In the summer of 1987, the three of us were appointed to teach Rhetoric and Composition to summer provisional students at The University of Texas at Austin. Initially developed as the University's attempt to provide higher education opportunities to Texans historically unable or underprepared to attend college, the provisional program became a crash course in survival at a large research institution. Consequently, students in this program were expected to undertake a significantly complex core curriculum (including math and English) and achieve higher overall GPAs in a shorter amount of time and under more adverse conditions than most entering fall semester students.

The idea to teach composition collaboratively using computers was born of necessity and efficacy. We were all assigned to teach in the Computer Research Laboratory and turned to one another as resources, to pool our years of teaching expertise and devise syllabi that initiated rather than disintegrated our students' sense of themselves as writers. After a few informal meetings, we not only discovered that we each dreaded the lonely prospect of teaching yet another labor-intensive, isolated summer composition course, but we also shared similar ideas about using computers to teach writing.

What resulted was a collaborative teaching effort, from the inception of the course through the daily use of computers to the grading of the last research project. In retrospect, we found the networked computer classroom an excellent environment for collaborative teaching and for teaching a syllabus based on student collaboration. We were able to expand the number of voices and opinions available to our students for collaborative activities because the networked computers and instructional software in the Computer Research Laboratory (CRL) extended the classroom, allowing separate classes to communicate and collaborate across space and time. The results of collaborative teaching in this setting, though not always as far-reaching as we would have liked, were promising. Although the students often struggled with notions of "socially constructed meaning," most of them valued the opportunity to discuss (in conventional modes and through the computer network) papers and ideas with one another and three instructors.

Having already taught individual writing courses in the CRL, each of us was familiar with the collaborative potential of networked computers combined with text-sharing software, and we believed that electronically linking three classes would provide superior technological assistance for modeling the academic community. Our sections met Monday through Saturday for five weeks at different times in CRL: Peterson's section met from 8:30 to 10:00 a.m.; Halasek's from 10:00 to 11:30 a.m.; and Balester's from 11:30 a.m. to 1:00 p.m. The classroom was equipped with 24 IBM PCs connected by local area network (LAN). The software used was the prototype of the Daedalus Instructional System (DIS)--in particular, electronic mail (MAIL, presently known as Contact) and synchronous conferencing (FORUM, presently known as InterChange), supplemented by the McGraw-Hill College Version of WORDPERFECT (1.0). The hardware/software configuration was far superior to the photocopying we relied on for critiquing in our traditional collaborative learning classes because it provided students with "publishing" capability, ideal in a collaborative learning environment that emphasized peer critique. Any text file could be viewed by network users in any and all of our classes. Students could work on their papers up to the last minute before giving them over to peer critics, and the number of critics was not limited to the number of available photocopies of a draft. A draft could also be printed at a moment's notice, so students did not have to bring photocopies to class.

DIS software provided two communications programs that made collaborative teaching possible while also enhancing the collaborative learning environment. The first of these, synchronous conferencing or FORUM, allowed our classes to hold group "discussions" that were captured in computer files as electronic transcripts. (Unfortunately, FORUM was in an early stage of development that limited its flexibility; consequently, early transcripts were preserved only in hard copy.) This sort of communication took place between the teachers and students and among students assigned to a particular section who met at a given hour--we call this intrasection communication. Such sessions could also be shared asynchronously or sequentially among sections in electronic transcript form, so students in Peterson's section could leave a transcript file of its FORUM discussion for Halasek's and Balester's sections to read, which allowed each section to continue the discussion during its class period and leave its own transcript files. FORUM provided a tool that enhanced traditional classroom discussions; the discussions in our classes ran longer and in greater depth because we could edit transcripts for interesting points and bring them to later classes for further elaboration. Furthermore, FORUM, because of its real-time editing features, encouraged all students to participate in discussions simultaneously without having to wait for the openings or "breaks" that traditional conversation conventions often require.

The second communications option, the e-mail program (MAIL), allowed our students to send more specific messages to each other and to each of the three instructors across sections, regardless of the time the classes met--we call this intersection communication. In other words, Peterson's section could leave messages for Halasek's students (or for Halasek, herself), who would arrive at 10:00, or for Balester's students (or for Balester), scheduled to come in at 11:30. We could still "converse" without trying to negotiate personal schedules in order to meet face-to-face. Intersection communities replaced the more traditional single classroom community of peers, thus extending our classes spatially and temporally. Because students gained access to three classes and three instructors instead of one, MAIL provided us with a tool not available in a single classroom setting. We also used MAIL for private teacher-to-teacher, teacher-to-student, and student-to-teacher communication to address pedagogical problems (e.g., how to introduce an assignment), respond to and evaluate final drafts, answer questions regarding assignments, and distribute grades.

In designing our sections so they were identical in organization, content, instruction, and evaluation; integrated the computers in every facet of classroom activity; and facilitated student writing and thinking through the collaborative learning pedagogy, we tailored Kenneth Bruffee's (1985) A Short Course in Writing to suit our particular needs. During the five-week session, students completed four major writing assignments. Our assignment sequence--an ungraded "Reminiscence," "Family Story," or "Accounting for Change" essay; a book review; a "Context of Issues" report; and a "Judgmental Synthesis" research project--did not make up the entire sequence of the Bruffee program but did represent a large share of it.

In the first assignment, students composed a personal experience narrative in which they wrote the "true story of something that happened to you once that deeply moved you, upset you, frightened you, or made you angry," told a " 'family story'. . . that somebody tells when your family gets together for holiday dinners, parties, picnics, or other celebrations," or wrote a "story of some aspect of your life that has changed noticeably during the past six months to a year" (Bruffee 1985, pp. 31-32). This was followed by a book review in which students described the subject of the book they had read, evaluated the book's worth, and supported that evaluation. The third assignment, the context of issues essay, required students to synthesize the information they gathered from the book's reviewers by reporting on the issues surrounding the subject of the book. Students were given the option of reporting several issues or selecting a single issue and describing how it was addressed by the reviewers and the author of the book.

The assignments culminated in the Judgmental Synthesis essay, what our students and we came to call the "Bruffee research paper." In this final essay, students addressed an issue raised in the book they had read, integrating from the earlier essays ideas, information, and opinions while also including the opinions of recognized scholars or experts whose work they had found through library research and interviews. The result was a researched paper that highlighted the student's stance as it confirmed or opposed published thinking on the selected issue.

The sequenced assignments and collaborative pedagogy, supported by the networking technology, provided the means by which we could impress upon students their own authority and expertise on issues such as self-healing, illiteracy, poverty, or war. MAIL promoted collaboration by allowing students to leave e-mail messages responding to drafts of propositions (thesis statements), essay drafts, and inquiries on topics of discussion. FORUM provided an engaging and lively arena for collaboration by allowing students to exchange ideas, resolve the meaning of difficult passages, or create propositions--all with immediate feedback from several colleagues.

By collaboratively teaching our writing courses, we were able to lead small intersection group discussions through MAIL or on FORUM with students on the books they had chosen to read. We saw this as a way to solve a central problem with the Bruffee research paper, which required that students read a book selected from a list of 42 titles. For example, a student reading Norman Cousins's Anatomy of an Illness might research the opinions held by prominent physicians and medical researchers about the ethics of administering placebos to chronically or terminally ill patients. A student reading Jonathan Kozol's Illiterate America might report on the variety of measures taken by other countries to combat illiteracy. How, we wondered, could we really engage students in discussions of these books without each of us having to read all 42 titles?

In our traditional classes, we did not have the luxury of monitoring and helping students with the quality of the overall discussions they might have about the books they chose primarily because there were too many students and too many book choices and only one teacher. Consequently, each student chose a book from the list and worked on it in isolation. Typically, we skimmed the book to be sure the student was reporting on it with some degree of accuracy, but we really were not able to model for our students the ways in which novices become experts on a subject because, depending on the book a student chose, we were often virtual novices ourselves.

Collaboratively linking our classes in a networked computer classroom helped us address this dilemma. Because FORUM and MAIL provided an impetus for the students in our classes to contribute to meaningful discussions and review such contributions from an electronic transcript, we devised more challenging methods of stimulating important, engaging conversations, aided invention, and encouraged students to explore with others the issues on which they were writing. We compiled our own reading list, narrowing the choices to six texts, created six small communities of peers, each working with the same primary text (see Figure 1). We divided the list among ourselves, each reading two texts and ensuring that an approximately equal number of students from each class read one book on the list.

Each of the six groups was comprised of (1) a "local expert" on the text (the instructor who had read it and who became a "participant-observer" within the intra- and intersection electronic mail and discussion groups); (2) two to five students within each section; and (3) eight to twelve students across sections. There were, for example ten students in the three classes reading Illiterate America: five from Balester's class, three from Halasek's, and two from Peterson's. Students no longer worked in isolation or relied on the teacher as the sole determiner of "correctness" or "truth." Electronic conversations included reacting to the author's style, discussing positions on issues raised by the texts, questioning one another on interpretations of difficult passages, or responding to one another's work in progress. By placing students in reading groups, we formed committees of peers undertaking similar tasks with shared goals and discourse conventions. Because these groups demanded that students put forth ideas and often defend them, students were encouraged to begin recognizing their own individual authority, as well as the growing authority of the group. But with individual authority came responsibility. Students learned that authority and respect required commitment. And as their own levels of authority increased, they were introduced--through texts and the insights of "local expert" instructors--to the larger academic community. We also introduced them to the ways informed groups of people become experts in their respective fields and how those experts negotiate and legitimate the knowledge structures they claim to value. As they continued to work, we explained to our students that this is how academic disciplines come to define themselves.

One surviving FORUM session illustrates how students used the messaging system to negotiate meaning, develop a sense of community, and collaborate to create propositions. The exchange, recreated below, took place among four students in a single class who had read Jonathan Kozol's Illiterate America. The students were asked to discuss the main issue of the book and work toward composing propositions for their book reviews. Ellipses represent three excised entries that focused on Kozol's use of statistics in early chapters of the book.

(1)	Renee Warren: Karen, I sent you a message in the mail system asking you if you wanted to talk about how Kozol is bias [sic] in the book. He also seems to be somewhat of a hyprocrite [sic]. What do you think?
(2)	Karen Dusang: I think that I'm confused as to why Kozol is so angry. I think that his main point is to make people aware of illiteracy in America. The only problem I have with this is his strong alligations [sic] against the government. I'm not saying that I don't agree with him but there is something about his arguement [sic] that really disturbs me. Can you help me?
(3)	Ashley Jackson: I think that the book was very interesting but boring. To me Kozol's main ideas was that illiteracy is more of a problem in the world than anyone realizes and that the government should do more to prevent this problem by setting up programs to help the illiterate. (Or to help people while they are young to read before they become illiterate.)
(4)	Hung Vu: I think that basically what Kozol is saying is that illiteracy in America is due to the educational system and the lack of support from the government.
(5) Reply to (3)	Karen Dusang: What is it about Kozol's ideas that bothers us so much? I agree with Ashley that the main idea was that illiteracy is more of a problem than we realize but it's all of the hostility that he shows that bothers me.
(6) Reply to (2)	Hung Vu: I think all of the hostility that he shows is due to his personal experiences
(7) Reply to (1)	Karen Dusang: Renee, I would like to work with you on this. We should go to the library today. What kind of proposition are you guys looking at writing? I feel like I might get into the issues more than just discuss them.
(8) Reply to (2)	Renee Warren: I think he is so hostile because he has worked for more than 15 years on trying to solve illiteracy but the main thing holding him back is lack of funds. Therefore, he takes this out on the governmen [sic]. Everybody feels they must shake the finger at someone. I do not agree with Kozol on the fact that the government must pay for this. . . .
(9) Reply to (7)	Ashley Jackson: Karen, I have no idea what my proposition is going to be. The whole book totally confuses me and I really think it is going to be tough writing a review on this book. I know I am going to talk about how the government should play more of a role in this issue, but I really have not thought about it that much. I would also like to meet ya'll at the library, maybe we could all get together and discuss the book overall to make it easier on all of us?
(10) Reply to (9)	Karen Dusang: I was going to come in here today at about 1:30 to start my rough draft and then to to the library afterwards. I have to get a new review.
(11) Reply to (7 &9)	Renee Warren: I think that is a good idea about getting together to discuss the book. I don't know what my proposition is going to be. Maybe I'll do the opposite of Ashley's.
(12) Reply to (7)	Karen Dusang: think I'll write on the government angle. I was really intrigued by his statement that if all of the illiterates could read than [sic] President Reagan wouldn't be president.
(13) Reply to (12)	Renee Warren: I think that might be true, Karen. Although I don't think Mondale would have won if the illiterates could read.

The first four entries represented each student's initial contribution to the FORUM session and were probably typed simultaneously. Not until the fifth entry, in which Karen agrees with Ashley's evaluation of Kozol's main ideas, did the "conversation" begin. From this point, every entry responded to a previous comment. Hung Vu and Renee addressed the issue of hostility; Ashley, Karen, and Renee offered tentative topics for propositions; Karen, Renee, and Ashley made plans for the group to meet later to continue the discussion. Hung Vu, who made the fewest number of entries (two), was marginalized in this conversation, as was often the case. Whether this was due to his tentative command of English, gender, or nationality difference was not clear, but his relatively few comments illustrated that in electronic conversations--as in oral group discussions--some students remain less active participants.

This FORUM session, which lasted from 15 minutes to 20 minutes, was typical of most intrasectional communication. Students did not generate as much text as they would have had they conversed orally, but they remained focused on the task and used the electronic medium to establish and time and place to continue the discussion. More important, they had a transcript that could be printed and taken to that meeting to aid their discussion of Kozol's purpose and their own propositions. Later classes commented on their FORUM session (it was made available to subsequent classes in a read-only mode during their FORUM sessions), which provided this group further commentary on their reading of Illiterate America and their propositions.

In defining the goals of college-level writing courses, scholars have argued that among an instructor's primary responsibilities are introducing and acquainting students with the discourse communities of academia. The instructor (a representative of the academic discourse community) instructs students (novices who must learn to write as members of that community) by teaching them how decisions about thinking and writing are made within academic disciplines (Bartholomae 1985; Bruffee 1985; Wiener 1985). The computers allowed us to "create social structures in which students can learn to take over the authority for learning as they gain the ability and confidence to do so" (Bruffee 1985, pp. 12-13) because they were conducive to student writing and increased interchange among students apart from direct instructor supervision. In traditional process-oriented writing classrooms, social structures have consisted primarily of peer groups working together with the guidance of a single authority from the academic community they wished to join, a practice that troubled us with its implicit assumption that one person, the classroom instructor, could represent the academic discourse community. Collaborative learning theory and our own experience told us that knowledge emerges from struggle and discussion among like-minded peers. We sought to model the formation of ideas and policy in the academy by presenting students with a trio of instructors working actively to make sense of texts, assign grades, and guide novices in joining them.

Our students knew and became familiar with all three instructors because we divided the reading list and explained to our students that different instructors were responsible for different texts. We also agreed to act as teaching assistants in each other's classes to orient the students to the computers and the software and to hold office hours for students from all sections, making us physically and electronically available to one another's students. This arrangement probably increased the number of hours we spent in class working with students, but the experimental nature of the course and the collaborative teaching made it difficult to determine whether the increase was a result of the pedagogy, computers, or teaching methods and to what degree.

Being present in one another's classes to help with technical questions also led, without foreknowledge, to open discussions between instructors about writing. For example, Balester might turn to Halasek, the "visiting" teaching assistant, to clarify or expand on a point, or Halasek might offer additional examples to illustrate Balester's lesson. By having three experts available, whether physically or electronically, students were able to get alternative points of view on their theses, peer critiques, and research techniques. They were able, in other words, to get second and third opinions, which they weighed and evaluated for themselves and with their peers. We discovered that we were literally illustrating the ways in which decisions about writing are made because we could never be counted upon to agree. We might, in the course of MAIL or FORUM discussions, openly debate a stylistic point or the reading of a text. We were also prone to live disagreements. On one occasion, all three of us were working with students in Balester's classroom on Bruffee's book review assignment when it became quite clear that we disagreed about the function and purpose of the writing task. For us, reaching consensus was not necessarily paramount; showing our students that disagreements can and will occur among even like-minded members of an academic community was. It was equally important for them to see that we could also reach agreement, validating for them the fact that, despite what students sometimes believe, college English teachers do indeed represent a group with some shared conventions about their discipline.

We were concerned, nevertheless, that three instructors would simply triple the instructor's authority. The aim in our classes was the opposite. We sought to diffuse authority, to share it with one another and with our students. We negotiated the "proper" way to write or carry out an assignment and discussed the issues raised in our readings, hoping students would be less inclined to invest us with ultimate power over their words. Other team-teaching programs have cited similar goals and advantages to their collaborative teaching efforts: authority was divided, student-instructor ratio was reduced, and the faculty instructor was no longer isolated in his or her classroom (Held and Rosenberg, 1983).

The final undertaking of collaboratively teaching the course was grading the essays as a group of expert readers. The three formal writing assignments based on the reading groups--the book review, the context of issues, and the judgmental synthesis--were graded by the instructor of record and the reading group local expert. If these two instructors varied significantly on their evaluations of an essay, the third read and critiqued the piece. In every case, however, the instructor of record assigned her students' grades and relayed them to students in private MAIL messages. Essays were graded both holistically and atomistically--that is, we read first to evaluate the overall effectiveness of the essay and later to point out and address problem areas. We judged an essay on development, unity, coherence, style, and mechanics, just as Bruffee (1985) suggested peers evaluate each other. His method required that some class time be spent reaching consensus about the definitions of each of these criteria, and we made an effort to follow, generally, the same guidelines we agreed upon with our students. All in all, our evaluations served not only as measurement tools but also as models of peer responses.

As the summer session progressed, students became increasingly aware of our strong commitment to the collaborative nature of the three courses and to each other as teachers and colleagues, and they learned to see us as experts with shared attitudes and values about writing instruction and evaluation methods. As one student reported four years later,

I remember all your faces and I remember almost everything I wrote about in that class that summer, but I just can't remember which one of you was my actual teacher. I think I just went to whoever [sic] was available. It just seemed clear that if one of you told me something, the other two would probably have agreed--or at least if you all disagreed, you'd explain why you maybe saw it differently. Sometimes I understood and sometimes I didn't, but I never felt like one of you was more bossy or in control than the other. I always figured you all pretty much knew what you were talking about, and that helped me to relax a little and not worry so much about who was in charge.

We don't want to give the impression that these classes ran perfectly, that all of our students rushed to contribute to FORUM sessions and then miraculously wrote (with the assistance of their peers) flawless and effective prose. They performed at the same level as most new college writers at The University of Texas. Regretfully, we concluded that we had not completely succeeded in sharing authority with our students. Fresh from high school, they were not altogether willing to take responsibility for their learning. Most had never been asked to before, and it was a strange and foreign request. To be fair, we were not entirely willing to turn them loose on each other, and the realities of instructors evaluating students' work invariably highlight the power differential in any classroom. Yet, the end-of-term student evaluations of the course suggested that many had learned an important lesson from seeing us share authority: More than one authoritative opinion about a piece of writing could be had and was, in fact, useful.

Student responses to a class evaluation questionnaire provide our best evidence of the results of our collaborative teaching. To our pleasure, some students, when asked what aided them most in their writing, cited the group grading and evaluation. As one put it, "I don't think I would have survived without all three of the teachers to comment on my papers." When specifically asked, "Was it helpful having three teachers available for comment and instruction? Why or why not?" most students, 48 out of 55 (87%), answered "Yes." For most of the students, the greatest benefit was having three sources of opinion, ideas, and explanation available. Some of them spoke of the general availability of help, while others specified the value of having more than one authoritative reader of a draft. Many also cited the convenience of greater accessibility to a teacher--some even mentioned faster feedback--as a positive result of collaborative teaching. Having three sources of (sometimes conflicting) opinion forced students to evaluate several positions and decide for themselves what they would do, as Held and Rosenberg (1983) recorded.

Perhaps the most interesting comments on collaborative teaching came from Halasek's class concerning a public debate of Bruffee's book review assignment. As one student wrote, "The teachers need to agree on what is to be done before the class starts, not arguing in the middle of class." She is, of course, correct to some extent, but she has missed the broader point--that open disagreement, while sometimes confusing, can lead to greater understanding. Another student's response illustrated this:

I felt it was helpful having three teachers although there are times when the teachers argue; for instance, when [they] disagree over the proper format of Bruffee when doing the book review. At times it can confuse the student[s] having so many different views[,] yet at the same time it can help broaden their literary skills.

In retrospect, the three of us might have preferred to hash out our differences over the book review in the confines of our office because the students were confused by our various opinions. But for cases other than assignments (where new university students understandably need a great deal of certainty), disagreement and negotiation was helpful, primarily because such public debates allowed students to see that opinions are informed by belief systems born of social interaction--initially within our family systems and later with the world at large. As one student wrote:

Some students found the collaborative grading troublesome: . . . I didn't like the fact . . . [that] another teacher was going to grade a paper that I had written under some one [sic] else's instructions. . . . I felt that it wasn't fair that another teacher was grading my paper because what if one teacher explained the assignment differently from another teacher? Then when the student writes the paper the way one teacher explained it then the teacher grading the paper may have explained it another way . . . so the papers may be written differently according to the teacher.

This student's frustration came from her discovery that instructors may disagree over how a paper is to be written, that they cannot provide the security of ultimate authority in any real sense, and that writing is not an exact science that can be broken down into sets and subsets of skills and transmitted uniformly from teacher to student. This student was learning the lesson of audience awareness--understanding whom she is writing to and why, and what that reader's expectations are.

Follow-up interviews with eight of our former students (all academic seniors) at The University of Texas at Austin during the fall semester of 1990 provide a compelling postscript to our collaborative experience in the CRL. They remembered our courses as the ones with three teachers in the computer classroom, and most of their recollections were positive as they described the general sense of community and support our courses fostered. But these interviews revealed some disturbing realities about the plight of writing instruction at large state universities: The question was asked, "In retrospect, do you think your E306 Rhetoric & Composition course helped you in future writing courses at the University?" Five out of the eight students responsed that they never took another writing course, an unsettling fact of undergraduate education. As they explained, they purposely avoided "real" writing courses because they were afraid these would be too difficult or would lower their GPAs.

Our students' comments showed that we needed to prepare them for collaborative teaching, even more than for collaborative learning. In most cases, teaching has been undertaken in an educational climate that privileges a strong central authority, what Tompkins (1990) refers to as the "performance model" (p. 654). Tompkins has critiqued the commonly held belief that teachers must carry most of the teaching and learning responsibilities while students passively, and sometimes hostilely, dare teachers to engage them in any significant or challenging way. Tompkins explained that this teaching model, which especially reproduces itself within the academy, was based primarily on fear: fear of exposure, fear of not knowing the answer to every question students might ask, fear of teaching itself (1990, p. 654). In contrast, more interactive teaching methods allow teachers to model more active, critical learning environments inside their classrooms in hopes that their students will become more active, critical people outside their classrooms. In our opinion, collaborative teaching contributed to the active learning practices of our students.

In the end, this course was successful for us as instructors. We tested our instincts and found beyond any doubt that the networked computer classroom--combined with flexible software and an appropriate composition pedagogy--offers an atmosphere conducive to collaborative teaching. We took traditional team-teaching to some important limits by involving one another in every aspect of our teaching, class preparation, and evaluation. We also extended the concepts of peer response and collaborative learning by opening every student text to the critical eyes of more than 60 fellow writers.

We believe that teaching, like writing, is a collaborative, social act. Like the writer whom we have for so long envisioned as a "solitary writer alone in a cold garret working into the small hours of the morning by the thin light of a candle" (Brodkey 1987, p. 396), so, too, are pictures conjured up of the instructor: hair frazzled, eyes weak from the strain of grading one essay after another in a windowless office. Because this image exists, like Brodkey's image of the writer who writes alone, it distorts for both teachers and students what instructors do and know and how they acquire and establish their knowledge, methodology, standards, and criteria.

Most of us recognize that writing instructors collaborate regularly by sharing lesson plans, classroom practices, and evaluation techniques in hallways, at colloquiums, and during professional meetings, and the institutionalization of collaborative learning will find instructors more "inexorably involved with each other's teaching in new and sometimes uncomfortable ways" (Kail 1983, p. 594). But it is the computer--on local area networks and even on wide area networks such as BITNET, or bulletin boards such as Megabyte University (organized by Fred Kemp of Texas Tech University)--through which collaborative teaching can be made infinitely richer, for it allows us direct access to one another's students and extends our professional associations.

Valerie Balester, Kay Halasek, and Nancy Peterson teach in the English Departments at Texas A & M, the Ohio State University, and the University of Texas at Austin.

Bartholomae, D. (1985). Inventing the University. In M. Rose (Ed.), When a writer can't write: Studies in writer's block and other composing problems (p. 134-154). New York: Guilford Press.

Brodkey, L. (1987). Modernism and the scene(s) of writing. College English, 49, 396-418.

Bruffee, K. (1985). A short course in writing: Practical rhetoric for teaching composition through collaborative learning. (3rd ed.). Boston: Little, Brown and Company.

Held, G., & Rosenberg, W. (1983). Student-faculty collaboration in teaching college writing. College English, 45, 817-823.

Kail, H. (1983). Collaborative learning in context: The problem with peer tutoring. College English, 45, 594-599.

Tompkins, J. (1990). Pedagogy of the distressed. College English, 52, 653-660.

Wiener, H. (1985). Collaborative learning in the classroom: A guide to evaluation. College English, 48, 52-61.