RADICAL CHANGES IN CLASS DISCUSSION

USING NETWORKED COMPUTERS

Computers and the Humanities 24(1990):49-65

Jerome Bump

Dept. of English, University of Texas, Austin, TX. 78712-1164

 

Jerome Bump, Professor of English at the University of Texas, is the author of Gerard Manley Hop­kins (1982) and "CAI in Writing at the University: Some Recommendations," Computers and Edu­cation 11,2 (1987), 121-33. He explores the interface of CAI, psychology, and the humanities.

 

Abstract

This study examines the effects of conducting class discussion on a local area network. A real time networking program (INTERCHANGE) was used for class discussion in  freshman and senior litera­ture courses and in a graduate humanities computing class. Pseudonyms, collaborative exams and es­says, and computer-assisted reading were tested, along with organization of the students by sex and personality type. At the beginning and end of each semester in each class 50 to 70 multiple choice questions were asked of the students. Their answers revealed that the many advantages of computer-assisted class discussion (CACD) clearly outweigh the disadvantages.

 

 

Key Words: CAI in literature, LAN, networks, pseudonyms, gender, introvert, extrovert, creativity, minorities,  reader response, collaborative writing and exams.

 

                  Individual micros, the focus of research in computer-assisted instruction (CAI) to date, seem to encourage writing as an isolated, individual activity (Heim, 1987, p. 152; Payne, 1987, p. 24), but particularly when clustered in computer labs or classrooms they demonstrate that writing is also the  result of social interactions among individuals (Feldman, 1987, p. 6; Logan, 1989). Increase in compe­tition between individual students has been ob­served because of the public nature of the computer screen (Herrmann, 1987, p.85), but collaboration (cooperation between students working together in groups of two or more) is the most obvious result (Arms 1984, 1987; Kelly, 1987). Single microcom­puters are often used by two or more for a project (Kelly, 1987, p. 29; Herrmann, 1987, pp. 89-90; Arms, 1987, p. 69; Kemp, 1989; O'Connor, 1989), and in many courses comments on drafts are inserted into a student's electronic file by teachers or other students (Schwartz, 1984; Bump, 1987, p. 131). However, the most intense collaboration occurs when computers are electronically linked to each other to form networks (Heim, 1987, p. 157; Bump, 1987, p. 131; Herrmann, 1987, p. 87). While termi­nals connected to a mainframe have inspired some projects (Jennings, 1987), micros linked in a local area network (LAN) have stimulated the most innovative research in the social aspects of CAI be­cause in the new real time, synchronous LAN CAI software students send messages concurrently to groups or to the class as a whole and the messages are displayed on all the screens almost instantly. After using such a real time program (INTERCHANGE), Lester Faigley observed, "That reading and writing are inherently social rather than individual activities is demonstrated when a class communi­cates electronically" (n. d.). The primary social activity is collaboration not only between students, but also between students and teachers, and even between teachers. Valerie Balester and Kay Halasek, who used the same program, realized that "In teaching the course collaboratively, we redefined our roles by sharing authority with one another and with the students, creating, as Paulo Freire advocates, a problem-posing environment in which students become student-teachers and teachers become teacher-students" (1989, p.4).

                  This kind of electronic communication within a group in the same room has been envisioned since 1945 and experimental sys­tems have been constructed since 1960 (Hiltz, 1978), but computer conference rooms have rarely been suc­cessful, primarily due to hardware problems (Kraemer, 1983). Some of the more promising initial research came from Xerox PARC, which has set up an experimen­tal meeting room known as Colab for groups of two to six people sitting at Xerox Lisp Machines con­nected by Ethernet. They engage in such activities as brainstorming in which different individuals type in words or phrases on their large individual display screens, transmit them to everyone else's screens, and then proceed to discuss them as a group (Stefik, 1987).  However, even if CAI software were available from such a company, expensive business conference rooms are probably not a feasible model for a computer classroom.

                  Hence, until recently, partly because of the expense involved, most of the reports about the results of networking microcomputers in CAI to date have been based upon less than a full classroom of truly networked computers (connected to each other in a LAN as well as to a server) and lack of ap­propriate software, often caused by the radical difference of perspective between the teacher of En­glish and the computer programmer. CAI LAN software for English classrooms is now beginning to appear, inspired by Trent Batson's pioneering work with deaf students at Gallaudet University and by the ENFI project (Electronic Networks for Interaction), sponsored by a grant from the Annenberg/ Corporation for Public Broadcasting Project. Admittedly, there has been some initial confusion and misinformation about these new networking programs for English CAI. InfoWorld, for instance, re­ported that Realtime Writer (RTW) was "the only commercial product specifically targeted for this function" and that "RTW software is also being used at the University of Texas" (Stephens 1988), but neither statement is true. Because the University of Texas was not included in the grant, we asked the Daedalus company[1] to develop INTERCHANGE, and thus there are now at least two different com­mercially available programs specifically targeted for synchronous class discussion in English courses.

                  Whichever program is used, the basic assumption of research on computer writing networks has been that students will benefit from truly collaborative writing. As Batson points out, "most col­laborative learning classes stop short of actual group writing. They may think together and plan to­gether and then, after they write individually, critique their writing together, but they probably won't write together. They don't observe each other's writing process. ENFI makes this last step possible" (n.d., p. 14).

                  Faigley's research with INTERCHANGE suggests that at least students' memories of elec­tronic conversations tend to be of the whole class rather than the few individuals who dominated: "Not only do the many voices act out Bakhtin's principle of dialogism, but the movement recalls the opposi­tion he described between the monologic centripetal forces of unity, authority, and truth and the dia­logic centrifugal forces of multiplicity, equality, and uncertainty" (n. d.).

                  The dominant individuals in the ordinary classroom are often males. Christine Neuwirth found no significant gender differences in her study, but for a wide variety of rea­sons it is often argued that males are much more oriented to computer-assisted instruction than fe­males (Marcus 1987, p.134; 1983), and of course we have all the research on how women tend to be interrupted more and are less assertive verbally (Rich, 1979; Spender, 1980; Selfe, n.d.), as well as all the stereotypes about what men and women tend to talk about when they are segregated by sex.

                  Because a few individual males are not allowed to dominate in the electronic classroom, mi­norities are liberated (Peterson, 1989). As Faigley puts it, "Instead of being tools of repression in the skills-and-drills curriculum, computers joined in a network can be a means of liberation, particularly for those students who are often marginalized in American classrooms" (n. d.). In one of his INTER­CHANGE classes a woman told Faigley that before INTERCHANGE she had never said anything in a classroom since the tenth grade. INTERCHANGE "releases students from some of the socially de­fined limits of being a student as well as those of gender," as it eliminates appearance, paralinguistic behavior, and the gaze of others as factors in the communication. A foreign student in one of Faigley's classes noted that the computer removed his accent and a Hispanic student observed that "the com­puter has only one color" and one printface (n.d.). The result is more honest communication from more people. Nancy Peterson, who also used INTERCHANGE, observed "dialogues whose candid­ness and honesty I have yet to encounter in a verbal class discussion, especially from students who or­dinarily will not volunteer opinions in a conventional classroom setting" (1989, p.8).

                  However, a problem in any computer-assisted instruction system, especially for minorities, is the diffi­culty of learning to use a particular computer system at the same time one is learning the sub­ject mat­ter of the course. This can lead to  "technostress," now broader in meaning than it was in the original title of a study of psychiatric patients who began "'internalizing the standards by which the computer works: accelerated time, a desire for perfection, yes-no patterns of thinking'" (Heim, 1987, p. 201). Another cause of tech­nostress is the huge amount of communal text generated by networked computers which can create a "forest and trees" problem (Langston, 1987, p. 6) especially for the indi­vidual who is not selective and feels responsible for carefully assimilating all of it. This can be a prob­lem particularly for the teacher, who also may be troubled by a loss of control, one more result of al­lowing students to use electronic networks (Kaplan, 1989).

                  Another potential danger in the electronic classroom is

that aspect of computerized telecommunications known as "flaming," the "tendency to write messages on the computer so directly that the usual norms of civility and polite­ness fall away" and the result is "'confrontational style'." The connotation is usually negative: "the di­rectness of digital writing some­times surprises the writer -- and may even upset the reader. So writing without barriers can also prove to be writing without restraint" (Heim, 1987, pp. 209-210). Such conclusions are usually based on studies of electronic mail, that is, leaving messages in asynchronous one-to-one electronic communica­tions. Sproull and Kielser, for instance, found more profanity, negative affect, typographic energy (capitalization and exclamation points), and self-absorption in their study (1986). Neuwirth (1987) followed up on Kiesler's tests with pairs of linked computers, testing for subjective affect, emotional state, evalua­tions of partner and self, expressive behavior, uninhibited behavior, self-disclosure, etc. She found that the computer interactions produced no significant differences except for greater ex­clamations and superlatives.

Research Questions

                  Almost all of this research on the networked classroom has been focused on writing, especially freshman and basic writing, the preoccupation of most CAI in English and of RTW (Batson, n.d., p. 5). At the University of Texas we too have conducted extensive research in this field. However, we devel­oped a separate, one-on-one networking program akin to electronic mail for peer editing in freshman[2] writing courses, CONTACT, and looked to new horizons for our synchronous network program, IN­TERCHANGE. We have gone on to a new frontier of research: the use of a LAN for what we might call computer-assisted class discussion (CACD) in literature (Butler, 1989) and in humanities com­puting courses. Writing is still a requirement in such classes, of course, and thus we can still test stu­dent interest in collaborative writing on an electronic network. But we are able to ask a new question: how often would students like to use CACD in a literature or humanities computing class? In addition, we wanted to test our assumption that students prefer the new real time CACD programs to the old electronic mail techniques. In other words, do students prefer asynchronous or synchronous network­ing for CACD?

                  There are some other basic questions that also need to be asked. Is the ability of CACD to al­low all members of the class, including minorities, to contribute to class discussion the most important advantage, as previous research suggests? How does it compare with its capacity to allow the class to break down into smaller, more practical discussion groups; to allow the instructor to give more indi­vidual attention to students; or the printed and electronic transcripts the software generates of the en­tire class discussion? In addition, we wondered how students feel about using writing rather than speaking to converse? How does CACD affect their styles of persuasion and argumentation? What ef­fect does the opportunity for brainstorming in a group have on individual creativity? 

                  Moreover, if we make various assignments we can ask other important questions. For instance, concerning the problem of certain individuals dominating discussion, what would be the effect of as­signing students to groups on the basis of gender or on the basis of personality traits such as introvert vs. extrovert (on the Myers-Briggs scale) or dependent, counterdependent, independent, and interde­pendent? What would be the effect on class participation of using the "share, not compare" rule used in the treatment of dependent personalities, that is, instructions to make only "I" statements, share only one's own experience, not compare or give advice to others? Would this decrease confrontation and increase self-disclosure? What effects would the use of pseudonyms have in CACD? Would they en­courage honesty and expression of emotions or would they increase confrontation?

                  Other assignments allow us to ask other questions, such as, would the students really want to take advantage of the opportunities for collaborative writing and collaborative exams in CACD? How would reading be affected if a text would be read in a window on the screen and students would type in their responses?

                 Often overlooked in the research on CACD are the problems posed by the new technology. Hence we wanted to ask the question, which are the most important disadvantages of CACD: the re­liance on keyboarding, the loss of voice communication, the slow speed compared to speaking, the loss of coherence, the absence of a controlling instructor, dehumanization, technostress, or the room ar­rangement of the typical computer lab?

Population Samples

                  We have asked these questions of students at the University of Texas. In the Computer Class­room attached to the Computer Research Laboratory of our English department, INTERCHANGE has been tested on over a dozen classes and the widest possible range of students: from freshman to graduate. Butler (1988, 1989), Balester and Halasek (1989), Peterson (1989), Slatin (n. d.) and Faigley (n. d.) have all tested INTERCHANGE in freshman English classes, primarily the second-semester course, E309. I tested it in a two-semester honors Freshman English class, E603A and E603B, World Literature and Composition, limited to 18 students, in 1987-88. In the spring of 1988 I tested the pro­gram on 33 students, primarily seniors, in E376L, "Family in the Victorian Novel," which focused on ac­cess to emotion in reader response.               In the summer of 1988 I tested the program on 12 graduate students in E388L, "Introduction to Computers and English." This was primarily the Holist type of humanities computing course described by Ide (1987), though it was also a comprehensive sur­vey of applications and resources with frequent fo­cus on programming and methodologies with the ex­pectation that most students would write their own software by the end of course. Where other Holists teach the general history of computing and basic computing concepts, we did so primarily in terms of artificial intelligence research, emphasizing it and hypertext much more than the courses discussed by Ide (1987), Tannenbaum (1987), or Oakman (1987). To encourage innovative programming, we had six graduate students from a previous incarnation of the course demonstrate and discuss the advanced programs and methods they had developed. Students could not only go on to adopt humanities com­puting as a minor field of study, but also as a Ph.D. con­centration.

Software

                  In order to compare synchronous and asynchronous CACD, at times we used CONTACT, also developed by the Daedalus company for us. CONTACT is a version of electronic mail adapted to the needs of English CAI.  Like most traditional asynchronous net­working programs, it allows only one-to-one conversation with a specific in­dividual and is used primarily for peer editing of student essays. Students can send or read messages within or between classes at any time whether in the computer classroom or outside it communicating by modem (Carter 1989).

                  INTERCHANGE is for synchronous networking of an entire class session. Like CONTACT, it accommodates a full classroom, with one student per micro, runs on Novell or the basic IBM token ring network, and requires no special hardware. In our classroom, thanks to a grant from IBM, we have thirty-three networked IBM computer workstations, each with its own printer and hard disks, all free from reliance on a mainframe.

                  The differences between INTERCHANGE and the only other currently available synchronous CAI program, RTW, have affected the research they have generated. In both pro­grams, the teacher's contribution to class discussion is just one of many messages on  the screen. However, in the research in most of the ENFI classrooms using RTW, a second video network is used (requiring a second mas­ter monitor, a teacher video switchbox and additional cabling to each student computer) which allows the teacher to control the screens and hence the discussion in the classroom.[3]  INTERCHANGE, on the other hand, because it is not packaged with such hardware, levels differences between teachers and students (Balester and Halasek, 1989), and thus promotes more power sharing or "status equalization" (Sproull and Kiesler, 1986) between student and teacher.

                  The most striking difference between INTERCHANGE and RTW is the hypertext option. However, as that was developed to solve the coherence problems endemic to real time software and was not available for the tests reported here, that feature will be discussed briefly in connection with coherence.

                  A less important difference between INTERCHANGE and other programs is the nature of the "private area" for composing messages, etc. In RTW, as in the Xerox model, "Each user's screen contains a shared area in which messages appear in a kind of ongoing chatter and a private area where users compose messages to send" (Langston, 1987, p.6). In INTERCHANGE the "private area" is a pop up window which disappears after the student has sent her message, leaving the student with a full screen of messages from others. Students are allowed to join various conferences (comparable to the "channels" in RTW) and can even create some of their own, if allowed by the instructor. Once they join a conference the messages that have already been sent by the other students in that conference begin to scroll by, each automatically preceded by the name of the sender, as in RTW. Another option in INTERCHANGE is a special window in which a student essay, a previous INTERCHANGE session, or a text supplied by the instructor can be read as it scrolls by and the student can respond in his/her own window directly to this primary text.       

Procedures

                  At the beginning and end of each semester in each class 50 to 70 multiple choice versions of the research questions listed above were asked of each student, with the last one being a request for a detailed critique. The answers to these questionnaires are the basis of this report.

                  In the graduate class, students used CONTACT to talk to each other about their projects. Some people sought general advice on research pa­pers from specific individuals, and the students writing programs in BASIC sought suggestions and help from each other as did those writing in PAS­CAL. Most of the communication on CONTACT oc­curred before or after class. In the second semester of the freshman course, however, we used CONTACT for an entire class pe­riod for discus­sion of literature.

                  INTER­CHANGE was used for class discussion of literature every fourth class in E376L and in E603A and B and about a fourth of the total class time in E388. Discussion was not graded and often in E376L there was no teacher presence in the discussion. In the first semester of the freshman course the instructor set up conferences in advance and provided initial questions and statements. Students joined different conferences and were able to move around as they chose.  Hence, there was a ten­dency of students to join the most popular confer­ence in ever increasing numbers as the class pro­ceeded and it became difficult to keep up with all the comments. In the second semester, freshmen were assigned in groups of four or five, insuring small group discussion, though they got to read the printed transcripts of all the groups.

                  The test of INTERCHANGE in E376L was designed more for those ori­ented to the social sci­ences, as it involved a larger, more statistically significant group of students and provided for control groups. The students  read five novels over the course of the semester and began discussion of each novel with the class meeting as a whole. For sub­sequent classes we split into two groups, with the first half of the alphabet going to the computer class­room to discuss the novel in conferences of four or five students on our networked computer system, while the second half of the alphabet discussed the novel in small groups of four or five students in the regular class­room. For the next class meeting, we would switch, with the sec­ond half of the alphabet going to the computer classroom to discuss the same novel and the first half meeting face to face. Thus each of the five novels was discussed both ways by all stu­dents, alternating between regular and electronic classrooms, giving them ample opportunity to com­pare.

                  In the literature courses we also experimented with pseudonyms, such as parts of social secu­rity numbers, rather than names, and with conference organization by sex. In E376L we tested the "share, not compare" rule and conference organization by personality type using the Myers-Briggs scale and a measure of dependence developed by our counseling center. In addition, all the second semester freshmen and most of the graduate students took their three-hour final exams on INTER­CHANGE. As a concession to tradition, they were allowed to type in their own individual contribution for the first 30 minutes, 60 in the case of the graduate students. But then anybody could see anybody else's exam on his/her monitors and the rest of the time was spent re­sponding to others. The impact of INTERCHANGE on composition was tested in E376L by having students write analyses of the tran­scripts of the class discussions generated by the program and compare them to other writing assign­ments. We also used INTERCHANGE for more direct reader-response experiments in E603B by having a text scroll by in the extra window and freshmen type in their responses to it as they read it for the first time. The reading window was adjusted so that they could see only a few lines at time, though of course they could scroll through the poem at will.

Results/Discussion

                  The result in all three courses was a truly egalitarian, student-centered inter­change which sup­ported relatively democratic discussion by all concerned of the goals and methods as well as the sub­ject of the course and thus can serve as a model for more radical changes in education called for by Friere and Selfe. In E376L, for instance, asked to rate "the most valuable technique in the course" 41% of the students chose the com­puter-assisted small group discussions, 20% chose face-to-face small group discussions, 20% chose discussion with the class as a whole, and 17% chose watching psycholo­gists discuss family systems on videotape. That INTERCHANGE would be more popular than small face-to-face and whole class discussions combined was perhaps the most  remarkable result of our ex­periment. One of those who voted that way added, "I felt that I could express my opinions a lot more openly. I also felt like every­one had more of chance to say what they wanted and however much they wanted without having some­one interrupt them." Another simply said, "I was more apt to express my feelings using the computer than I was in group or face-to-face discussion." Freshmen commented: "I looked forward to the [INTERCHANGE] days a dozen times more than I did the roundtable discus­sions"; "I feel like in here [INTERCHANGE sessions] I really expressed my true opinions, ...really got into the material and the discus­sions a lot more so than during regular class times. Also, I enjoy writ­ing. Not papers and essays and stuff like that, but I do enjoy expressing my feelings and opinions in this manner."

Asynchronous vs. Synchronous Group Discussion

                  In their questionnaires, 14 out of 16 freshmen respondents voted to always use INTER­CHANGE rather than CONTACT for class discussion of literature, 2 voted for using CONTACT a third of the time. They commented: "INTERCHANGE moves much faster and allows for quicker thinking and transfer of ideas"; "CONTACT can get cumbersome and isolate you from a lot of ideas, reducing real dis­cussion"; "the disadvantage to the CONTACT system is that it was no different from writing notes in class." In the graduate class, however, which included some collaboration on projects as well as networked class discussion of CAI, 5 out of 10 wanted INTERCHANGE most or all of the time, but the other 5 wanted to use CONTACT half the time.

                  Most freshmen were content to keep the same pattern of usage for INTERCHANGE, every fourth class, though one voted for every class in the computer class room. Most seniors voted for every third class, with the number rising from 54% to 61% by the end of the semester. In the graduate class, where INTERCHANGE use had been comparable, 11 out of 12 respondents voted for more INTER­CHANGE use, with 6 voting for every third class, 4 for every 2nd class, and 1 for every class. 

All Students Participate

                  Nine of the eleven graduate students, 84% of the seniors,  and half the first-semester freshmen felt that the primary advantage of INTERCHANGE was that it allowed all members of the class to contribute to class discussion, corroborating the research finding that electronic conversation tends to be more equally distributed among the participants. Students commented, INTERCHANGE "allows for a more detailed and lengthy discussion since everyone can speak at the same time"; "even the peo­ple who don't talk as much in class find it very easy to communicate through the computers. Also, ev­eryone has a chance to share their views on everything without running out of time as we would in class." Whereas in the ordinary sessions certain individuals tended to dominate, leaving up to half of the class contributing only if the teacher called upon them, in INTERCHANGE everyone was able to contribute as much as they liked, and back up what they had to say with word-for-word quotes from the book. Those who tended to dominate resorted to capital letters and exclamation marks (see Sproull, 1986; Neuwirth, 1988), but even they liked the format; as one put it, INTERCHANGE "gives me a chance to speak whenever I want without the rudeness of interrupting someone." In general there was less fear of other people, more opportunities for honest confrontation: "we always get into much more controversial, and thus more interesting, discussions than in class, perhaps because people are not afraid to start arguments because no one can refute them to their face, just on a computer screen. "

                  More importantly, shy students were liberated; one student wrote, "I myself am a bit shy and, in class, am nervous to just blurt something out for fear that it might be wrong. However, with this system, I get a chance to think things out before I say ...them.  When questions are asked verbally, stu­dents do not have a chance to answer them with their best thoughts because they feel rushed. So, in short, this system helps me to give better, more concise answers because I have a chance to think be­fore answering."

Liberation of Minorities

                  One of the senior women stated: "I, as a less aggressive person, had a fair part in class discus­sions with the use of computers." She was referring to CACD even without gender segregation in con­ferences, a comment which suggests connections between CACD in general and less aggressive femi­nist models of reading. For instance, the desire to dominate class discussion is akin to the desire to "master" a text. Hence Clara Juncker suggested that Helene Cixous's "receptive, rather than aggres­sive, mode of reading a text might also work in our interaction with student writers" (433). She cites the experience of a teacher who insisted on teaching in silence: "The result was, he tells us, an amazing eagerness, particularly on the part of the most quiet, mute(d) students, to grab the chalk and con­tribute to the quickly expanding 'silent' discussion on the blackboard. By allowing his most marginal students to use the silence in/with which they felt comfortable, he ironically gave them back their voice(s)" (433-4). CACD restores voices to all such students more effectively, whatever their sex, race, class, or age. Our network has a telecommunications link and thus we can also expand CACD to in­clude others marginalized by geography or other handicaps (Carter, 1989).

Small Group Discussion

                 Five of the freshmen at the end of the first semester and eight at the end of the second, along with four per cent of the seniors and four of the graduate students identified the primary advantage of INTERCHANGE as its ability to allow the class to break down into smaller, more practical discussion groups. One second semester comment was "Small groups allow people to pursue and support an ar­gument more com­pletely than the ordinary class discussion." 

Individualized Instruction

                  Another advantage of INTERCHANGE, that it allows the instructor to give more individual attention to students, was checked as the third most important advantage by four first-semester freshmen, 25% of the seniors,  and five graduate students, but one graduate student voted for this as the primary advantage and two as the second most important advantage. This feature might have been more popular if the students had sought or needed individual attention more, and/or if the teacher had found more creative ways to make use of this option. This is perhaps the most radical feature of INTERCHANGE, however: its tendency to make the controlling instructor obsolete.               

Visual/Writing vs. Auditory/speaking

                 Although word processing, and electronic mail in particular, results in a style more like spoken language, students who felt they expressed themselves in their writing better than in their speaking were also liberated by INTERCHANGE.  To see one's part in a discussion and others' and be able to return to them instantly is a big change in class discussion obviously. Praise for the program was often tied in this way to the visual or writing aspect. "On the whole," according to one student, "it also makes us think more and analyze more and it leaves us up to our own writing abilities."

Improved Thinking

                 In part because of this connection to writing, many students felt their thinking was improved by the program: "computer discussions allow a person to collect his thoughts more efficiently, and prepare something that won't get lost in the shuffle of a 15-person debate"; because visual and audi­tory cues are not allowed "it forces the students to use a clearer approach than they might in verbal discussion. You can think about logical reasons to back up your comments, look up citations, and in­clude them in your arguments."

Improved Creativity

                  Perhaps an even more important aspect of CACD is the synergistic effect of the communal brainstorming. The seniors wrote "Interchange was great, we should have done it more. That's where I got most of my ideas"; "It helps you to get your own ideas to surface when someone else gives an ob­servation about a passage that is familiar to everyone in the class." A freshman stated, "I think it allows for a lot more imagination and creativity than does oral class discussion." This characteristic, along with our experi­ments with collaborative writing and exams, led us even more obviously to a radically different model of creativity than we have now, one more in tune with feminist models of collabora­tion  than the cur­rent myth of creativity as the product of the isolated male.

Written Transcript

                  Twelve per cent  of the seniors, one of the freshmen and one of the graduate students felt that the primary advantage of INTERCHANGE is that it provided the instructor and all members of the class with a written transcript of the class discussion. However, forty-two per cent of the seniors, com­pared to only three of the graduate students felt this was the second most important feature of the program. The seniors valued it more highly probably because they were required to write analyses of the transcripts and thus read them more carefully. The transcript allows everybody to see what stu­dents in other conferences are talking about and makes class discussion much more meaningful in some ways. It encourages students to think seriously about significant contributions which can be used as the kernels of future essays, for they know in advance they will have transcripts of their contribu­tions. It also makes the brainstorming more productive, demonstrating how it works, and allowing it to continue after the class has ended. In the literature class, huge transcripts resulted from our brain­storming, reminding us once again that in general computers encourage us to produce more text, a common experience in computer-based composition classes (Payne, 1987, p. 26).

The Effect of Pseudonyms

                  Remarkably ev­ery one of the freshmen and 80% of the seniors liked the use of pseudonyms. Six out of 15 freshmen  respondents even felt that pseudonyms should always be used, though the true authors should be re­vealed eventually. The remaining 9 of 15, like 64% of the seniors, felt that pseudonyms could be used on occasion, but also wanted the true authors revealed eventually. Pseudonyms were more popular with the freshmen probably because theirs was a small, intimate class where they got to know each other only too well in some respects, while the senior course was so large many people knew only a few other members of the class anyway.

                  Twelve out of 14 freshmen respondents and 20% of the seniors felt that the primary advantage of the use of pseudonyms was that they allowed students to try out different roles and arguments with­out being charged with inconsistency (9, 17%) or insincerity (3, 3%). This was particularly valuable in dis­cussions of controversial, often stereotyped subjects, such as sex roles and biases, where dissenting voices could be easily subjected to peer pressure. Three felt the primary advantage was that pseudonyms prevented stereotyping of each other. Students commented, "anonymity removed the built-in prejudice of sex -- something I didn't realize affected me until this experiment"; pseudonyms allowed us to "send out messages to others [with whom we] normally do not communicate" (see Sproull, 1986), and "to be daring and elicit stimulating responses"; this method "increased adherence to the text...There was an unprecedented quantity of quotes to back up most assertions. [Pseudonyms] en­hanced our rhetorical skills." On the other hand, especially for some graduating seniors, pseudonyms encouraged more playful, irreverent, and personal  versions of class discussion, less tied to the text.

Emotional Honesty vs. "Flaming."

                  Because the primary emphasis of the senior course was on expression of emotion it is not sur­prising that 58% of the seniors, compared to only 5 of the freshmen, felt that the primary advantage of pseudonyms was that they allowed true feelings to be expressed without fear of future recriminations, an option not available in the normal classroom, where one immediately identifies who is speaking, and, as one put it, "might either be afraid to hurt someone else's feelings or ...just be embarrassed." 

                  Unlike Sproull and Kiesler (1986) our results suggest that even with pseudonyms computer-as­sisted discus­sion of literature at times increases individuality and self-disclosure in apparently healthy ways.  Stu­dents commented, "not only were the students able to communicate with one another more easily but I saw that using the computers the students seemed to be a littler freer to share what was running around in the brain, especially when we use our numbers [pseudonyms]"; "I was able to ex­press myself a bit more freely, for some reason, and appreciate more the criticism and praise that I got"; "the hon­esty and strength of the messages are enhanced."

                  Not everyone liked the freedom, however: one of the seniors complained that "the pseudonyms, though they were well worthy trying, took way from the discussion on the book and other relevant issues. People tended to really get into their own personal problems with their boyfriend, etc. and not talk about the novel at all." Of course, the goal of the senior course was to elicit more emo­tions and personal relevance than most courses. Hence tests with this class demonstrated the irony that pseudonyms allow students to be more, rather than less, personal.

                  Most people liked the greater access to emotions in CACD: one of the freshmen noticed that  "we did not really hold back our emotions [for] fear of what the rest of our classmates may think of our logic ....we really don't have to take time out to [be] eloquent so as to not hurt anybody else's feelings." A senior said, "My favorite aspect of this class was the computer-assisted discussions. This allowed me to discuss my feelings more freely. It is hard for me to tell my feelings and emotions to a bunch of peo­ple. Also, I liked the use of the pseudonyms better; it made me feel safer in expressing my feelings."

                  Of course as with any release of emotion, we can not always pick and choose which emotions come to the surface. Nevertheless, I felt it was better to know about hostilities, say, between the sexes, rather than pretend they weren't there in normal class discussion. One person praised the "seemingly more honest discussion [but noted that] hostility also seemed more rampant." There certainly were some parallels with "flaming," though the connotations were not always negative: [INTERCHANGE] "helped to get people say things that they might have been nervous to say out loud"; "there is no reason to be afraid to say exactly what you think"; "people are less inhibited"; "People seem a lot more vehe­ment"; I felt free to express my feelings "as bluntly and concisely as I wanted. When I write under my name, though, I always soften my opinions and fill in my messages with little phrases that lessen the impact of my response. These include stating that it is only my opin­ion (which is superfluous ...) and stating specific conditions so that my response can't be shot down [as] broad generalizations."

                  Pseudonyms also allowed us to set up a debate about the opposite sex in which each person was asked to argue for a position which was the opposite of the one he or she re­ally believed in. Stu­dents commented, "this exercise made us step into our opponents' shoes to see what he sees and why he sees it....it taught me to look at both sides ... before staying on one side"; "forcing yourself to take a look at a reasonably controversial topic from a different point of view gives you a better understanding of your own position on issues....I tried to agree where I would normally contra­dict, be delicate where I would usually antagonize and suggest without being arrogant." One student suggested that "the class could vote on which team won each debate after reading the transcript."

Experiments in Conference Organization

                  Most students preferred  introvert/extrovert conference organization segregated by Myers-Briggs personality tapes, that is, introverts in one conference and extroverts in another. Most students also preferred the share, not compare rule, and 42% preferred it with pseudonyms. Introverts felt most strongly about this; one student said, "The interchange between the students was much more open with the computers, especially for the introverted members of the class, including myself." In fact, at one point the introverts thought an extrovert had entered their conference and stated their re­sentment. Others found the share, not compare rule even more effective; one student wrote that "It gave me more freedom to express what I felt about the novel even more so than using the pseudonyms." However, one of the students in a control group composed of both extraverts and intro­verts complained that "the share not compare rule didn't work that well for me, but that might just be because I was in the mixed conference. Because the introverts and extroverts were mixed together there was a personality conflict and I couldn't discuss with my conference because there was nothing we agreed on."

Effects of Gender Segregation

                  When conferences were organized by sex some people were liberated but the stereotypes be­came quite clear. With all the men in their own conferences, some women in the freshman class felt more able to be heard: "in the new arrangement, no one person dominated the conversation. Usually John, Doug, and Franklin end up dominating the computer and even when other people comment, those are seldom read."  As this comment suggests, the gender bias toward males in CAI extends to attempts to dominate on computer networks as well.

                  In addition, one student observed that "the female conferences have a different emphasis than the male conferences." For instance, both were concerned with "the question of which sex dominates the other" but the women blamed the men for failed marriages in Lawrence's The Rainbow and the men blamed the women. Another noticed a male sense of the relation between the sexes as a "battle" and observed that in one of the male conferences "the battle was much more threatening. The whole of [that] conference was spent discussing the female dominance they found in the book."

                  The deep emotions each sex felt about the other began to be identified: "One of the most in­teresting things which came out in the conference was certain guys' actual fear of women, or [of] women reaching true equality." Another difference was that, as one woman put it, "It is interesting how many of the guys thought that the men in the story had the most problems. I thought it was the women who had the most trouble in the book." At times the differences seemed quite fundamental: a woman noted that "the boys are always more interested in the sexual aspects of what we read"; a man observed that "the women .. found a common bond in childbirth ... something that makes them dis­tinctly different form men. Childbirth is very painful for women and gives them a feeling of satisfac­tion upon completion." In Jane Eyre gender seemed to determine the degree of identification with the central character. One male wrote, "The females really seem to relate to Jane like [she] could re­ally be them" because the ability of females "to be something on their own" is "doubted," whereas the men can be more "objective" because none of them "have probably ever had to defend the fact that they could make it on their own; we are expected to."

                  When we had a session on modern women poets, one student wrote, "the thing that struck me the most was the differences in the way men and women treated the subject. I had to guess, of course, which ones I thought were male or female, but it seemed pretty obvious....The guys tended to disre­gard what the female poets said, and dismiss some of it as gross exaggeration."

                  Although there were some strong feminists in the class, and the class discussion usually re­volved around very controversial assertions about sex roles, every one of the freshmen and, what is more remarkable, all thirty-three of the seniors voted for conferences to be balanced between the sexes rather than segregated by sex. Though some women said they were liberated when the more ag­gressive males were segregated by gender, most students argued for "letting the males and females trade ideas directly"; one added, because then "the discussions could be substantially more heated."

                  The issue of gender bias or sex role stereotyping was never raised in the graduate class, where the composition of the conferences depended on student interests rather than gender. Nevertheless, in this class as well everyone voted for conferences to be balanced between the sexes, if the teacher were to assign people in advance.

Gender and Pseudonyms

                  In the mixed sex conferences, the use of pseudonyms helped prevent stereotyping, "forcing me to look for clues in the statements to determine sex rather than just glancing at the name."  However, when the students were asked to argue for their opponents' views, "seeing things through the eyes and minds of those of the opposite sex," one student commented that "some people simply could not hide their gender, even when advocating the views of the opposite sex." Others were liberated. A particu­larly macho male liked the pseudonyms because they gave people the "ability to act however they want, be whatever sex they want."

Direct Reader Response Experiments

                  Another experiment which produced remarkable results was the computer-assisted reading assignment. In almost every case, this format slowed students' reading so much, making them ponder the significance of each individ­ual word, that it soon became apparent that few would even finish reading a 60 line poem in 75 minutes of class time! Some students liked the new method right away, but others became desperate and got out their anthologies, found the poem in the book, and were thus able to oscillate between their tradi­tional mode of reading, where the entire poem could be seen at once as a whole, and the new format. This technique has many implications for experiments in reader-response criticism (the number of lines of text visible can be easily varied in INTERCHANGE), and for those who argue the need to read more slowly (Sire 1978). When the freshmen were asked their opinion of this technique, which, like CONTACT, we only tried once, 12 of 17 voted to do more of that sort of reading, 7 preferring shorter selections from texts we have already read, 5 voting for shorter selections from new texts. And of course we can go on to ex­periments in which we let each student change the text, that is, revise it, interpolate his or her own comments, etc. (Evans, 1988). Then we can  compare the changes with those of other students, creating a composite text incorporating their per­sonal, emotional reactions as read­ers. Beyond that we have the whole world of hypertext and interac­tive fiction opening up before us (Bump, 1987, p. 129) as we move beyond page-bound reading alto­gether.

Collaborative Exams and Essays

                  Perhaps the most radical experiment, however, was the collaborative exam. After taking col­laborative exams at the end of the course students in E603B and E388 and the option of voting to have no more exams on computers, but no one selected it: every freshmen and 8 out of 11 graduate respon­dents voted to have all exams be of this type!

                  As Batson suggested (n. d. 14), CACD also makes collaborative essays possible, and thus the students were asked the general question, "should one or more of the essay assignments be replaced by essays composed as a result of collaboration," and given five options. 9 of 16 freshmen and 7 of 12 graduate students voted for essays to be composed in advance and merely revised in response to sug­gestions by others, either in INTERCHANGE or CONTACT, the technique used in our other fresh­men computer-assisted classes. One freshman wrote, "I think people should be able to do their original work, but have the benefit of people to comment on what they have written. That helps a lot. ...I'm ambivalent about writing papers on the basis of INTER­CHANGE, maybe for inspiration, but not from scratch based on other people's input. I wouldn't like the group to get together and write an essay."  However, 6 freshmen and 3 graduate students did vote for the formal essays to originate in, that is, be inspired by, conference sessions in INTERCHANGE and develop from the transcripts of the sessions. In all these options, grades were to remain individ­ual. Only one freshman and two graduate students voted for essays to be produced from start to finish by the conferences and the individuals' grades to be those of the conference as a whole, though three freshmen and two graduate students voted for this option as their second choice. Obviously there is great resistance in our culture to any threat to indi­vidual grades in education; as one graduate student put it, "anything but group grades!" 

                  Nevertheless, however applied, the basic electronic network of CACD is a model of the kind of non-hierarchical, multivoiced, broadly inclusive, collaborative creativity supported by feminists and others who emphasize egalitarian cooperation within groups more than competition between individu­als. Further experiments with collaborative reading and writing may well challenge our most basic concepts of authorship.

Disadvantages

                  When it came to identifying the disadvantages of INTERCHANGE some students found no disadvantages and thus the numbers on the answers to the questions no longer come close to adding up to the total number of students. 12 freshmen at the end of the first semester, for instance, found no "primary" disadvantage of the program.

Reliance on Keyboarding                

                  One of the most serious limitations of computer-assisted-discussion is the reliance on key­boarding, a much less widespread skill than speaking. Four first-semester freshmen, nine the second semester, and 46% of the seniors felt the primary disadvantage was the dependence on typing rather than speaking skills.  However, at the end of the second semester the number of freshmen who typed not only class essays, but other materials as well, had doubled. As only two students had taken a typing class, presumably much of this improvement can be attributed mainly to the use of INTERCHANGE in class. The grad­uate students, who tended to be better typists, did not regard this is an important dis­advantage, though even these  good typists spoke more quickly than they could type. 9 out of 12 had a typing course, and all at least typed their papers for class. 5 Always typed, even their notes. 

Insufficient Voice Communication

                  One first-semester freshman, two second-semester freshmen, and 38% of the seniors felt that the primary disadvantage of the computer classroom, if not the program, was that it made voice com­munications between students difficult. This complaint was cited by the most graduate stu­dents, four out of the nine who rated the disadvantages of the program. One student was in the fresh­man class by mistake, an engineering freshman whose poor writing skills prevented him from going on to the next semester. He relied on his speaking ability to get him through the class and thus he insisted that "Voicing your opinion gets it across better.  Using a computer is also more time consum­ing, because [it is] limited to the individual's ability to type." Others pointed out that "correcting mis­takes in typing is somewhat difficult and when you're in a hurry it can be disconcerting" and that "it's hard to emphasize with the desired tone in the computer because there is no sound or vocal communi­cation involved." On the other hand, as we have seen, by preventing voice communication we eliminate accents and other paralinguistic clues which invite prejudice.

Primary disadvantage: slow speed

                 The primary disadvantage of CACD was thus its relatively slow speed compared to speaking. The problem of speed extended to reading as well as writing: "it gets frustrating sometimes when a conference gets really busy and you would have no time to type anything in if you worried about read­ing absolutely everything."  Overcrowding in a conference during the first semester of the freshman course created the big time lag problem. Hence, 10 of 17 second-semester freshmen and 66% of the seniors voted for the teacher to assign people in advance to conferences of no more than four or five people. The graduate students also experienced this problem but most of them preferred the freedom to join any conference they wished.

Less Coherence

                 However, even in a small conference there has been some frustration about the "time delay -- unlike verbal discussion, everyone can get their word in, but by the time they do the moment may be gone -- it's not relevant anymore." This weakens coherence -- "conversations so easily go off on tan­gents" -- and in this respect may make the program seem inferior to speaking: "during oral dis­cussion, ideas flow from one topic to another in a more logical, fluid manner." However, Neuwirth (1988) has shown that the focus of attention in fact is much greater in computer interaction than in face to face interaction.

                  Students seem to be comparing the coherence in INTERCHANGE to written rather than spo­ken prose. Comments refer to the loss of the "thread," to "breaks in the text," and to loss of "flow," the sorts of critiques usually made of writing rather than speaking. The result is more like a communal journal, and hence it is perhaps not surprising that the seniors, whose primary writing assignments were journals, did not protest as much as the freshmen about the less formal prose style.  

                  For those who need more coherence it may be that in this respect the absence of an instructor in constant control of all conferences at all times may seem a disadvantage: "I don't feel like there is much instructor guidance with this method of discussion; I like it best you [the instructor] put ques­tions up at the beginning to start us." One option, of course, is simply to celebrate the new kind of in­formal, heteroglos­sial prose which CACD generates, broken up as it is by many voices, by playfulness, and by irreverent creativity.

                   A better solution is to use a hypertext method of attaching responses to the comments which inspired them. An important difference between the latest version of INTERCHANGE and other programs is that INTERCHANGE can now be run in either a hypertext or a normal mode (Taylor 1989a and b). Students and teachers can choose either the traditional linear mode or the new nonlinear mode. In the linear mode messages from the participants in the collaborative session appear in strict chronological order, with the result that related messages are separated by any number of unrelated intervening messages, lessening coherence. In INTERCHANGE's new hypertext mode an electronic document is constructed in which all related messages are linked. When writing and sending a mes­sage, a student can send an entirely new message or choose the option "Reply." When she does so she establishes an invisible link between her response and the message to which she is responding and these links form a dynamic electronic network. When reading the messages she can choose the word "Hypertext" from the menu bar and then has three options to choose from: "Chronological list of all messages," "Reply thread (replies to a message)," or "Search for textual connections." The first option is the only option on all the other currently available programs. The second choice, the true hypertext option, allows her to select one message and see only the messages that have been attached to it by use of the "Reply" option. The third choice allows a search through all messages to find and link all those connected by key search terms chosen by the student. As Taylor pointed out (1989a), the new INTER­CHANGE is a response to Halasz's call for greater collaborative capabilities and improved search and query functions for hypertextual systems.

Loss of Controlling Instructor

                 One freshman and one senior felt that the primary drawback of INTERCHANGE was that the classroom made holding the usual, instructor-based class difficult. Five freshmen and 24% of the se­niors felt it was the second most important disadvantage and four freshmen and 29% of the seniors felt it was the third most important. For the graduate students, on the other hand, almost as many voted for this as the primary disadvantage as for the  loss of voice communications: three out of nine. Teach­ers may find this comforting, however, as it reveals that are occasions even in class discussion when their intervention would be beneficial. 

Dehumanization

                  Because they were allowed fewer opportunities to get to know each other in ordinary class dis­cussion, the graduate students focused more on the dehumanizing aspects of the paucity of voice communications and the intervention of the machinery between people: "What is written often does not reflect personality in the same way that face to face conversation does." When the graduate stu­dents made their problem clear to me, I had the class meet once in the student union and one student's comment was: "I think it was a very good idea for the class to meet at the union so that we get to know each other. After that meeting,the monitors did not seem to get in the way of communicating with real people." One would think this problem would have been worse in the large senior course but the alter­nation with small face-to-face groups apparently alleviated the problem; one senior wrote, "I strongly disagree that the code-name interchange contributes to a faceless university. In fact, this class seems to do the opposite -- the class is much more open and free, which for me encourages a better spirit of learning."

Technostress

                  We assumed technostress would be a problem primarily for E376L. Where the previous courses had eighteen and nine students respectively, this course had thirty-three students of widely varying backgrounds. Most of the students were seniors yet 33% of them had never used a com­puter before for verbal rather than mathematical experience, and 12% had done so only infrequently. Nev­ertheless, instead of experiencing technostress, most students in this class felt that, as one put it, "the computer network is an interesting new mode of emotional analysis" which releases stress. 

                  In the graduate class three students had used a computer frequently and two had a symbiotic relationship with it, and these numbers doubled by the end of the course. Hence one would expect that "technostress" would not be a problem in the graduate class. One student wrote, "I didn't feel much technostress, perhaps because I have had three programming courses before this one." However, even in this class we experienced a lot of overload, partly because we were learning  new computer systems at the same time we were trying to meet the ambitious goals of a Holistic humanities computing course: learning dozens of new computer programs  at the same time we were trying to cover our sub­ject in more traditional ways and write individual software or research papers. In fact two of the origi­nal fourteen graduate stu­dents had to withdraw from the course and technostress was identified as a problem for ten out of the twelve who survived.  One of the two who did not experience technostress throughout the course, a college professor, sud­denly succumbed to it during the collaborative final exam and had to leave the room, unable to re­turn. Trying to switch from a Macintosh to an IBM he was not able to do what he thought he would be able to do and when the clicking of the keyboards of the other students became a clatter he panicked and gave up, an example perhaps of how dependent we can become on a particular computer or pro­gram (see Heim, 1987, p. 151). Hence our experience with technostress in this particular course rein­forces Ide's caveat that "increasing the amount and difficulty of the material in such courses may ini­tially discourage many humanists from computer use" (213).

                  Fortunately, technostress was one of the subjects of the course and it was a great relief to be able to discuss the problem in class. In fact the discussion of this concept and the philosophical aspects of wordprocessing was judged by the students to be more valuable than learning about the design and programming of simple word processors, or using and evaluating various outliners, thought organiz­ers, and advanced word processors. Another way to deal with technostress, of course, is to follow Her­rman's suggestion of separating the teaching of the particular computer systems used in the course from the teaching of the subject matter (1987).

Room Arrangement

                  Finally, there were some comments about room arrangement, especially the inability to see the faces of most of the other students due to the arrangement of the hardware in rows in the com­puter classroom. Many students felt that "The arrangement of the room has very little to do with the class because there is no speaking or visual contact necessary." However, three stu­dents suggested that the workstations be arranged in a circle or square so that one could look at the other people and more personal communication could develop. Xerox's Colab room is a possible model here. One graduate student recommended that the machines be placed in a circle and that stu­dents have higher chairs so that they could look over the monitors at one another. Of course another solution is to lower the mon­itors, embedding them partially in the table. A senior wrote, "it would im­prove the environment if there was/were one big circle or several small circles ... (a circle for each conference). This would make the sessions 'feel' a lot more friendly and, I believe, induce more emo­tional thoughts."  Another wrote, "I would prefer a more modern setting with greater outside exposure. In fact I would be happier in a glass enclosed setting, exposed to the outside world." Such a room may in fact be more important for a computer-assisted class than for a normal class to prevent tunnel vision and to lower stress and fears of dehumanization.


 

                        Works Cited

Arms, Valarie Meliotes. "Collaborative Writing with a Computer." The Technical Writing Teacher, 11 (Spring 1984), 181-5.

---. "Engineers Becoming Writers: Computers and Creativity in Technical Writing Classes." In Writ­ing at Century's end: Essays on Computer-Assisted Composition. Ed. Lisa Gerrard. New York: Random House, 1987. pp. 64-78.

Balester, Valerie, with Kay Halasek, "Sharing Authority: Collaborative Teaching in a Computer-Based Writing Course." In Proposal Abstracts from The 5th Computers and Writing Confer­ence, University of Minnesota, Minneapolis, May 12-14, 1989. Ed. T. Batson. Wash­ington D. C.: Gallaudet University, 1989. pp. 4-6.

Batson, Trent W. "Computer Networks in the Writing Classroom." Unpublished essay, n.d. HMB 120. Gallaudet University, Washington, D. C., 20002.

---- "The ENFI project: A Networked Classroom Approach to Writing Instruction," Academic Com­puting (February, 1988): 32-3,55-6.

---- ed., Proposal Abstracts from The 5th Computers and Writing Conference, University of Minnesota, Minneapolis, May 12-14, 1989. Washington D. C.: Gallaudet University, 1989.

Bump, Jerome. "CAI in Writing at the University: Some Recommendations."  Computers and Edu­cation, 11,2 (1987), 121-33.

Butler, Wayne. "A Report on Student Attitudes." English Department Computer Research Lab, Uni­versity of Texas, Austin, 1988. Unpublished MS., 33 pp.

-----"The Construction of Meaning in an Electronic Interpretive Community." In Proposal Abstracts from The 5th Computers and Writing Conference, University of Minnesota, Min­neapolis, May 12-14, 1989. Ed. T. Batson. Washington D. C.: Gallaudet University, 1989, pp. 8-9.

Carter, Locke, "Telecommunications and Networked Personal Computers: Opening Up the Class­room," Conference on College Composition and Communication, Seattle, Washington, March 17, 1989.

Evans, John F. "Reader-Response Pedagogy and Computer-assisted Composition." Computers in Writing and Language Instruction Conference. Duluth, Minnesota, 2 August, 1988.

Lester Faigley, "Subverting the Electronic Workbook: Teaching Writing Using Networked Comput­ers," The Writing Teacher as Researcher: Essays in the Theory of Class-Based Writing (Upper-Montclair, N.J.: Heinemann-Boynton/Book, forthcoming).

Feldman, Paula R., and Buford Norman. The Wordworthy Computer: Classroom and Research Applications in Language and Literature. New York: Random House, 1987.

Friere, Paolo. Pedagogy of the Oppressed. Trans. M. B. Ramos. New York: Seabury Press, 1968.

Gerrard, Lisa. ed. Writing at Century's end: Essays on Computer-Assisted Composition. New York: Random House, 1987.

Halasz, Frank, "Reflections on NoteCards: Seven Issues for the Next Generation of Hypermedia Sys­tems," "Hypertext 87": Proceedings of a Conference at Chapel Hill, N. C. Department of Com­puter Science, University of North Carolina, 1987.

Heim, Michael. Electric Language: A Philosophical Study of Word Processing. New Haven: Yale Univ. Press, 1987.

Herrmann, Andrea. "An Ethnographic Study of a High School Writing Class Using Computers: Marginal, Technically Proficient, and Productive Learners."  In Writing at Century's end: Essays on Computer-Assisted Composition. Ed. Lisa Gerrard. New York: Random House, 1987. pp. 79-94.

Hiltz, S. R., and M. Turoff. The Network Nation: Human Communication via Computer. Reading, MA.: Addison-Wesley, 1978.

Kaplan, Nancy. "As we May Teach: Some Problems with Computer-Supported Collaboration in the Writing Curriculum," Proposal Abstracts from The 5th Computers and Writing Conference, University of Minnesota, Minneapolis, May 12-14, 1989. Ed. T. Batson. Washington D. C.: Gallaudet University, 1989, pp. 49-50.

Kemp, Fred. "Computer-Based Collaborative Writing Instruction Without a Computer Network," Proposal Abstracts from The 5th Computers and Writing Conference, University of Minnesota, Minneapolis, May 12-14, 1989. Ed. T. Batson. Washington D. C.: Gallaudet University, 1989, pp. 50-51.

Ide, Nancy M. "Computers and the Humanities Courses: Philosophical Bases and Approach." Com­puters and the Humanities, 21 (1987), 209-215.

Jennings, Edward M. "Paperless Writing: Boundary Conditions and Their Implications." In Writing at Century's end: Essays on Computer-Assisted Composition. Ed. Lisa Gerrard. New York: Random House, 1987, pp. 11-20.

Juncker, Clara. "Writing (with) Cixous." College English, 50,4 (1988), 424-36.

Kelly, Erna."Processing Words and Writing Instructions: Revising and Testing Word Processing In­structions in an Advanced Technical Writing Class." In Writing at Century's end: Essays on Computer-Assisted Composition. Ed. Lisa Gerrard. New York: Random House, 1987, pp 27- 35.

Kraemer, K. L. and J. L. King.  "Computer Supported Conference Rooms: Final Report of a State of the Art Study." Dept. of Information and Computer Science, Univ. of California, Irvine. Dec. 1983.

Langston, M. Diane. "Invention Aids for Computer-Based Writing: Expanding the Horizons through Collaborative Invention." ERIC, 1987. ED 280 055.

Logan, Shirley W. "Social Interaction Among Writers, Tutors, and Teachers in a Writing Computer Lab for Undergraduates," Proposal Abstracts from The 5th Computers and Writing Conference, University of Minnesota, Minneapolis, May 12-14, 1989. Ed. T. Batson. Washington D. C.: Gallaudet University, 1989, pp. 55-6.

Marcus, Stephen. "Computers in Thinking, Writing, and Literature." In Writing at Century's end: Essays on Computer-Assisted Composition. Ed. Lisa Gerrard. New York: Random House, 1987, pp. 131-40.

---. "Sexism and CAI." Computers, Reading, and Language Arts 1,2 (1983).

Neuwirth, Christine. "Effects of Computer-Mediated Collaborative Writing." Computers in Writing and Language Instruction Conference. Duluth, Minnesota, 1 August, 1988.

Oakman, Robert L. "Perspectives on Teaching Computing in the Humanities." Computers and the Humanities, 21 (1987), 227-33.

O'Connor, John. "Teaching Collaborative Writing on a Computer," Proposal Abstracts from The 5th Computers and Writing Conference, University of Minnesota, Minneapolis, May 12-14, 1989. Ed. T. Batson. Washington D. C.: Gallaudet University, 1989, pp. 65-6.

Payne, Don. "Computer-Extended Audiences for Student Writers: Some Theoretical and Practical Implications." In Writing at Century's end: Essays on Computer-Assisted Composi­tion. Ed. Lisa Gerrard. New York: Random House, 1987, pp. 21-6.

Peterson, Nancy L. "The Sounds of Silence: Listening for Difference in the Computer-Networked Collaborative Writing Classroom," Proposal Abstracts from The 5th Computers and Writing Conference, University of Minnesota, Minneapolis, May 12-14, 1989. Ed. T. Batson. Washington D. C.: Gallaudet University, 1989, pp. 6-8.

Rich, Adrienne. On Lies, Secrets, and Silence: Selected Pr