Regarding the use of computers in composition, one of the more persistent calls put forth in articles in College English, CCC, Computers and Composition, and on BITNET lists like MBU and WPA, and in conference presentations, workshops -- both local and national -- and even in casual conversations in elevators and hallways, is a call for proof.
Proof.
People want proof about the following:
In short, we want to discover the universal efficacy in a process, lay it out before the community untainted by the debilitating quirkiness of special enthusiasms, pure in some clinical way that asserts, unequivocally, that this thing works here, uninfluenced by special conditions, and it will work anywhere with anybody. Guaranteed.
We can do that with many of the things and processes we value in society. We can do that with air conditioning systems, vaccines, rockets, voting procedures, military training, and automobile repair.
But we haven't been able to isolate universally effective procedures for reforming criminals, rearing-children, making marriages last, rehabilitating drug users, and educating our children. This last is what we in this room are most professionally concerned with. We come to meetings like this in order to determine with as much certainty as we can what will work and what won't in our classrooms. And for this we want proof, or the closest thing to proof possible.
But I suggest that in writing instruction, and in any kind of instruction, actually, it is infuriatingly difficult to prove anything worth proving. All too often the evidence put forth in order to assemble the trappings of empirical proof are rhetorical devices that convince the already 80 percent convinced and are ignored by those already convinced otherwise.
Let me explain my point using several examples.
For at least twelve years in our small academic community of computers and writing, people have been trying to prove that students do or do not write better on a computer. Many of the studies transparently reflect the prior attitudes of the writers, no matter the obligatory nod toward empirical methods, either qualitative or quantitative. The subject of computers, especially in English departments, often brings to the surface deep-seated animosities that reflect entire wormbeds of Huxlean technophobia and Arnoldian literary essentialism, and it is hard for these to hide beneath the patina of "proof" no matter how thick the writer lays it on.
But as we in computers and writing have been diligently piecing together a range of attitudes about whether people write better on a computer or not, almost all people in the industrial world who do any regular writing have come around to using a word processor. They don't know about the efforts of academics to prove whether writing on a computer is better or not, or if they did know about those efforts, they would have ignored them. That fact alone, which any of us can observe in any office place in the country, should support what anybody who uses a computer for writing already knows, that for most people it is easier and more productive to write on a computer.
Of course, to scholarly researchers, ease and productivity may not be convincing characteristics. Ronald Sudol, in an article in College English in 1991, argued against word processing. "The ease of production leads to voluminous output, everything in the writer's mind gushing without the monitoring and regulation that a slower, more laborious writing process imposes" (921). Mr. Sudol's conclusion would suggest that computers make writing too easy, and that the key to encouraging clarity, brevity, and specificity is to impose a "more laborious writing process." One could, of course, carry this reasoning back to stone chisels and rock walls.
In spending so much time trying to determine in some abstract academic sense whether writers write better on computers, we have completely ignored what I am calling the "Look Around You" factor. Who else but an English professor would suggest that a process that makes writing harder would generate better writing?
The magazine Academic Computing, in 1990 described a study at the University of Delaware which drew the conclusion that students write better on IBM compatibles than they do on a Macintosh. The reason? Macintoshes used a graphics user interface, more commonly called a "point and click" interface, while IBMs, at that time, used a command line interface, which meant that the user had to remember commands or keystroke combinations rather than just move the mouse and point and click at items on a menu. The Macintosh interface, the article reported, did not encourage seriousness in the writer, an argument very similar to that used by Professor Sudol. If the researchers at Delaware had employed the "Look Around You" factor, they would have realized that, for better or for worse, all microcomputing was soon to adopt the point-and-click, graphics user interface. On the IBM this is called Windows. Whatever Academia may conclude about a Mac interface or windows interface, the vast majority of computer users have come to demand it in their computers. And so, as our rapidly expanding body of ancient computers rust and sigh their last, so too will the command-line interface. The question of whether students write better using it instead of a mouse -- which in my heart of hearts I would deny but have no means of proving -- has, blessedly, become moot.
Now the search for academic proof has shifted to the use of computers in classrooms and in instruction, as if instructional activities are qualitatively different from occupational activities. "Granted that almost everyone who needs to write uses a computer to write" -- the argument goes -- "but that doesn't mean that computers have a place in the classroom." The implication here is that writing instruction doesn't involve writing, that whatever it is we do in the classroom, it is qualitatively different from whatever it is that people do when they actually write in life. This doesn't make sense to me.
In preparing for this paper I called 13 people on the telephone and contacted 21 others by electronic mail. In so doing I established one incontrovertable fact. It is a heck of a lot easier to get people on the Internet that it is to get them on the telephone.
I wanted to acquire a more in-depth examination of what instructional emphases people were using when they used Interchange than was possible in our national survey. I'd like to provide you some general conclusions, which I freely admit are my interpretations of a number of heterogenous comments.
While it is undoubtable that many people become highly enthusiastic at the concept of ENFI and the kinds of electronic real-time discussions that Interchange supports, I have been receiving indications for a while now that many people who buy Interchange actually employ it more sparingly in their computer-based classrooms than one might suppose, and, when discussing their use of synchronous messaging, find it difficult to explain what value it is in any theoretically consistent or powerful way. That is, they almost always report (though not always!) that students love the discussions and the classroom activity level rises considerably, but they have trouble linking such attitudes and behavior to an articulated instructional benefit.
At places like Texas Tech and the University of Texas at Austin, where there is a long history of ENFI and Interchange and where experienced cadres of mostly graduate students have evolved over time, there seems to be no doubt about the efficacy of network-based writing instruction and both synchronous and asynchronous messaging in the classroom. But in these schools there have evolved genuine human support systems, groups of basically younger professionals who have no stake in long established methods and who feed off of the enthusiasms of their peers, who protect themselves from the difficulties and disappointments inevitable in adapting radically new pedagogies by sharing their efforts regularly with sympathetic colleagues, both locally and through the Internet.
In other places, the major problems deal principally with control issues. Those who love Interchange speak glowingly of being forced to re-examine the nature of writing instruction, being forced to follow the discursive authority of their students, being forced to stay out of the limelight, and so forth. Those who had serious doubts about Interchange or downright hated it, and this group was only about a sixth of the respondents, but that percentage is highly suspect since except for several cases the respondents were not anonymous, basically didn't like it for exactly the same reasons as those who liked it; it forced them to rethink or react in ways they just didn't want to rethink and react regarding questions of what is productive teacher and student behavior. Essentially, like Ronald Sudol and the Delaware study, they don't like the apparent chaos or open flow of what happens in Interchange, what Shoshanna Zuboff calls the "post-hierarachical relationships" implicit in networked communications. To them, hierarchy means controlled, incremental progress. Interchange seems to suggest that the students are always just about to break out of the hold and take over the ship. Consider some of the comments:
In an article in College Composition and Communication in February of 1991, Gail Hawisher and Cindy Selfe defined what they called the "rhetoric of technology and the electronic writing class" as, in part, "the enthusiastic discourse that has accompanied the introduction of computers into writing classes" and bemoaned "the uncritical enthusiasm that frequently characterizes the reports of those of us who advocate and support electronic writing classes"(56). In their presentation at CCCC in San Diego last year, both Hawisher and Selfe repeated the need for a more critical view of computer-based writing instruction by those who profess its virtues. At the 9th Computers & Writing Conference at Ann Arbor in May of last year, Dawn Rodrigues criticized what she saw as a dominating influence of network theory on the field of computers and writing, calling for individuals to develop their own methods of using computers and not be intimidated by the largely unsubstantiated enthusiasm of networkers. In that same session, Pat Sullivan called for a more rigorous and yet varied research methodology in our field so that we would not be subject to broad, unsubstantiated claims.
To some degree these calls for a more critical review of what we are claiming -- and we've experienced them here in Nashville -- are similar to that clearly put forth by my friend and colleague at Texas Tech, Patricia Goubil-Gambrell, in a 1992 article in Technical Communication entitled, "A Practitioner's Guide to Research Methods." Although she was describing the field of rhetoric and composition and not just computers and writing, I think her concerns are pertinent.
In trying to encourage what she calls a "methodological literacy," Professor Goubil-Gambrell defines four types of research on writing. The first two, which she terms "empirical methodologies," quantitative and qualitative, "come to writing research primarily by way of the social sciences and through educational research." The other two, scholarly inquiry and practitioner inquiry, come by way of the English department, and those terms she has borrowed from Steve North's The Making of Knowledge in Composition. She states that "one type of research is not necesarily better than another," but out of approximately 910 lines of her text she devotes only 30 lines to scholarly inquiry and practitioner inquiry and over 870 lines to qualitative and quantitative methods. She lists the characteristics of productive research as "credibility," "transferability," "dependability," and "confirmability," and I think she and most of her readers would agree that, based upon that criteria, empirical methods succeed much more than scholarly or practitioner inquiry.
The purpose of methodological clarity and consistency in a discipline is, as she says, to "let one know how much a given piece of research can be relied on." To return to computers and writing, then, the boundary between a "critical" and an "uncritical" enthusiasm is drawn by a presumed universally agreed upon set of criteria for what claims "can be relied on." Underlying all these calls for a greater methodological consistency in research is an implicit or explicit desire to support computers or writing or rhetoric and composition as a clearly defined academic discipline with a discipline's usual political stability and clout.
Regarding these statements about the need for methodological consistency and a more self-critical stance, I completely agree.....almost. But in the qualifier "almost" there lies a characteristic of progress in our field that I feel needs to be emphasized. In a sense, I am juxtaposing my own word of caution against my colleague's words of cautions regarding what constitutes valid research.
(We in computers and writing are a cautious lot, as you can tell from Trent's presentation.)
The problem is that much of what changes teaching in the humanities, at least, does not follow any clearly articulated methodology or rules of evidence. That which changes our teaching often comes as an idea that strikes us like a bolt of lightning. When I watched Trent discuss ENFI at the 1988 CCCC in Atlanta -- and I was a classroom teacher of 15 years experience -- the sheer sense of what he was describing was inescapable. Later that spring, at a small conference in Corpus Christi, I heard Wayne Butler, Kay Halasek, and Valerie Balester describe classroom collaboration and the pedagogy of Kenneth Bruffee, and again, the sheer sense of what they were describing in terms of what I knew about networks seemed, to me at least, irrefutable. I have heard numerous testimonies to the same effect regarding ENFI and networks. Conviction, but not as a function of proof.
Sharon Crowley, in her 1990 book The Methodical Memory: Invention in Current-Traditional Rhetoric, reminds us of Aristotle's distinction between scientific demonstration, dialectical certainty, and rhetorical conviction, and makes the claim that much of what has reduced the academic essay to a moribund current-traditionalism has been too great a reliance on scientific proof and formal logic and an abandonment of ethos and pathos, or arguments from authority and emotion, that, together with logic, compose the means of securing rhetorical conviction. One might see in the term "enthusiasm," even "uncritical enthusiasm," an element of rhetorical pathos, and if transmitted, a valid means of securing conviction and change, and disciplinary progress.
Often, perhaps more often than we realize, we change what we do in the humanities because of our enthusiasms, not because somebody has proved something. It is the ideas we experience as truth in our own minds, in our own convictions, that transforms our lives as teachers, not necessarily the proofs assembled under specific rules of evidence. As Thomas Kuhn has made clear, by specifying a single valid method of inquiry, a science excludes as much as it includes within its scope of inquiry. Each successive paradigm of inquiry finds entire realms of suddenly significant data that had been ignored under the rules of evidence of previous paradigms. This has most recently been made clear by choas and complexity theory, whose paradigm of inquiry has revealed vast and significant aspects of nature almost totally invisible to previous means of discovery.
In my voice discussions and e-mail with those who use Interchange, the persistent impression I got was of people who believed firmly that classroom email and electronic discussion stimulated a major improvement in their instruction, but at the same time they apologized for being so unable to couch their convictions in objective terminology, in effect for being so uncritically enthusiastic, as if their own convictions, floating free of some scientific rules of evidence, were not only negligible, but just plain wrong.
As Patricia Goubil-Gambrell said, "one type of research is not necesarily better than another," referring to her four types. We must truly believe that and maintain practitioner inquiry and rhetorical processes as valid means by which progress is secured in a field, even if by so doing we forfeit a disciplinary clarity that would enhance our professional status in Academia. Frankly, we needn't go out of our way to incorporate our own cautious brakes on the process of change: there is a vast, human brake out there constructed of generational caution, bureaucratic defensiveness, and institutional timidity that will more than ensure that we all aren't dragged over the cliff of radical progress. But, without explicitly valuing at least some of our uncritical enthusiams, we stack the game in favor of the comfortable forces of habit and tradition, forces that may bleed the possibility of change to death before it has a chance to demonstrate itself.
(I invite comments on or responses to the preceeding essay. I will link to such responses from this page for others to read, if the sender requests me to. Fred Kemp, ykfaq@ttacs.ttu.edu)