[as published in Computer-Mediated Communication and the Online Classroom. Volume II: Higher Education, ed. Zane Berge and Mauri Collins. 93-110. Hampton Press, 1995.]

In spite of powerful and dramatic developments in theories of how people learn language and literacy, the "default mode" of university teaching remains today essentially unchanged from what has been for the past century: students read a textbook, listen to lectures, engage in occasional class discussions, infrequently write a paper, regularly take an examination (more often quantitative and machine scoreable than essay-type). And this method works about as well as it always has: that is, it favors those who need help least, separating out those who come prepared to learn in spite of the method from those who, for whatever reasons, could profit from educational methods which took more account of the nature of learning.

Collaborative Investigation at St. Thomas

There are some signs, however, that things are changing. The spread of new ideas, public calls for curriculum reform, university initiatives to support and reward teaching, and the advent of new technology are beginning to make inroads in traditional practice. One instance that I can describe in some detail is what is called unofficially the "collaborative investigation" program at St. Thomas University. After more than eight years of development, this work is beginning to receive increasingly widespread attention (see the appended list of references for publications and conference presentations arising out of it).

Although details vary from teacher to teacher and from discipline to discipline, certain principles are widely accepted, whether the course is introductory psychology, Native American Religions, an eighteenth century literature seminar, or a senior-level class in the theory and applications of rhetoric. In each case student research and writing are the central activities of the course, largely displacing textbook reading, attendance at lectures, taking and studying class notes, and group discussion. But it is student writing of a rather different kind than is normally found in classrooms. Students write neither traditional term papers nor essay exams. They do not, that is, write in order to demonstrate the acquisition of knowledge, or for the teacher in the role of examiner. Writing in these classes, insofar as possible, has authentically social purposes. Students write to amuse or interest or inform or persuade a known audience -- an audience capable of being, in actuality, amused, interested, informed or persuaded. Another way to say this is that insofar as possible the written language is what is described by Mikhail Bakhtin (1973, 1986) as dialogic -- that is, it originates in, and constitutes a response to, someone else's query, assertion, or argument, and it is written in anticipation of someone else responding to it.

An illustrative example of this contrast: A few years ago in my own class, I might have asked on an examination, or as an essay topic, why Jonathan Swift is an important author. Now, the class as a whole ask each other, who are the important authors in the eighteenth century, and which should we read, and why? Based on wide (individually selected) readings in a range of textbooks, literary histories, guides to literary study, and so forth, groups of class members come up with lists of important writers, and write arguments for reading some as opposed to others. (Most of the groups usually include Swift; it's difficult to read background material on the eighteenth century without seeing that there is a consensus that he is an important writer.) The class read all the arguments and make the decisions about whom to read based on whether they were persuaded by them. The arguments are not read by the teacher and evaluated. They are not overtly evaluated, even by the students' classmates. They either achieve their end -- persuasion -- or don't. That is, people are either persuaded to read and study the author, or they are not. We select a "short list" of writers, and one group investigates Jonathan Swift's reputation and work. They then persuade the rest of the class to read specific works of Swift. Again, people choose what to read (and what everyone should read) on the basis of the arguments and information presented.

To use written language in these ways in a classroom context involves some fundamental changes in the structure of the teaching process, and in the means by which it is organized and conducted. For example, much time that conventionally would be spent in class discussion or lecture is spent in reading or in writing. Often students read the teacher's photocopied written assignments and explanations rather than listening to oral explanations of assignments or organizational strategies; or they might write impromptu texts on a given issue or subject, which are then circulated around small groups or the whole class for reading and response -- a form of written discussion referred to as "inkshedding." Similarly, time that could have been spent reading a course textbook might in this case be spent in the library, reading in a selection of textbooks or, more important, in the public literature of the discipline. Class time is regularly given up for library work or for groups to assemble reports or arguments, or for such reports to be read and responded to. The instructor's time is distributed differently, as well. For example, written "prompts" are composed, printed, and photocopied rather than lectures prepared. Regularly, a great deal of student writing must be distributed quickly to multiple readers. Sometimes this can be accomplished by circulating individual copies around a group; more often it entails massive amounts of photocopying. Organizing this sort of thing takes time which might otherwise be used in marking papers or planning classes.

It became clear a number of years ago that there are important practical, procedural, and logistic problems associated with utilizing writing in these ways. Circulating written text by means of photocopies is not only expensive, but time-consuming and awkward. Individual access to the writing of others -- especially at the time you need it most -- is logistically complicated and often inconvenient. Consistent production of originals which are susceptible of photocopying (dark enough, clearly enough handwritten) is not a trivial difficulty. Responding to the writing of others entails creating at least somewhat legible marginal comments, or extra sheets of paper, paper clipped, stapled or stuck to the original. And keeping track of all the paper -- making sure that there is a central record of who has written what, and who has responded to whose writing -- poses serious mechanical difficulties.

Computerizing Collaborative Investigation

For such a course and such a concept the next logical step was clear. Computer word processing technology and computer networking clearly have a great deal to offer such a situation. Since 1987 we have been moving toward a situation where we could take advantage of the technology in pursuit of these aims. During the 1989-90 year St. Thomas University installed a Local Area Network on the PCs in the university computer lab. It was a Novell network, running a classroom software program distributed by IBM, called ICLAS. After some initial startup problems, it became clear that such a course could be based on such a network. Particularly important considerations included the ease of signing on, the availability of a uniform and easily-learned word processing program, and the general accessibility of the technology. Accordingly, after Christmas I began encouraging my year-long eighteenth- century literature class to use the account I had set up for the course (students all signed on to the same account, "18C"). Although the computer-based aspect of the course had not been announced in advance, and could hardly be required, by the end of the academic year at least 17 of the 26 students had submitted or edited files on the LAN (in some cases by composing on other computers and arranging to have them loaded onto the server). Although we had not found ways to computerize most of the kinds of writing characteristic of a collaborative investigation, we had managed to put the entire corpus of reports for the "course book" -- in excess of 50 files, ranging from two pages to ten or more in length -- on the network. Many of these files had been radically revised, or constituted separate reports combined into single files, by individuals or groups. At the end of the course, each was copy-edited, usually by two other members of the class, and the whole was printed using the network printer in the lab. As finally combined, the course book took up half a megabyte of disk space (557739 kilobytes, to be exact) and printed out at over 200 8 1/2 by 11 pages.

The logistic problems, however, remained great. While students generally found it possible to learn to handle the lab and the software, it was often difficult for them to get access to the lab, which was not only small, but was used for teaching computer literacy courses and for other purposes as well. This contributed to a decision to delay completion of the book until after classes were over. The physical difficulties of getting files from my own computer, in my third-floor office, to the lab in the basement and back, and maintaining the account, were considerable. There were extremely time-consuming problems with duplicate files (often, copies existed on individual student disks, on my computer, and on the classroom server; occasionally it was very difficult to tell which had been edited and which should be discarded as obsolete; often we had need for backup files because errors were made on the files on the classroom server, etc.)

It was clear, however, that the results of this trial were encouraging. In large part the computer technology actually did facilitate the work of the course -- the dialogic exchange and use of written text -- rather than being only a difficulty to be overcome. It was also clear that using only slightly more elaborate technology, a great deal more could be achieved. Many of the other genres and purposes of writing typically used in such classrooms could, it was apparent, be facilitated by the computer. Over the years since, in fact, our experience suggests that virtually all of these ways can be implemented, and without overwhelming expenditures in hardware or software. Here's how we have worked out methods of rendering such strategies easier through the network.

Technological issues. Some major hard-and software issues that have been addressed since that first trial are the following:

(1) The network has been physically extended beyond the lab to include the machines in teachers' offices. This has had profound effects, including the possibility of using email for virtually synchronous assistance with problems, and the elimination of the problem of duplicate files on various machines: all are now on the server, and teachers have access to the same common directories as students. In the near future the lab will be extended to student residences and will be provided with dialin access, which will make it accessible from home computers, on telephone lines. Obviously, this will make communication even more easy.

(2) A elegant shareware mail system (Pegasus Mail) has been installed on the network. This has not only made collaboration easier, it has facilitated communication between teachers and students. The fact that it is an extremely simple program to use has eased the initiation of many students into computer use.

(3) During one year, we had a usable bulletin board program (that freeware program, BrainStorm, has since been replaced by an unusable commercial version, and we are currently searching for a replacement).

(4) Abandoning ICLAS allowed us to facilitate student-to- student communication and collaboration; a simpler menu program allowed us easily to set up a directory structure which promoted these activities and allowed students easily to share files and exchange messages.

(5) A specific room has been designated as the collaborative investigation classroom, and cabled appropriately. In this particular case, it operates on the same server as the larger, university lab, and students can use either. The CI lab has six or seven (depending on how many are functioning; some are aging 8088-based machines) computers in it, and the larger lab a dozen or fifteen. We expect to expand both, but at the moment this serves with some difficulty the needs of the hundred or so students currently using the network for Collaborative Investigation courses. The network now has a server with 1.4 gigabytes of storage, of which a third is dedicated to student and other user files. The server runs Novell version 3.11 and currently uses a shareware program called DougMenu as a user interface.

Because of the creation of this smaller lab, it turned out, there were serendipitous benefits. Groups meet in the lab not only for electronic editing but for more informal and wide-ranging oral discussions. The lab, over the course of the term, becomes a regular meeting place. Not only is the electronic medium used for collaboration, so is the cork bulletin board (print documents with hand-written annotations are regularly posted there), so is the top of the filing cabinet (boxes of printed bulletin-board strands are left there for annotation) and the chalk boards (graffiti of various kinds accumulate steadily).

Generic issues. Similarly, there are a number of ways in which we have found that genres of writing which these courses promote can be facilitated on the network.

(1) Teacher's messages, instructions, assignments to the class ("Prompts"). Much of the organization and conduct of such courses, particularly at the beginning of the year, is accomplished through written "prompts" which are photocopied and handed out at the beginning of a class meeting, or posted in an envelope on a bulletin board or on the instructor's office door if class isn't meeting. Over the course of the year in the pre-network eighteenth century course, for example, there were over 80 such documents composed, printed, duplicated and distributed, in 30-35 copies each (some ran to two or three pages). Currently, however, such documents can be written and held in a read-only file on the computer network. This not only saves a great deal of photocopying (students make their own printed copies as and if they need them), and the prompts are permanently accessible for reference. Further, questions regarding them can be posted through email as they arise, and dealt with immediately. And in fact, although the difficulties of structuring the network to make that sort of access possible turned out to be greater than we had anticipated, once achieved, the process turned out to be extremely productive. Although finding a reliable and user-friendly bulletin board program to run on a Novell network has not been easy, students have proved eager to engage in such discussions.

(2) Transactionally embedded impromptu writing ("inksheds"). Regularly, as I have described, in-class impromptu writing is generated and circulated as a way of strengthening and structuring discussion, testing ideas, sharing responses to presentations or papers, etc. While it seemed obvious it would be difficult to computerize the writing that is done physically in the classroom (unless there were twenty or more keyboards in the room), such writing is often scheduled to be done between one class meeting and another. Students with access to the lab could compose such writing on the computer and print it for circulation. Even more important, it seemed likely that the principle could be extended by establishing running conversations in accumulative files on an electronic bulletin board; this could be done entirely outside of class time, and printed and distributed to the class at intervals (or read online). Again, this proved to be much easier than we had anticipated: although many of the details of reading and commenting on each others' writing were modified, almost all of this is readily achieved on a bulletin board, or, even more simply, by setting up a common read-write file to which students add their own comments.

(3) Group reports. Besides the research reports which are the central material of the course, there are proposals for study, annotated lists of questions to be explored, proposals for scheduling, and so forth. Without the network, these were done in traditional ways -- individuals drafted sections, and editorial groups or individuals combined sections, revises and edited hard copy. Finally, "designated secretaries" from the group arranged for final copy -- either typescript or computer file. The final copies were photocopied and distributed, sometimes prior to class by being posted on the board outside my office, sometimes at a class meeting. Access to the LAN greatly facilitated this process. Just as computer word processing programs afford revision, they make collaboration easier. Files are written individually and combined electronically; spell checkers and editing software are be used; hard copies are printed as needed.

(4) Responding to reports. Regularly, when a report is circulated, everyone in the class reads it and responds in writing -- in my own courses, usually with a set of questions for clarification or further information. Before the network, texts were usually turned in to me by the responders, whereupon I assembled them into packages according to the report being responded to and distributed them to the authors. The authors then revised their report, using the questions in whatever way they found appropriate. This was a very difficult process to administer, especially when the reports and responses were not tightly synchronized (it was particularly difficult to keep records of who had responded to which reports). But it is perhaps the most powerful aspect of the writing process in a collaborative investigation course: the questions are a very effective way in which students can tell whether their report has been understood, and an extremely productive stimulus for revision. If the reports were on read-only files on the computer, we thought, students could enter their questions in associated files, enabling them to avoid duplication and to elaborate or reinforce questions already asked (because they would automatically have access to all the questions already asked of a given report); it would also provide an automatic record of who had read and asked questions about which reports. Although this, too, turned out to be more difficult than we had anticipated, because of the peculiarities of the network software, eventually we found a method for creating easily accessible read-only and read-write directories for students.

(5) Revisions of group reports. Each such report is (at least potentially) revised in light of the questions of the others in the class. What happened previously was that the group who wrote the original report might take all the questions and a copy of the report and work with all of that at a meeting -- editing, recasting, perhaps doing more research. At the end of the process a secretary would be designated to retype or rekey the report. Putting all the files involved on the computer network obviously makes such a process easier and involves more students more actively. The processes of passing the questions and the various annotated copies of the report around the members of the group are, if not eliminated, rendered much less confusing.

(6) Question & Topic Generation. Regularly, everyone in such a class proposes lists of questions, topics, or subjects to be studied. The process of discussing and deciding among such proposals is often a complex and confusing (though in important ways productive) one. If, for example, 25 students each come in to class with different lists of five or six Restoration writers they wish to argue the class should read, and short arguments for each, circulating the information around the class and helping everyone to make an informed decision can entail a great deal of paper shuffling and reorganizing of groups. It seemed clear that if each recommendation were loaded into a database and sorted according to the author recommended (or question asked, or topic proposed), it would be easy to arrange for everyone to read all the recommendations before making a decision. This, so far, has not proved possible. No database we've become aware of is easy enough to use to allow such manipulations for the computer novices in our classes.

(7) Individual correspondence. Because a great deal of the work of a class like this is carried out in writing, I regularly encourage students to write me with questions or comments about the course or the subject matter, before we meet (if necessary) to discuss things. Sometimes I ask specific questions or raise specific concerns and invite comment. I encouraged them to pin notes to the literal cork bulletin board outside my office (in envelopes if they were private). I wrote responses and pinned those to the bulletin board in turn. It was obvious, and it has turned out to be the case, that this literal cork bulletin board could be made more flexible and convenient by being converted into an electronic one. Messages can either be private or public, depending on the concerns; a network mailing system (Pegasus Mail, as it turned out) could be set up for correspondence between student and teacher (and between student and student), and a bulletin board for concerns that might be of interest to everyone.

(8) Book publication. As I have described, what is in some ways the focal activity of the course is the preparation of a "course book." Usually this book constitutes a permanent record of what the course's subject matter, and is the result of varying amounts of revision and copy-editing of reports prepared by individuals and groups during the course. Such reports were photocopied (from typed, printed or handwritten originals), and permanently bound. Many of the ways in which this activity could be enhanced and made more convenient through putting the course on computer were obvious, and indeed had already been demonstrated through the computerization of the course book, The Eighteenth Century Frame: Studies in the Literature and its Contexts, during the spring term of 1990. More students can be instrumentally involved in more aspects of the book's production; moreover, the final copy can take advantage of the techniques of desk top publication and can be laid out attractively and uniformly typeset, rather than being constituted out of assembled photocopies of whatever individuals or groups submit.

(9) Evaluation. One of the most significant aspects of collaborative investigation teaching as it has developed at St. Thomas has been a novel approach to evaluation. Traditional courses evaluate students primarily on the basis of an instructor's judgment about performance on specific tasks -- essays, examinations, class discussion. Since an important aspect of collaborative investigation involves decreasing the extent to which the instructor is at the center of all discourse transactions in the class, methods of evaluation have evolved which also move the instructor away from the center. Also important is the extent to which the course depends upon students learning from, and teaching, each other. Thus, typically, two factors become central to a new sort of evaluation: (1) exhaustive records of the extent to which students are actively involved in the sequences of assignments and tasks which make up the course, and (2) the students' evaluations of each other's contributions -- in writing, in class discussion, in editorial committees and group work -- to their learning. These evaluations typically involve relatively complex forms, photocopied, by the students, filled out by individual students, and turned in in sealed envelopes.

Both of these matters posed some problems which could, it seemed clear, be at least in part solved by computerization. For example, it is much easier to keep accurate records of student participation in tasks such as writing questions on reports, proposing subjects for study or questions for investigation, or writing, revising, and editing reports. In spite of all its limitations, IBM's ICLAS made such record keeping virtually effortless. The basic Novell network software and menu system we are currently using is not quite so forthcoming, but currently we are exploring possibilities of using its accounting capabilities for such purposes. Perhaps equally important, peer evaluations could be greatly simplified if they were put on computer files from the beginning: such files could easily be converted to data bases which would allow the instructor to see readily all the ratings for a given student in any area. This has worked well, in spite of my own misgivings about the public nature of composing onscreen in a public lab.

An Example of a Course

What happens in a given, specific course, then? In my case, as in most, it all begins with my handing everyone a long, written introduction to the course, and giving everyone time to read it silently. I also hand out, as I do at the beginning of most sessions, a document headed "In Class Today." In September of 1991, that document said, in part:

As you'll discover, one of my central beliefs as a teacher is that reading and writing are powerful tools, and ones we don't use as often as we might. One of the ways in which that belief is acted out in my teaching is that I write a lot, ask you to read it, and expect you to write a lot and expect others (including me, sometimes) to read it. But I don't expect that the writing is going to be used in the way most educational writing is used -- that is, as a basis for evaluating the writer (can she write? does she know what she's supposed to know?). I expect it's going to be used the way you'll use most of these handouts -- to see what I have to say, and respond to it in some meaningful way (by doing what it asks, or arguing that what it asks doesn't make sense, for example).

. . . the second part of the class will involve everyone writing about the literature of the eighteenth century, reading each other's writing, and generating responses and questions. This is a way of ascertaining the sorts of things we all know, and need to know, about the period, about its literature, and about literary study, and generating some issues and concerns that we're going to be addressing over the first few weeks.

In the meantime, we began a running conversation on the electronic bulletin board. Everyone was required to log on and read the board -- and contribute something -- each week. Over the year, contributions varied from Merry Christmas messages and complaints about the heat in the computer lab to an extended, multi-voiced discussion of whether Moll Flanders should be regarded as primarily the author of her own fate or a victim of society. Everyone was also required to touch base with me once a week through the more private electronic mail system. Letters there varied from "nothing to report this week" to long exchanges about the reasons why some people find it harder to participate in oral discussions than in written ones (that one, in fact, expanded into a bulletin board discussion).

Between then and April, according to my estimate, the thirteen students in that course generated well over 40,000 words on the bulletin board and over 30,000 words in electronic notes to me. How much they may have generated in notes to each other I have no way of knowing, but it is considerable. Even without that, the total works out to a bit over 5000 words per student.

Beyond that, of course, there is a great deal of non-computer writing -- inkshedding, question generating, commenting on other people's reports, questioning them, and so forth -- which I have no way of counting or tracking.

But also, of course, and most important, there is all the electronic writing done in the more formal context of written reports to the rest of the class, and comments on and reactions to those reports by their readers. Although the mechanics of this varied as the course (and our familiarity with the computer network) developed, the last cycle of reports, all of which had to do with some facet of the class's reading of various texts of Pope and Johnson, were handled this way. Questions and issues were discussed (in part through in- and out-of-class inksheddings) and then proposed individually, in files in a common directory on the network. Each person in the class was invited to read the questions posed by all the others and add comments and suggestions to the individual files. As the comments accumulated, the authors read them; in some cases these led to modifications of the questions, and in some the authors were offered strategies for finding answers; in most there was a good deal of comment suggesting that others were interested in the questions. Over the next week or so, as the authors began finding answers to their questions, they began putting drafts into the same file, immediately following the sets of questions and comments. As the drafts lengthened, others read them and added comments on, and questions and suggestions about, the drafts in the same files, following the drafts. As authors checked back on the responses to their work, they regularly edited and changed the drafts in response to their audience's questions. Comments on the bulletin board suggested that this was, from most of the students' point of view, the most successful way of managing this collaborative form of writing we had yet tried.

It is difficult in a few pages to give any flavour of the discussion in these files. I didn't work fast enough to arrange to archive the original drafts, so as I go through the files it isn't clear how the comments on first drafts affected subsequent modifications, but let me pull a few examples out of one such file. It's the one which began with this question, from Darice (all names have been changed):

My question has to do with Pope's repulsiveness both physically and personality wise as I feel his ability to write satiric literature may be connected (simply because I'm sure since he is described as looking like a toad that he knew that people found him repulsive and therefore promoted this repulsiveness in his personality, which ultimately led to an ingenious ability to compose satiric literature as a way to overcome the public's view of him). I may be way off, but I feel that this may be the case as Russ explained in class that Pope used to get very irate if someone had said his parents had been poor and also that not many people who knew Pope liked him. If I'm unable to get any information upon this connection between his physical and personality repulsive character which may have influenced his poetic ability, then I thought I might just pursue the reasons behind his physical deformity.

Any suggestions? What do you think: a dead path or possibility? I realize this is not a question orientated specifically to historical background, but more a background on Pope (Russ is this o.k.)?

Darice: This sounds interesting. It's nice to get another side of things -- a background, or at least some kind of sense of this sort of thing. -- Pat

There are arguments about some of these issues (different biographers have different views). One way to focus it would be to present some views of it, specifically ascribed to the authors; it's certainly a question worth asking. -- Russ

Darice, I didn't pick up on the fact that Pope wasn't a 'very handsome' fella, or as you say "repulsive"! I think you might have to look at Biographies etc to see what his background was like and family life which might have influenced his personality, but, (not to discourage) I think it would be difficult to determine that someone's physical appearance affected their personality. -- Greta

Darice, I just read something that described when Pope developed his disease and the pain it caused him. The book relates that as a young man Pope was first stricken with the disease and was convinced he was dying. He even went so far as to write letters of goodbye to his friends. Perhaps the constant expectance of death influenced his nasty attitudes and helped sharpen his satirical tongue. -- Tamara

I just read something else about him too: a description of him at fourteen: "He is small and pale, fragile, and already not quite straight in the back . . . , but he has a frighteningly sensitive face, large wondering eyes, and an enchanting voice which will earn him the name of 'the little nightingale'". I don't know if this is relevant to what you are doing, but I thought it might be nice to hear a pleasant description of the poor guy. -- Tamara

Pope was often described in quite attractive terms by his friends, often quite similar to what you found, Tamara. (Who is that, by the way?) -- Russ

I took the quote from Bonamy Dobree's book Alexander Pope, published in 1952. -- Tamara

. . . Since they were Catholics at a time when England's religion was Protestant, the Pope's were forced by antipapist legislation to move often, which prompted Mr. Pope to retire from his successful linen business. There is little known of Pope as a child, except that he experienced several traumatic experiences. Although Pope was not physically deformed as a child, his half sister, Mr. Rackett informed Pope's biographer, Mr. Spence that when he was between the age of three and five "a wild cow that was driven by the place where he was filling a little cart with stones struck at him with her horns, tore off his hat which was tied under the chin, wounded him in the throat, beat him down, and trampled over him" (Russo, p.27). Further Pope studied under four priests, one of which was said to have whipped and ill-used him for writing a satire (isn't that ironic!).

Darice, I enjoyed this report very much; it was an interesting way to look at Pope and his work. However, I think you may have overlooked something of relative importance: what can explain his friends' kind attitudes towards him? Surely, he must have had some attractive qualities. For instance, his voice was quite enchanting--could that have affected his ability to create such rhythmic, lyrical verses? Perhaps not, but I do think his positive attributes should also be explored. -- Tamara

Darice, I have been fascinated with the physical descriptions of Pope since reading about him. This report is very helpful in giving me a more vivid picture. Have you read Johnson's "From the Lives of Poets", the section on Pope? I read it for this week and I must say, it is a very informative piece. Not only does it talk about his works, but about his personal life too. It said that he"never took tea without a stratagem" - his mind was always on the go. It also said he thought quite highly of himself. My question is, "did his brilliantly sharp mind and maybe, his somewhat conceited air have something to do with his physical deformities - was his mind compensating for something else? Something to think about and you should read it if you have not already - its really interesting. -- Barb

Ultimately, the relationship between comedy and its audience cannot be measured because society is not homogeneous in nature; there can be no absolute because there is no universal standard.

I found my initial report to be very formal. I think we were trying to impress you, the professor, rather than our classmates because that is what we are used to doing. I think when we write essays we tend to try to aspire to academic heights and we try to sound as academic as possible. When we write for the benefit of our classmates, we know that they are at the same academic level, so we don't have to sound so professional. The writing in class is more friendly; more personal and less formal. I think, too, I am more relaxed in my writing because there isn't the pressure of a paper that is worth 40% of the mark. With this type of class, I am able to relax and this changes my writing style, I believe.

From what I've read about the often diseased food at the time, I don't think I would have wanted to have eaten back then.

I think, as well, there is evidence of similar learning going on in other situations in this class. One place I find it most obvious is in the postings to the electronic bulletin board. Besides the growing concreteness and specificity of the messages, what I am struck with is the new consciousness of this medium as a device for forging and maintaining social relationships as well as carrying on an intellectual discussion. There is strikingly more casual and efficient references to the positions of others ("The question was originally brought up by Barb"), to other documents in the class ("When you consider Tanya's report on employment for women"); there are in-jokes ("for any Soc. people out there"), etc. All these suggest a context and relation between writer and readers very different from what is implied in those early, formal messages, written in obvious consciousness of the teacher reading over the author's shoulder.

There is an increasing consensus (at least among the educational writers and practitioners I have the most respect for) that the best sort of teaching is the kind that engages people in what Frank Smith (1983; see also Dixon & Stratta, 1984) calls "an enterprise" and then observes closely to see what they can do, what they actually do, what -- as Vygotsky (1962, 1986) insisted -- they can almost do, and can do with a little help from their friends; and then finds ways to promote learning that's specific to where each learner is and to where she needs to be. I think, in general, that's what's going on in classes conducted like the one I have described. And the computer network facilitates this in ways that seem to me to be fulfilling some of the long-promised and long-delayed benefits of computers in education.

Bakhtin, M. M. Marxism and the philosophy of language. (1973). ["By V. N. Volosinov"]. Tr. L. Matejka and I. R. Titunik. New York: Seminar Press [Harvard University Press].

Bakhtin, M. M. (1986). Speech genres and other late essays. Tr. V. W. McGee, ed. C. Emerson and M. Holquist. Austin: University of Texas Press.

Dixon, J. and L. Stratta. (1984). "Student enterprises with personal and social value." [Series B: Writing 14 to 18, Discussion Booklet 3]. [ERIC Document No. ED268522.]

Smith, F. (1983). Essays Into Literacy. London: Heinemann.

Vygotsky, L. S. (1962). Thought and language. Ed. and tr. E. Hanfmann and G. Vakar. Cambridge, Massachusetts: MIT Press.

Vygotsky, L. S. (1986). Thought and language. Tr. A. Kozulin. Cambridge, Massachusetts: MIT Press.

Hunt, R. A. (1989). "A horse named Hans, a boy named Shawn: The Herr von Osten theory of response to writing." Writing and response: Theory, practice, and research. Ed. C. M. Anson. Champaign-Urbana: National Council of Teachers of English. 80-100.

Hunt, R. A. (1991). "Modes of reading, and modes of reading Swift." The experience of reading: Louise Rosenblatt and reader-response theory, ed. J. Clifford. Portsmouth, New Hampshire: Heinemann-Boynton/Cook. 105-126.

Reither, J. A. (1985). "Writing and knowing: Toward redefining the writing process." College English 47(6): 620-628.

Reither, J. A. (1990). "The writing student as researcher: Learning from our students." The Writing Teacher as Researcher: Essays in the Theory and Practice of Class-Based Research. Portsmouth, New Hampshire: Boynton/Cook Heinemann. 247-255.

Reither, J. A., and D. Vipond. (1989). "Writing as collaboration." College English, 51(8): 855-867.

Hunt, R. A. (1987). "'Could you put in lots of holes?' Modes of response to writing." Language Arts 64(2): 229-232.

Hunt, R. A. (1989). "Learning to converse with texts: Some real readers, some real texts, and the pragmatic situation." SPIEL: Siegener periodicum zur internationalen empirischen literaturwissenschaft 8(1): 107-130.

Hunt, Russell A. (1991). "R Texts Us?" Inkshed 10(1): 5-6.

Hunt, R. A. (1991). "Subverting the literary system: Nonhegemonic literary socialization." Proceedings of the second IGEL-conference: Amsterdam 1991. Ed. E. Ibsch, D. Schram, and G. Steen. 175-180. Amsterdam: Rodopi.

Hunt, R. A. "Texts, Textoids and Utterances: Writing and Reading for Meaning, In and Out of Classrooms." Constructive Reading: Teaching Beyond Communication. Ed. Deanne Bogdan and Stanley B. Straw. 113-129. Portsmouth, New Hampshire: Heinemann-Boynton/Cook, 1993.