jakobsonian Analysis
1. Writer: He is a newly-hired Vice-President for marketing who is young and ambitious, and he is the new boss of the sales force. In this memo he is announcing a new sales plan designed to increase the volume of business and letting the sales force know that a tough new manager has been hired to shape things up.
2. Readers: They are middle-aged salesmen who have been fairly comfortable servicing old accounts and who are not ambitious. They want to keep their jobs.
3. Context: The relationship is boss to employee in an old established company that is now threatened by several new, aggressive competitors.
4. Message: This sales force will help put the new competition out of business or a new sales force will be found. Salesmen will make at least six new cold calls each day. Voluntary cutting of commission margins (e.g., secret kickbacks) to reduce prices will be regarded as company loyalty.
5. Code: Managerialese
6. Contact: Extremely intimidating tone.

Austinian Analysis
Locution

The company's, ultimate long range objective is the attainment of full laundry services market dominance in the city wide area.
Illocution

Ordering
Perlocution

High pressure sales visits to companies' using other services

Transformational Record

The Kernel sentence was probably something like, "I want a monopoly." No doubt the author was never conscious of the kernel, which passed through at least five obligatory transformations before it was inscribed.

1. Adjectivalization, which adds seven adjectives to an adjectiveless kernel

2. Polysyllabification, which adds ten polysyllabic words to a kernel with only one

3. Nominalization, which doubles the number of nouns in the kernel

4. Redundancy, which calls for doubling adjectives or adjective-noun combinations (e.g., ultimate long range or full dominance)

5. Depoliticization, which changes ugly words like "monopoly" to "full laundry services market dominance" and personal pronouns ("I") to abstract nouns ("the company").

After doing this sort of analysis, students are in a position to write a follow-up memo. They fictionalize themselves as the young, ambitious (and probably unscrupulous) Vice-President and fictionalize the audience as they might be two or three weeks after the first memo. Written language is the tool the students use to bring this fictional audience into the desired mental state.

V. Implications

The course is method, not content. It is not training in a specific kind of writing; it is training in how to find out what to do. Once students learn that written language is objectification, and thus falsification, of natural language, they are freed from the tyranny of abstract correctness. Once they learn how to make up an audience and then make up a self to speak to that audience, most of their problems are solved. The result is the most powerful, most productive heuristic I have ever encountered. The course does, however, have two implications, the second troublesome to many composition teachers: (1) writing is a value-free technology; thus, (2) learning to write is learning to manage a technology, not training to be a moral person.

That writing is a technology invented by humans is self evident. Of course like any technology it can be made into an art form, but it was a technology first and an art second. Like any technology, it is value-free. Written language is neither good nor bad until someone uses it. It is a technology just like mechanical engineering nuclear physics, and plumbing. As a technology it can lead to other technologies, such as nuclear physics, which would be unthinkable in a preliterate culture. But it can be separated from other technologies. Preliterate cultures can invent wheels and even sewage systems.

All this seems obvious enough. Few composition specialists would dispute it. The catch comes at the second implication. George Orwell's essay "Politics and the English Language" articulates a moral position about language accepted by many composition teachers. (Judging by the number of freshman anthologies that include the essay and by the almost violent reaction many of my colleagues have to my position, I am tempted to say most composition teachers.) Orwellians would argue that the sentence "The company's ultimate, long range objective . . . " is immoral while the sentence "I want a monopoly" is moral. They would also argue that training in precise, controlled, clear writing (the sort embodied, I hope, in his essay) improves the moral fibre of the student. As a result, many composition teachers try to teach and will reward only this style. Composition courses, thus, become training in morality, not training in technology. This is just as wrongheaded as the old literature-based composition courses that attempted, with precious little success of course, to teach every American student to write like Cleanth Brooks and Robert Penn Warren.

Forcing someone who chooses to write "The company's ultimate, long range objective . . . " to write "I want a monopoly" is not instruction in writing. It is instruction in morality—the morality of a highly educated, poorly paid humanist, who probably has very little experience managing a large company. In fact, the Vice-President in question could not have written "I want a monopoly." Had he revised his memo this way, he would have revised it right out of existence. The result of such revision would have been the death of one fictionalized writer and the birth of another. The new one would not be ambitious, greedy, arrogant, and cruel and thus would have no reason to write the memo in the first place. The fictional author of the long sentence is clearly a very different sort of person from the fictional author of the short one. The fictional audiences are also quite different as are the conceptions of writing.

It is possible in one or two semesters to teach fairly bright students how to fictionalize themselves as a variety of authors. It takes a lifetime to teach them what sort of authors they ought to want to fictionalize themselves as. The profession has learned that instruction in grammar produces students who score high on multiple-choice editing tests and the instruction in writing literary criticism produces adults who do not write well and do not read literature at all. The profession has not, however, learned that teaching writing is essentially an amoral enterprise. Undoubtedly it is frightening to think that students who master the technology of writing may use it for ends and in ways that we find repugnant. But it is a fact that we have very little control over the sort of fictional writers our students wish to become or over the sort of mock readers they are capable of creating.

Writing is surely the most powerful tool we humans possess, for through it we can become things we are not and make others into things they do not wish to be. But learning to use the tool well and learning to use it morally (even if we could define "morally") are quite different and probably mutually exclusive. No doubt all of the theorists summarized in Sections I-III attempt to be moral people. No doubt they hope their students will behave morally. Even so, the theories they have advanced about the phenomenon of writing suggest strongly that it is a highly abstract technology through which human beings and human society can be changed. Unfortunately, their theories also suggest strongly that gaining mastery over the technology has no effect on the soul of the new master. Someone did an excellent job of teaching the rental uniform service Vice-President how to write. He was, and is, quite successful, and his success depends in large part on his ability to use language as a weapon against weaker, less gifted people. We all wish he were more compassionate. Possibly he could be taught to be so. But teaching him compassion would have nothing to do with teaching him to write, which he already does masterfully.

Northern Illinois University
Dekalb, Illinois

Appendix I

Roman Jakobson developed this diagram in 1960.29 Ross Winterowd modified it, making a heuristic of it in 1975.29 This is a modification of Winterowd's modification. In the diagram there are six primary aspects of the contract between writer and reader. The relationship among the six is as follows:

Context

Writer--------------------Message-------------------Reader

Code

Contact

The minimum amount of information about each aspect for writing to occur is:
Writer. Explain who the writer is by giving as many characteristics as possible. Give the most important reasons why the writer is writing.
Reader. Explain who the reader is by giving as many characteristics as possible. Give the most important reasons why the reader is reading.
Context. Describe the relationship between writer and reader.
Message. Describe the one or two most important effects the writer wishes to have on the reader.
Code. Describe the appropriate set of linguistic conventions in which this piece of discourse must occur. (The jargon, syntax, punctuation, etc. varies from code to code. No one, for example, would confuse the language of post-structuralist literary criticism with supply-side economics or economics theory with a CPA's audit report.)
Contact. Describe the writing strategies (phatic devices) used to keep the reader reading.

Appendix 2

This "writing act" diagram is developed from the work of John Austin and John Searle.30 Statements, according to Austin, have three aspects: grammatical, intentional, and effectual. He names grammar "locution," intention "illocution," and effect "perlocution." For example:
Locution Illocution Perlocution
Fred, jump in the lake) order (Fred jumps in the lake
Play it again, Sam) remember (nostalgia
That's my last Duchess) frighten (warn the Count's emissary

Appendix 3

Students do not need to know generative-transformational linguistics to learn to recognize kernel sentences and the optional transformations that can occur. As they analyze the prose of their model, they look for the most frequent transformations, which are then designated "obligatory" for the code they wish to write.

1. Example sentence found by a student: "It is the professional opinion of the investigating officer that, all things carefully considered, the parolee should have his parole entirely revoked and that he should be sent immediately back to the penal institution."

2. Possible kernels suggested by student.

a. "I recommend reincarceration."

b. "I recommend revoking parole."

c. "Send Jones back to jail."

3. Transformational record of the growth of the sentence (assumed kernel, "I recommend reincarceration"):

Obligatory transformation 1, depersonalization: "I" becomes "the investigating officer."

Obligatory transformation 2, passivization: "recommend" becomes "be" + subject complement + expletive "it."

Obligatory transformation 3, extension and redundancy: "reincarceration" becomes two subordinate clauses, both saying the same thing.

Obligatory transformation 4, modification: a modifier-free kernel adds "professional," "investigating," "all things carefully considered," "entirely," "immediately," and "penal."

1(Cambridge: Belknap Press of Harvard Univ. Press, 1963), pp. 169-190.

2Interfaces of the Word: Studies in the Evolution of Consciousness and Culture (Ithaca: Cornell Univ. Press, 1977), pp. 17-21. See also Havelock, pp. 197-214.

3 Ong, pp. 21-22.

4 Psycholinguistics and Reading (New York: Holt, Rinehart, and Winston, 1973) and Understanding Reading: A Psycholinguistic Analysis of Reading and Learning to Read, 1971), pp. 1-28. See also Kenneth Goodman, "Reading: A Psycholinguistic Guessing Game," in Current Topics in Language, ed. Nancy A. Johnson (Cambridge: Winthrop, 1970), pp. 370-383.

5 Huey, The Psychology and Pedagogy of Reading (1908; rpt. Cambridge: MIT Press, 1968). Dewey, Art as Experience (New York: Minton Balch, 1934).

6 The Psychology of Reading (Cambridge: MIT Press, 1975).

7 Kenneth Goodman, ed., The Psycholinguistic Nature of the Reading Process (Detroit: Wayne State Univ. Press, 1968); Haryy Levin and Joanna Williams, eds., Basic Studies in Reading (New York: Harper and Row, 1970).

8 For an introduction to Reading Miscue Analysis, see P. David Allen and Dorothy J. Watson, Findings of Research in Miscue Analysis: Classroom Implications (Urbana: NCTE Press, 1976). For a brief overview of schema theory, see Anthony Petrosky, "From Story to Essay: Reading and Writing." College Composition and Communication, 33 (1982), 19-36. For a more thorough introduction, see Robert Schank and Robert Abelson, Scripts, Plans, Goals, and Understanding (Hillside, N. J.: Erlbaum, 1977) and Marilyn J. Adams and Allan Collins, A Schemata-Theoretic View of Reading (Urbana: Center for the Study of Reading, 1977).

9 For an introduction to Freire's theories and their implications, see Jonathan Kozol, "A New Look at the Literacy Campaign in Cuba," Harvard Educational Review, 48 (1978), 341-377; and Kyle Fiore and Nan Elsasser, 'Strangers No More': A Libertory Literacy Curriculum," College English, 44 (1982), 115-128. Freire's three major books are Pedagogy of the Oppressed (New York, Seabury Press, 1970), Cultural Action for Freedom, Harvard Educational Review Monograph Series, No. 1 (Cambridge: Harvard Educational Review, 1970), and Pedagogy in Process (New York: Seabury Press, 1978).

10 (Baltimore: Johns Hopkins Univ. Press, 1980).

11 Is There a Text in the Class? The Authority of Interpretive Communities (Cambridge: Harvard Univ. Press, 1980), especially Chapter 11, which also appears in Critical Inquiry, 4 (1978), 625-644.

12 (English 692), "Poem Opening: An Invitation to Transactive Criticism," College English, 40 (1978), 2-16.

13 Subjective Criticism (Baltimore: Johns Hopkins Univ. Press, 1978) and "The Identity of Pedagogy and Research in the Study of Response to Literature," College English, 42 (1980), 350-366.

14 The Pleasure of the Text, trans. Richard Miller (New York: Hill and Wang, 1975), p. 16.

15 The phrase "sous rature" comes Of Grammatology, trans. Gayatri Chakravorty Spivak (Baltimore: Johns Hopkins Univ. Press, 1974). The most accessible of Derrida's books is Writing and Difference, trans. Alan Bass (Chicago: Univ. of Chicago Press, 1978).

16 The Anxiety of Influence (New York. Oxford Univ. Press, 1973). See also "The Breaking of Form," in Deconstruction and Criticism (New York: Seabury Press, 1979), pp. 1-37.

17 Allegories of Reading: Figural Language in Rousseau, Nietzche, Rilke and Proust (New Haven: Yale Univ. Press, 1979).

18 "The Figure in the Carpet," Poetics Today, 1 (1980), 107-118. For a more detailed explanation of Miller's theory, see Fiction and Repetition: Seven English Novels (Cambridge: Harvard Univ. Press, 1982).

19 The Fate of Reading (Chicago: Univ. of Chicago Press, 1975), pp. 3-19; "A Touching Compulsion: Wordsworth and the Problem of Literary Representation," Georgia Review, 31 (1977), 345-361; "Preface," Deconstruction and Criticism (New York: Seabury Press, 1979), pp. vii-ix.

20 Lane Cooper, "Introduction," The Rhetoric of Aristotle, trans. Lane Cooper (Englewood Cliffs, N. J.: Prentice-Hall, Inc., 1932), p. xx.

21 The Rhetoric of Aristotle, trans. Lane Cooper (Englewood Cliffs, N. J.: Prentice Hall, 1932), pp. 1-4.

22 Ong, pp. 60-62.

23 "Authors, Speakers, Readers, and Mock Readers," College English, 11 (1950), 265-269.

24 The Rhetoric of Fiction (Chicago: Univ. of Chicago Press, 1961), pp. 71-73.

25 "The Structure of Narrative Transmission," in Style and Structure in Literature: Essays in the New Stylistics, ed. Roger Fowler (Ithaca: Cornell Univ. Press, 1975), pp. 213-257.

26 The Fate of Reading, pp. 20-40.

27 For a different look at "CBS" style as play, see Richard A. Lanham, Style: An Anti-Textbook (New Haven: Yale Univ. Press, 1974).

28 "Concluding Statement: Linguistics and Poetics," in Style and Language, ed. Thomas A. Sebeok (Cambridge: MIT Press, 1960), pp. 350-378.

29 "Introduction: Some Remarks on Pedagogy," in Contemporary Rhetoric. A Conceptual Background with Readings, ed. W. Ross Winterowd (New York: Harcourt Brace Jovanovich, 1975), pp. 1-37.

30 John L. Austin, How To Do Things with Words (New York: Oxford Univ. Press, 1962). John R, Searle, Speech Acts (London: Cambridge Univ. Press, 1968).