I. INTRODUCTION

Composition theory, in the last four decades, has gone through many significant changes. Berlin, North, Faigley, and others have attempted to define and categorize the various "camps" resulting from the changes. In 1982, Hairston proclaimed that the shift from "current traditional" to "process-based" pedagogy amounts to a paradigm shift on the scale of those described in Thomas Kuhn's The Structure of Scientific Revolutions. Process-based theories of teaching writing were originally on the fringe, then acknowledged but marginalized, and finally took over in a radical power shift, according to Hairston. Indeed, many composition researchers have taken the leap of faith into the fray of multiple drafts and revisions.

Unfortunately, though, a large number of composition courses are taught by graduate teaching assistants or instructors trained in literary analysis who have not been exposed to the increasing mounds of research being done in this area, nor do they care to be. They continue teaching as they have been taught, in a style which Barker & Kemp, Cooper and Selfe and others disdainfully refer to as "proscenium" pedagogy or lecture-based teaching, even though Hillocks' 1984 meta-study and several smaller ones have shown this method to be one of the least effective means of "transmitting knowledge" (a concept also recently under attack).

In an effort to find other means of teaching people to write effectively, teachers such as Elbow, Murray, Emig and others began turning their classes into writing workshops where students would read each other's work aloud to each other, give each other suggestions for revision, then go back to their desks or their rooms and write. This method of teaching was based on the premise that by sharing their writing with others, students could benefit from the feedback of their classmates, gaining a more realistic sense of audience, and therefore write "better" papers--in other words, express themselves more effectively. Although this method employed collaboration, it was still based on the romantic notion of an inner self which, under the proper circumstances and in the right environment, could be inspired to greater heights and glories.

More recent social-constructionist thought, however, tends to discount the idea of this inner or "autonomous" self, even though it has historically held a strong tradition in the discipline. In this respect, an important addition to process-based writing pedagogy was Kenneth Bruffee's 1972 textbook, A Short Course in Writing, and his subsequent 1984 College English article "Collaborative Learning and the 'Conversation of Mankind.'" The underlying basis for Bruffee's theories was not the idea of a single person transmitting his or her inner thoughts and feelings to another similarly self-contained being, but, borrowing from Bakhtin and others, the concept of a socially-constructed self, and therefore, socially constructed knowledge. Other theorists have explored this notion, which suggests that all knowledge, and therefore all writing, is not merely an expression of one individual's inspired thoughts, but instead an active reconstruction of all ideas, thoughts, readings, comments, conversations, political and societal influences, and dominant ideologies that the writer has been exposed to.

This concept of socially constructed knowledge, dubbed "social epistemic rhetoric" by Berlin, "social-constructivism" by Barker and Kemp, and the "social view" by Faigley, has won acceptance among many current theorists, such Bruffee, Kemp, Selfe, and even Kinneavy, and has led to the identification of discourse communities by such theorists as Porter, LeFevre, Gere, Killingsworth, and Lyotard. For these researchers, discourse communities are defined as groups of like-minded peers that construct their own localized "truth," defined as the knowledge generally accepted by that particular group (Bruffee 646). There are positive and negative views of this socially-constructed view of knowledge. Some, such as Lyotard, Wittgenstein, and Faigley, frequently warn of its potentially negative impact on knowledge-sharing, claiming that it only helps to further fragment a society already plagued by increased specialization, while others, such as Handa, Selfe, and Hart-Davidson, welcome it as an opportunity for previously marginalized groups to enter into the academic "conversation."

At the same time as these theories of collaboration and social constructivism have become prevalent, colleges and university student populations have changed from their traditional predominantly upper-middle to upper class youth to a more culturally, ethnically, socially, economically, and ideologically diverse population; veterans programs, open admissions policies, the change from an industrial-based to an information-based economy and the resultant need for retraining, the increase in working mothers, and the emphasis on recruitment of minority students have all contributed to the increasingly diverse student body. The idea of a classroom full of a "community of writers" has evolved into a "community of diversity" in which many voices share equally in the construction of the knowledge which takes shape in the classroom.

Moreover, the technology of composition has been affected by the advent of computers in the composition classroom. At first, computers were mainly used for word-processing, which, due to the ease of cutting, pasting, and rearranging text, made the revision process less painful for students, and thus helped support process-oriented pedagogy. As Kemp, Hawisher and Selfe, and others point out, though, computers are not in themselves the agent for change; rather, they are a tool which can also be used to reinforce current-traditional, product-based instruction. In fact, much of the initial research on computers and composition focused on the effectiveness of word processors or the benefits of this or that skill and drill program for improving spelling or grammar usage.

With the addition of the local area network, though, writing instructors began experimenting with various forms of interaction between students, using real-time conferencing software such as the Daedalus Integrated Writing Environment (DIWE), as well as asynchronous e-mail or file-sharing programs. Projects such as Galludet University's Electronic Network for Interaction (ENFI) began to investigate the pedagogical implications of computer-mediated communication. Writing instructors began to share their experiences, then conduct research into the seemingly dramatic changes that could take place, given the proper theory underlying the pedagogy. Many of the instructors claimed that writing courses could now be more writing-centered, student-centered, and idea-centered, rather than teacher-centered, as most writing courses had been up until that point. With the added dimension of computer-mediated communication, many voices could be heard at once, and electronic "conversations" could focus on particular goals such as invention or revision, and be stored on disk or printed out for later referral and/or analysis.

Another benefit noticed by many in the computers and writing community was that previously unresponsive, shy, or marginalized students could now participate in conversations with their peers, and especially with the use of pseudonyms, could feel free to express their views (in writing) without traditional barriers such as physical appearance, gender, or socioeconomic status acting as "filters" for their comments. The instructor couldn't even get in the way, at least not as before, because he or she no longer "controlled" the conversation, at least not in a traditional sense, since students could ignore or respond to the instructor's comments at will.

Composition instructors, however, were not the only ones to see the enormous potential of this new form of student interaction, and computer conferencing was just one of many ways that computers were being used in English studies. Literature professors, learning of organizations such as The Guttenberg Project, The Library of Congress, and others who were busy creating databases of publicly-accessible electronic texts of classic literary texts, began to envision the possibilities of a world where texts would be available to be read, shared, studied, discussed, written about, and even added to, all in an electronic environment. And creative writers were seeing the potential to "e-publish" their texts, thus creating a wider readership and, according to researchers such as Kaplan, Moulthrop, and Bolter, a more "democratic" method of distribution.

In the Spring of 1986, George Landow, a professor of Victorian Literature at Brown University, participated in an innovative project using a hypermedia system called Intermedia. In this project students were introduced to another form of computer-mediated communication, hypertextually linked information storage and retrieval systems. A "web" of information was built around the life and works of Charles Dickens. Included were critical essays, biographical information, historical information such as the prevalent scientific, political, and religious thought of his day, the complete text of several of Dickens' works, and even scanned-in graphics. The information was stored on a mainframe computer, and was accessible through a group of workstations connected via a local area network. Students could not only read and/or browse through this information but could add their own comments and link them hypertextually to Dickens' work, any of the extra-textual material, and even to comments made by their classmates. One of the drawbacks to this system, though, was its reliance on an expensive mainframe computer, and even though a version was eventually developed that would run on Apple Macintosh computers, the project, though groundbreaking in its accomplishments, was eventually abandoned due to a lack of funding.

Around the same time, a novelist/college writing instructor (Michael Joyce), a classicist/computer scientist (Jay David Bolter) and a computer science professor (John B. Smith) teamed up to produce a software program called Storyspace, which they described as a "hypertext authoring environment." This program had many of the same characteristics as Intermedia, and could be used on a stand-alone computer or across a local area network, and the entire program could be stored on a single floppy disk. This feature alone helped move the concept of hypertext out of the research laboratories and into the English classroom. Storyspace could be used as an authoring tool for creative writers or for those assembling data bases of hypertextually linked information, such as the Dickens works and related material in the Intermedia Project. Storyspace could be used to create and distribute read-only "webs," (data-bases) of information which could be searched by topic, theme, or keywords. Eventually, in 1992, Landow's Dickens Web was translated to the Storyspace software and made available to anyone with access to a personal computer. Unfortunately, Storyspace and documents or databases created with it were only available to users on the Macintosh platform, and Storyspace "webs" could only be distributed on disk or via ftp protocol, a concept which was, and still is, unfamiliar and intimidating to many computer users. (In 1995 a Windows version of Storyspace was finally released.)

In 1991, a young computer scientist, Tim Berners-Lee, ventured across the Atlantic from Switzerland to San Antonio, Texas, where he presented a demonstration at the Association for Computing Machinery's international conference on hypertext. His idea, which he called the World Wide Web, was to create a method of "tagging" documents with certain commands (based on SGML) which would allow them to appear and be read on virtually any desktop computer, whether it was a Macintosh, an IBM-compatible, or even the sophisticated Sun workstation. In this way, Berners-Lee explained, documents could be distributed over wide area networks such as the Internet and be readable by virtually anyone anywhere. The idea sounded rather utopian at first, even though visionaries such as Douglas Engelbart and Ted Nelson had been predicting it for over two decades. Now, in 1996, the term World Wide Web is a household word, as are other descriptors such as the Internet and the "Information SuperHighway." What once was scoffed at by many skeptics as a "pie in the sky" dream is now seen in elementary, high school, and college and university classrooms, and is even accessible from home via a telephone line.

BACKGROUND FOR THIS STUDY

From 1992-1998 I was employed as a community college writing instructor, teaching almost exclusively in a networked computer lab connected to a local area network (LAN). Through attending presentations at various conferences and workshops, and by participating in e-mail discussion lists with other computers and writing specialists (specifically, but not exclusively, The Alliance for Computers and Writing (ACW-L) and Megabyte University (MBU-L), I have followed the development of Intermedia, DIWE, Storyspace, and the World Wide Web with great interest. In my classroom, I have used LAN software such as the DIWE program to promote discussion and collaboration among my students, and have witnessed first-hand many of the benefits of this form of collaborative learning in both composition and literature courses. The possibility of moving the course discussion to a larger venue such as a wide area network (WAN) seemed to be the next logical step, so in the Fall of 1993, with the aid of a very patient college computer technician, I set up a first-year composition and rhetoric course on a local electronic mail bulletin board system (BBS) operated by the community college where I taught. Although the BBS did not provide access to the "true" internet, it did allow students the opportunity to participate in e-mail discussion conferences similar to internet e-mail discussion lists. The next semester I offered an introduction to literature course, and throughout the next two years developed curriculum and offered courses in composition and rhetoric, introduction to literature, and technical communications, all using the college BBS as the main focal point for communication between me and the students and vice versa, exchange of rough drafts of papers, and other interaction between the students themselves. Students would freewrite, answer discussion questions, respond to each otherís comments, send and critique each other's essays, and send personal e-mail messages to me regarding course assignments and their progress (or lack thereof) in the course.

One interesting off-shoot of using the BBS to communicate via modem was that it allowed other BBS users who were not officially enrolled in the course, such as local computer enthusiasts, to participate in the discussions, and also led to the formation of what became known as "The Gatsby Project," in which students from an advanced placement English class at a local junior high school participated in discussions of literature (via e-mail) with college students.

DESCRIPTION OF THE CURRENT COURSE

I continued teaching with the BBS system until Fall 1995, when it became clear that there would be many advantages to the students if I were to move the course discussion to an internet e-mail list, and construct a World Wide Web site which would serve as not only a "place" for me to post assignments, due dates, course syllabi, etc., but, like Landow's Dickens Web, would also be constructed as a hypertextually-linked "learning space" through which students, at their own pace, could browse through sample research papers, guidelines for writing and descriptions of critical theories, other Web sites containing vast databases of literary texts, critical articles, transcripts of electronic discussions held at other schools, and more. The main difference between my course and the original Brown University course utilizing The Dickens Web is that the World Wide Web course is accessible, as the name implies, from anywhere in the world, as well as twenty-four hours per day, seven days per week. Since the majority of my students are working adults who have neither the time nor the energy to battle city traffic to access a large mainframe computer such as that which housed The Dickens Web, offering the course via the Web allows them to connect from home or work anytime they have a few spare moments to download their e-mail or follow a few links from the course syllabus to other Web sites which contain descriptions of literary criticism, biographical and historical information, and so on.

In this new "Web" version of the course(s), the students (in both the sophomore-level survey of American literature course and the first-year composition course based on analyzing literature) began the semester by attending a "live" orientation in which I showed them how to join an internet e-mail discussion list and how to access the course World Wide Web site, which included basic course information as well as links to all course readings except for the major novel, The Great Gatsby (the students had varying degrees of experience with computers and/or the World Wide Web).

For both courses, the first assignment was to read The Great Gatsby, and, via the gatsby-l e-mail list, discuss various aspects of it (such as traditional literary elements, social and historical implications, and views of the novel through the "filters" of several 20th century literary theories) with their "virtual" classmates for about 3-4 weeks. They then collaborated on an annotated bibliography which I compiled and posted at the Web site. After all of this discussion, exploration, and collaboration, they each wrote a traditional literary research paper on a topic of their choice (but related to The Great Gatsby), and in the paper were required to use quotations from not only traditional sources such as academic books and journals, but also information they found on the Web, and perhaps more importantly, comments that had been made by their classmates on the course e-mail list. Rough drafts of the papers were mailed to the list, and the students then read each other's papers and sent e-mail messages offering each other (usually) constructive criticism for revision. The papers were then revised and sent to me, and usually revised at least once more before the end of the semester.

In between all the writing and revising and commenting associated with The Great Gatsby and the research paper, the students read other literary works and discussed them via e-mail (at this point the students branched off into separate discussion lists based specifically on the readings for their particular course--the first-year composition course stayed on gatsby-l, while the sophomore course moved to amlit1-l). On each list, I continued the same pattern which I had established at the beginning of the semester: I assigned weekly readings of short stories, poems, and/or critical articles, asking the students to respond to an initial discussion prompt about the readings (they usually had a choice of at least two or three prompts to respond to), then to reply to at least three of their classmates' messages. Grades were determined fairly holistically, based on the following: participation in the e-mail discussions (50%), the research paper (25%), and a written final exam (25%).

Other than sending the original discussion prompts and answering an occasional question about due dates or a request to clarify a discussion prompt, my original intention was to rarely participate in the discussions. My reasoning for withholding my opinions was an attempt to structure the course in such a way as to (as completely as possible) eliminate my role as the "authority," to allow students anonymity (i.e., they were allowed to use pseudonyms to disguise gender or ethnic background if they desired) and, by requiring participation in but not grading the responses to the discussion prompts, to allow students a neutral "writing space" in which they could feel free to express opinions and explore new modes of inquiry, but which also required them to make use of rhetorical strategies and critical thinking skills to support their claims when they were questioned or challenged by their classmates. I gave no lectures, few outside readings assignments, and only minimal guidelines for their research papers. In other words, I was attempting to create a learning atmosphere as far removed from the "proscenium-based" pedagogy as possible.

When I set up the listserv for this course, I had the option of making it a "closed" list (in this way I could limit participation to students enrolled in the course, and if exceptions were made, people could only join with my permission, thus enabling me to "filter" the participants) or leaving the list "open"; (anyone with access to Internet e-mail could find out about the list and join it). I chose to leave the list open, mainly because I wanted my students to have the benefit of being exposed to a variety of responses and points of view.

One of the unexpected benefits, though, has not necessarily been so one-sided. Besides the students involved in The Gatsby Project, people from a variety of walks of life have joined the list, some as "lurkers" and some as active participants, and the list has given them an opportunity for intellectual stimulation they might not otherwise have access to. Non-enrolled participants have included students at other universities, faculty and graduate students in various academic areas, members of the general public who "discovered" the list on the internet, and authors whose stories were being discussed on the list.

The fact that such a variety of "types" of people participated in the e-mail list is encouraging, and lends credence to Lyotardís claim that:

The transmission of knowledge should not be limited to the transmission of information, but should include training in all of the procedures that can increase oneís ability to connect the fields so jealously guarded from one another by the traditional organization of knowledge. (52) By giving students the opportunity to interact with people other than their classmates and their professor, the list helped students to gain insight into The Great Gatsby and other works that were read and discussed throughout the semester.

In this study, then, I hope to demonstrate how the virtual environment has given birth to a unique type of academic discourse community, and that a new course curriculum and pedagogy is emerging that, although it is still in its infancy, promises to foster dramatic changes in the way literature is studied, taught, and written about. Section II discusses the methodology I have chosen, which is to conduct an ethnographically-based case study of the course.

Section III examines some of the new avenues of discussion which learning in an electronic environment opens, and the effect these discussions can have on invention and collaboration. Students are able to "test" their ideas by submitting them to their peers before they even begin their paper. When they see other students with similar ideas, or conversely, see that their particular idea is not only unique but respected, they tend to take more risks and be more confident than if they were composing alone or with only a small peer group meeting once or twice a week. The elimination of this time/space factor is of vital importance, especially to non-traditional students with families, full-time jobs, or other time-restricting responsibilities; students can send a message at 3 am or 10 pm and check for responses at 9 am, or the next day, or the next week--thus, they can contribute to, and benefit from, the construction of knowledge of their discourse community without space, time, or other traditional barriers.

The networked electronic writing space and its effect on the writing process is explored in Section IV. Just as collaboration and invention are affected by the electronic writing environment, so is the actual writing process. Writing instructors have long since realized the benefits of electronic word processors, such as cutting and pasting, ease of revision, and spelling and grammar aids, but the "networked" writing environment brings several other factors into play. Some of these include: the ability to access and save locations of information online; the virtually unlimited number of potential sources; the ability, while the thought is still fresh, to send messages to classmates and/or the instructor requesting input, advice, or information; and the ability to recall, word for word, previous "class discussions" by scanning through previous messages for specific quotes or to follow "threads" related to a particular topic.

Section V examines the curious phenomena of a virtual/ephemoral/temporal electronic discourse community. Although the discourse community of gatsby-l (which was at once local and global in nature) was short-lived, many students and other participants maintained connections beyond the semester. They conversed with each other, with non-enrolled "e-quaintances," and with me both on and off-list. Some also used their newly-acquired skills to join other lists and thereby gain access to other academic and non-academic discourse communities representating various fields of study or interest. The course allowed many, especially those who participated in both gatsby-l and amlit1-l, to experience first-hand the phenomena of shifting, evolving, and dissipating circles of discourse which they will no doubt encounter often as the business and academic communities continue their shift toward working and learning online.

Section VI discusses some of the changing roles and relationships of students and instructors in networked, computer-mediated learning environments. The role of the instructor in the web-based course has evolved dramatically from a "professor" of knowledge to a facilitator of learning experiences; from an evaluator to a coach; from the all knowing fount of knowledge to a signpost on the information superhighway. In the same vein, the role of the student has also changed; he or she is no longer simply expected to passively digest and regurgitate received knowledge; the student is now required to be an active learner--to follow links to information, to sift through and make connections between thousands of unrelated sources and hundreds of e-mail messages; to not only react in writing to an essay prompt, but to defend his or her views and challenge those of others. In a traditional classroom, it is highly unlikely that a student would have access to virtually every word every other student has written for the course; this access exposes students to a variety of opinions, interpretations and viewpoints that would hardly be available otherwise.

Moving the course to the Web has not only created an entire new world of possibilities, as we have seen, but has also raised an entirely new set of problems that are at once administrative, ideological, and technical in nature. Section VII will address some of these important issues that are brought to the forefront by this new "place" to teach and learn, and will examine the impact of teaching on the "cyber-frontier" on the pedagogical strategies and the work load of the "cyber-instructor." In addition, the section examines the drastic changes relationships between the instructor, college administrators, and technical personnel undergo once the instructor steps into the realm of virtual teaching. This section gives a realistic glimpse into some of the challenges that await the cyber-instructor and present advice for those that will come after as well as suggestions for future research.

Section II