In a recent Shoe cartoon, the "perfesser" attempts to use word processing for revision. The first frame shows Muffy bringing the professor a bottle of white-out: "Thanks, Muffy. I hope this stuff works better than the last batch ...." The second frame shows the professor painting the white-out all over the computer screen: "I don't think he'll ever adjust to the word processor," laments Muffy with a forlorn expression.

Computers are a part of our lives as writers and are increasingly more common in English departments and writing classrooms. But, like MacNelly's professor, we don't always know how to use computers to our best advantage. Which students can benefit from using computers for writing and which educational methods should be used to teach word processing? How can students be taught to adapt their writing habits to take advantage of the possibilities offered by the new technology? These are typical of the questions being asked in the profession.

p. 30

Certainly unleashing students in the computer lab, with or without elaborate text editors or style-analysis programs, does little to improve the quality of their writing, as a body of research in word processing and writing instruction is beginning to reveal. No studies to date have shown an improvement in writing quality by students using computers as compared to those not using computers (Clark, 1985; Collier, 1983; Kiefer & Smith, 1983; Pfaffenberger, in press). Furthermore, Collier hypothesized that word processing would encourage revision in student writing, but he was not able to confirm that hypothesis in his research (1983). Harris' study (1985) suggests that word processing does not, by itself, encourage revision, and Hult's study (1985) shows that word processing does not make student writing more correct. In the book Writing On-Line, several of the contributors caution teachers not to expect computers to effect changes in their students' writing habits and procedures (Selfe, 1985; N. Sommers, 1985). Other authors cited in the bibliography point to similar concerns.

I wish to suggest that using computers in writing classes, in the absence of appropriate instruction, may even reinforce the unproductive composing strategies characteristic of inexperienced writers. Like the professor, our students do not always make the transition from writing on paper to writing on screen very well; they too often simply transfer current writing habits wholesale. I will focus my discussion on revision strategies, since it has been hypothesized that word processing encourages revising; even if word processing does encourage revising (which is still questionable), what kind of revising does it encourage?

Elizabeth Sommers' seminal article "Revision Strategies of Student Writers and Experienced Adult Writers" (1980) identified two main ways in which the revision strategies of student and experienced writers differ: (1) student writers saw their compositions in discrete parts and considered revision to be a rewording activity; experienced writers saw their compositions as a complete unit and considered revision to be a communication activity, and (2) student writers viewed their texts as the embodiment of predefined meaning; experienced writers used writing and rewriting to discover meaning. I would like to explore more closely how word-processing and text analysis programs may potentially inhibit the very revision strategies we attempt to teach our inexperienced student writers.

p. 31

Elizabeth Sommers pointed out that students "understand the revision process as a rewording activity. They do so because they perceive words as the unit of written discourse" (p. 381). In contrast, experienced writers saw revising as a way of finding shape for their arguments; they saw their writing as a whole, taking on a reader's perspective and attending to communication in the broadest sense.

The second and related major difference between the two groups studied by Elizabeth Sommers centered on predefined meaning vs. discovery of meaning. While the student writers felt there was a predefined meaning that they need only find the right words to express, the experienced writers sought to discover or create meaning through the act of composing, and particularly through revising. The students' inordinate preoccupation with repetition of words or phrases, which they listed as something they worried about most, illustrates their perspective: eliminating repetition involves lexical (wording) rather than semantic (meaning) changes.

The only evidence that the students modified ideas while revising came when they tried different introductory paragraphs. They stopped revising when they felt that they had corrected any "rule violations," such as "never begin a sentence with a conjunction" or "never end a sentence with a preposition." Any changes made were changes to accommodate such sets of rules. The students failed to use reordering and addition when revising, but rather concentrated primarily on substitution and somewhat on deletion.

On the other hand, experienced writers defrayed concern about vocabulary and style to the end of the writing process. They made changes on all discourse levels and used all revision operations because they saw their composition both as a whole and as a way of discovering meaning. The predominant revision operations used by the experienced writers, in contrast to the students, were addition and deletion. The experienced writers stopped revising when they felt they had met their communication objectives and had come closer to an understanding of their own meaning. Through successive cycles of revision, experienced writers first focused their attention primarily on finding form for their argument, then on matters of expression and style.

p. 32

How might using computers for writing contribute to students' ineffectual revising processes as described above? Perhaps the already poor strategies of inexperienced writers may be reinforced by computers unless teachers consciously work to integrate the teaching of word processing with the teaching of the writing process. For example, the predilection to see text as parts (words) rather than as a whole (communication) can be reinforced by writing with computers. Only a small amount of text fits on the screen, and the entire text is relatively inaccessible until a printout is made. Thus reading the text on a screen may reinforce an inexperienced writer's parts approach to revising. In addition, active reading (so essential for revising) may be hampered by the small screen and the relative inaccessibility of the entire text. In fact, Haas and Hayes point to reading problems observed when students use computers for writing (1986). Many writers who use word processing have learned to compensate for their difficulties reading on-screen by relying on frequent printouts. We need to encourage our students to adjust their own writing processes to the changed writing environment.

The predilection of inexperienced writers to substitute and delete rather than add and rearrange may also be reinforced by word processors. For instance, the rewording functions on most word processors, allowing the writer to substitute and delete, are simple to use and easy to learn. Thus some students spend most of their computer writing time backing up with the backspace key, constantly erasing and rewording rather than getting on with composing. In contrast, the reordering function of word-processing programs is usually more complicated than some of the other editing functions. Teachers need to look for word-processing programs that make the cut and paste functions--which allow rearranging of text--easy to use. But teachers also must realize that, no matter how sophisticated the capabilities of the word-processing program, students must be taught to see why text reorganization is important and how best to accomplish it, given their own rhetorical purposes.

If we are not careful, inexperienced revising strategies may be heightened when students use text-analysis programs, such as HOMER, HBJ WRITER, GRAMMATIK, OR WRITER'S WORKBENCH. Analysis programs concentrate on words and rewording (for example, vague

p. 33

words or sexist language), thus potentially reinforcing the inexperienced writer's emphasis on words rather than whole text. Analysis programs often point out repetition of words and phrases-- something students already worry about unnecessarily. Lexical, rather than semantic, changes are encouraged by analysis programs; thus, a student who is told to choose another word for the vague word most may change it to an equally vague many. The inexperienced writer's concern about rule violations may be reinforced by spelling checkers and analysis programs since these programs typically point out usage rules (distinguishing between who or whom and that or which, for example). Furthermore, these programs may be misleading because they identify such a limited set of errors, yet students too often feel that their "grammar" has been "checked" by the computer. Lastly, students using analysis programs are often encouraged to bring finished drafts to the computer lab and type their drafts into the computer for analysis. This procedure, if not preceded by extensive revision for content, may reinforce the misconception that meaning is already defined and the text just needs to be cleaned up.

The following excerpt from a student's text, in draft and revised form, illustrates the concerns I have outlined:

 	There is an aspect of college life that is rarely discussed, yet 

		of greater      ce    
often considered -more- importan-t- than the actual college one will

				     eatible? edible?		good?
be attending.  For most, it even surpasses the concern of how hood

the food will be in the dorm cafeteria.  What could this vital life or 

death issue be?  FINDING A ROOMMATE!  Inevitably one's 
								 in
roommate will  become  the  center  and  most  important  part  -of- 
							     throughout
your college career.  For without one there would be no one to 

accompany you on those midnight snack raids (catered courtesy 

of Pizza Express), no one to engage in those necessary gossip

p. 34

sessions concerning who saw who with who and where, and no 

				     that annula
one with to plan that Spring Break(journey to Padre.)

     There is an aspect of college life rarely discussed, yet often 

considered of greater importance than the actual college one will 

attend.  For many, it even surpasses the concern of how edible the

food is in the dorm cafeteria.  What could this vital life or death

issue be?  FINDING A ROOMMATE! Inevitably your roommate 

will become the center of importance throughout your college 

career.  For without one there would be no one to accompany you 

on those midnight snack raids (catered courtesy of Pizza Ex-

press), no one to engage in those necessary gossip sessions 

(concerning who saw who with who and where), and no one with 

to plan Spring Break (that annual journey to Padre).

From this excerpt, you can clearly see the ineffectual revising strategies outlined by Elizabeth Sommers. The student sees revising as a rewording activity, never considering the communicative effectiveness of the whole text. Though the student used word processing (DEC Rainbow/SELECT) to produce the text and a text-editing program (GRAMMATIK) to analyze it, the changes evidence little real revision. The word-processing and style-analysis programs may even have reinforced her misapprehensions. Notice that the student has changed only those words flagged by the analysis program as vague or possibly incorrect ("most, very, good, and nice;" prepositions, and that/which).

Some of the changes are improvements, others just make matters worse (e.g., changing "more important" into the prepositional phrase" of "greater importance" when the analysis program had already suggested to the student that she used too many prepositional phrases throughout her entire paper). Although I have only

p. 35

included the introductory paragraph, throughout the entire text the student discovers little new meaning through revising, but rather sees the text as virtually complete, just needing a little cleaning up. Comparing the two versions, one is struck by how little has really changed . There has been little at tempt to engage the audience in the opening; there are no global changes--no block moves have been used to reorder parts, no substitutions other than words occur, and there are no additions whatsoever. This pattern of revision continues throughout the entire piece, which is (not surprisingly) five paragraphs long.

As Elizabeth Sommers so aptly put it, "The evidence from my research suggests that it is not that students are unwilling to revise, but rather that they do what they have been taught to do in a consistently narrow and predictable way .... The students do not have strategies for handling the whole essay" (p. 383). In the absence of appropriate instruction, computers may just exacerbate the problem. Students in writing classes often find word-processing instruction is largely divorced from the other classwork, as they did (unfortunately) in my own class. In computer labs, students may work independently at computers, with little or no peer or teacher interaction or encouragement to make use of the important capabilities word processing offers to writers. Too often, students do not have enough lab time available to them to actually draft at the computer. Consequently, students bring texts which they type into the computer for revision and analysis--thus reinforcing their habit of seeing first drafts as essentially finished products.

Yet those of us who use word processing to teach writing see the potential for real benefits and are unlikely to lose faith in that potential despite our mistakes and failures of the past. Besides, with or without our help, students are increasingly using word processing on their own, and the number of college word-processing microlabs continues to grow apace--doubling in 1985 alone (Barker, 1985). But if we are to use computers judiciously, we must design instructional settings and curricula that provide for an interactive computer classroom, using what Hillocks calls an "environmental teaching mode" (1984). Entire writing classes can work together to solve writing problems in collaboration with each other and with the teacher. Group problem-solving can be accomplished by projectors interfaced with computers, thus displaying text from one screen onto a large screen for everyone in class to view and discuss. Such projection set-ups allow teachers to demonstrate

p. 36

effective composition strategies in process. Also, networking systems can provide students access to each other's papers for comments and suggestions while work is in progress. Adequate facilities are essential--comfortable work stations, sufficient hardware and software, clear documentation, and immediate assistance for both technical problems and writing questions. Ideally, when working with a class in a lab, the teacher should have a lab assistant available to answer technical questions, thus freeing the teacher's time for writing instruction.

Once a congenial writing atmosphere has been established, students should first be exposed to potentially beneficial uses of computers and then be encouraged to experiment with them at every stage of the writing process, rather than simply to use computers as super-typewriters. Prewriting software and exercises may encourage reluctant writers to use the power of word processing to explore ideas, discover new ideas, make connections between these new ideas and related experiences and knowledge, and perhaps gather or retrieve relevant materials. Word processing may help students experiment and explore, through freewriting and brainstorming, for example. It can also help students gather and retrieve materials, serving as a repository of ideas and sources. Furthermore, word processing can help students plan and organize their writing in a fluid, flexible form that is easily manipulated during drafting.

Often prewriting can provide students with a rich source of ideas and information on disk from which to draw as they compose. Using word processing can facilitate a building-block approach to composition, filling in and expanding an outline stored on disk, for example, or writing from a predesigned frame. As they compose, students should obtain frequent printouts for easier reading of their drafts, and as they revise, they should gain feedback several times from peers and from the teacher. Collaboration (between peers and with the teacher) can be encouraged so that students gain experience from writing with others at the computer.

Revising, editing, and proofreading can all be facilitated by word-processing programs, provided students are instructed in effective revision strategies. Students need instruction in word processing functions that allow them to add and rearrange as well as to substitute and delete. Class time should be spent on rearranging text using block move commands. Analysis programs should

p. 37

not be run until the whole-essay concerns have been addressed through conferencing and repeated revising sessions. Once run, analysis printouts should be brought to class for discussion and for comparison among students. The appropriateness of the advice, in the rhetorical context of the piece being written, can be analyzed by the class. As did the experienced writers, students need to wait until very late in the process to attend to matters of expression and style.

Finally, students must understand the limitations of analysis programs and recognize that careful proofreading is as necessary with computer-printed papers as it is with typewritten or handwritten papers. Even when using a spelling checker, words will be missed, particularly homophones (to, too, and two) which the computer will not flag as misspellings. Furthermore, analysis programs cannot check grammar or usage in any real sense, so careful proofreading for correctness is still essential.

Although word processors cannot teach writing, I am convinced that word processing can be an important tool for writers. In order to use word processing effectively, however, students must understand the principles of effective revision and apply those principles to writing with a word processor. Writing courses that include word processing must do a great deal more than simply introduce students to the machines. Essentially, instruction in writing with word processing should encourage a process approach to composition that reinforces the difference between the substantive revision of content exemplified by experienced writers and the ineffectual rewording of finished texts too often exemplified by our students.

p. 38

Barker, Thomas T. (1985). The English department microlab: An endangered species? Research in Word Processing, 3, 2.

Clark, Richard E. (1985). Confounding in educational computing research. Journal of Educational Computing Research, 1, 137-148.

Collier, Richard M. (1983). The word processor and revision strategies. College Composition and Communication, 34, 149-155.

Haas, Christina, & Hayes, John R. (1986). What did I just say? Reading problems in writing with the machine. Research in the Teaching of English, 20, 22-35.

Harris, Jeanette. (1985). Student writers and word processing: A preliminary evaluation. College Composition and Communication, 36, 323-330.

Hillocks, George, Jr. (1984). What works in teaching composition: An analysis of experimental treatment studies. American Journal of Education, 133-170.

Hult, Christine. The effects of word processing on the correctness of student writing. Research in Word Processing Newsletter, 3, 14.

Kiefer, Kathleen E. & Smith, Charles R. (1983). Textual analysis with computers: Tests of Bell Laboratories' computer software. Research in the Teaching of English, 17, 210-214.

Pfaffenberger, Bryan. (in press). Word processing technology and composition instruction: Toward the environmental mode. Journal of Educational Computing Research.

Selfe, Cynthia L. (1985). The electronic pen: Computers and the composing process. In James L. Collins & Elizabeth A. Sommers (Eds.), Writing On-Line (pp. 55-66). Upper Montclair, NJ: Boynton/Cook.

Sommers, Elizabeth A. (1985). Integrating composing and computing. In James L. Collins & Elizabeth A. Sommers (Eds.), Writing On-Line (pp. 3-10). Upper Montclair, NJ: Boynton/Cook.

Sommers, Nancy. Revision strategies of student writers and experienced adult writers. College Composition and Communication, 31, 378-388.