[a somewhat shorter version of this piece is published in Conversations about Writing: Eavesdropping, Inkshedding, and Joining In, ed. M. Elizabeth Sargent and Cornelia Paraskevas. Toronto: Nelson Thompson, 2005. 134-141.]

We did this in our classes by asking students to freewrite in response to a shared experience -- a reading, a class discussion, an event -- and then arranging to have the freewritten texts passed around for reading, asking the readers to mark with a vertical line in the margin passages in which the writer said something "striking," something that seemed to them interesting or new or outrageous. We then arranged for the passages marked most often to be transcribed, photocopied, and distributed at the next class meeting.

It was Jim who came up with the word "inkshed," having found it in the Oxford English Dictionary. It wasn't exactly a description of what we were doing; the Oxford offers it as a "humourous" word, meaning "the shedding or spilling of ink; consumption or waste of ink in writing." The word was had been used earlier by Marvell and Sterne, but our favorite quotation was from Carlyle: "With no bloodshed . . . but with immense beershed and inkshed."

The ways in which inkshedding functions and the ways in which it has been used and modified in varying situations have grown and changed profoundly, of course, in the decades since then. In some ways, in fact, it has turned out to be a reasonable synechdoche for my general stance as a teacher, which has been focused on finding ways to foster growth in literacy by making it more directly and perceptibly instrumental. Inkshedding turned out to have a rich range of implications, many of which we hadn't anticipated at all.

For me, maybe the most important of these implications was composing text in such a situation increases the likelihood -- even with novice writers -- that the text will be formed with an anticipation of audience. As oral language is uttered, it is shaped in large part by the need to address a particular interlocutor or discourse situation. James Britton called this "shaping at the point of utterance." Like speech, writing, it has always seemed to me, needs this pressure; inkshedding can provide it. It's not always true the first few times a group or individual engages in inkshedding, but over a few experiences you can begin to see telltale evidences of the way audience pressure (which the writer may or may not be conscious of) affects the shape of the text, in things like clarity of handwriting and increased explicitness of reference.

It's also important that during inkshedding ideas, positions, and questions which would not otherwise attain a hearing have a better opportunity to get "on the floor" than they would in an oral discussion. A significant force in the original impetus for using inkshedding in classrooms was the perception that classroom discussions tend to be dominated by a few voices. This is natural, as the "bandwidth" for classroom discussion -- at least for whole class discussions -- is very narrow. Only one voice can be heard at any one time: for what everyone thought about an event to be articulated and discussed is not only practically difficult, even in a small class, but socially constraining: the first few utterances tend very strongly to determine and focus the range of discussion, and effectively determine the kinds of questions or issues which will be raised. Anthony Paré, in a comment on an earlier draft of this piece, says, "I've always felt that inkshedding allowed for the individual exploration of a top-of-the-head response before that response is deflected, diminished, or destroyed by the first question or comment spoken out loud. Inkshedding allows each member of the group to 'gather' her/his thoughts before they are scattered by that first, articulate, confident person who gets up to say what you weren't even thinking about."

What written discussion tends to do is broaden the bandwidth everyone in the room can, as it were, talk at once. But what is even more important is that every idea or response has a chance not only to be formed in the first place, but also to be "heard."

That "hearing" is important. What is often overlooked in this situation indeed, in any discussion of student writing is the importance of reading. Writing, of course, had been seen as central from the very beginning, in the origins of the practice as freewriting (which might or might not ever be read). But what differentiates the social practice of inkshedding from what we might call the expressivist practice of freewriting is that the text is read. And even more, that the text is read in what we can characterize as "dialogic" ways -- that is, read for what it says, dialogically, not in order to evaluate it or to "help" the writer "improve" her text. And because the reading occurs immediately, and in public, the writer has a good chance to come to realize this.

The nature of this reading, and of the social transaction it affords, is an issue which has come to seem to me increasingly important. The concept of "transaction" -- in the sense defined by Dewey and Bentley, in their Knowing and the Known -- is central here, because it reminds us that no component of the process can be understood or characterized outside the process. The writer is influenced by the reading, and her understanding of that reading, as much as the reader is influenced by the writer, and the whole event itself cannot be understood outside its relation to preceding and subsequent events, in what Peirce called the infinite web of semiosis. As Anthony Paré pointed out in that same commentary, "Texts are located in an intertextual web. This is something students don't (can't) get, since their texts are not linked to other texts. Students eavesdrop on the disciplinary conversation and report what they've heard; they don't join the conversation. They are intellectual voyeurs. Inkshedding gets them into the action."

A number of ways of organizing situations to make this reading more central and more influential -- more a part of the action -- have evolved. At the beginning, we asked for small teams of volunteer editors to read all the inksheds produced on a given occasion, marking with vertical lines in the margin passages which they found "striking." The more people who found a passage striking, the more vertical lines accumulated; a secretary could, in theory, simply transcribe passages in a fairly mechanical way, beginning with those with the most vertical lines. The transcribed passages could be photocopied and distributed for a subsequent meeting of class, to start off a further discussion (or another session of inkshedding) with the ideas which had survived this evolutionary pruning process.

Early on -- especially as we began to use inkshedding as a way of structuring discussions at conferences and workshops -- the decisions of the "editorial committee" became a subject of dispute. Everyone wanted to read everything (or, to be a bit more cynical, perhaps everyone wanted her own writing to be read by everyone else). One response to this was to post all the inksheds on the wall; what this meant, however, was that in practice very few were read -- in general, only those with extremely clear handwriting.

Another response was to allow a "reading time" immediately after the inkshedding, in which anyone could read anybody else's and mark "striking" passages for transcription. These "reading times" often become one of the most powerful moments in my own teaching and conference participation, as people silently exchange sheets of paper and a "discussion" occurs in almost complete silence, punctuated by sotto voce expressions of agreement or outrage, or laughter. There is something particularly powerful about the fact that the reading and selection is being done immediately, or as one anonymous commentator on an early version of this text put it, "in real time."

A further extension of the reading process -- especially if there's a pressure for an immediate discussion, rather than waiting for the process of transcription and copying -- involves asking individuals to find one passage from someone else's inkshed to read aloud. This elevates reading to an even more central position and is especially powerful for novice writers. One of the most important educational aspects of inkshedding, for me, is the way it foregrounds and dramatizes the transactional nature of text. For almost all students (and this is especially important for those who have difficulties, or limited experience, with writing and reading), text has never been the basis of an authentic social transaction -- beyond, perhaps, a thank you note to a distant grandmother or, more recently, e-mail exchanges with friends. The process of creating an identity and a role in a group through written text, as they do every day through oral utterance, is one in which they have only rarely engaged. And it is my belief that this process is the defining mark of the fully literate person.

Of course, one of the most persistent questions about inkshedding has to do with ethos: are inksheds signed? Oral contributions to a discussion are, of course, always "signed"; there's no question whose voice you're hearing when the person is across the room from you. Inksheds, however, can be signed or not, and there are powerful and persuasive arguments on both sides of the question. On the side of signing them is the argument that taking responsibility for what you've said encourages not only a deepened reflection about its possible consequences, and thus a more active awareness of the reader and a more attentive consideration of what might be called the "rhetorical back-pressure," the anticipation of response that shapes language at the point of utterance. On the side of not signing them is the argument that it is one of the characteristic strengths of written language that it needs to create its ethos within the bounds of the text, that -- as Socrates noticed -- once you've let that text go it goes around saying the same thing over and over, beyond your control or influence, to people who don't know you and are unable to read it with you there to warrant it and immediately respond to misinterpretations. Does it make a difference? Clearly. Is one better? It depends, it seems to me, on the situation. In practice, at conferences and workshops, I usually leave it up to the writer.

Inkshedding moved toward becoming an institution when, in response to discussions among Canadian delegates to the Conference on College Composition and Communication, and influenced by the powerful collegial informality of the Wyoming Conference on Freshman and Sophomore English, Jim Reither and I organized the first "Inkshed Working Conference" in Fredericton, in August of 1984. One of the features of the conference was that we conducted discussions after sessions by means of inkshedding, taking the "edited" inksheds and transcribing, photocopying and publishing the most-marked passages within a few hours of the session. The first morning of the conference, the edited inksheds from the opening evening session were distributed at breakfast, and their impact on the conference was profound.

The next year, Chris Bullock and Kay Stewart at the University of Alberta and Grant MacEwan Community College organized a second conference in Alberta, and within a year or so there was not only a regular newsletter (it had existed under another name since 1982, and was edited until 1987 by Jim Reither) and an annual national conference, dedicated not only to exploring literacy and learning, but to extending and testing the limits of the ways in which scholarly or professional conferences can be organized, and attracting writing teachers, English and education professors, public school teachers, graduate students, and others (regularly, the conference and the listserv have included a bracing range of disciplinary perspectives -- from writing teachers at the Bank of Canada to professors of religious studies, from tutors in writing centers to cognitive psychologists and professors of engineering).

There is now a national organization, the Canadian Association for the Study of Language and Learning (CASLL); an Inkshed Newsletter (under a floating editorship); an annual conference now in its thirtieth year; an occasional publisher of monographs, based at the University of Manitoba and the University of Winnipeg); a website (http://inkshed.ca) and an electronic listserv, CASLL-L.

A Postscript in Progress: Electronic Inkshedding and Contextualized Reading

What has perhaps been most important about inkshedding for me is the power of its basic idea -- that writing written in the expectation of reading and response, especially immediate reading and response, will do more to engage the writer in making, using Linda Flower's term, "reader-based" decisions in an active way than writing whose reader has to be imagined up out of whole cloth. It was clear from the moment I began to explore electronic networks as a medium for the exchange of texts that the potential for that immediacy was, at least potentially, powerfully enhanced by electronic mediation.

Thus, the fundamental ideas involved in the practice of inkshedding have extended inexorably into, and shaped, the increasing use of computer-mediated writing and reading in my teaching. As, over the past few years, students arrive in my classes with more and more experience with computers and computer networks, it has become increasingly rare for paper-based pen-and-ink inkshedding to continue in my courses much beyond the second or third week of class. Very quickly new forms supplant the use of pen and paper.

Such forums present new challenges and opportunities, of course. Contexts for reading can become a powerful and unpredictable force. Some years ago on a now-defunct electronic forum, Roger Graves said, discussing reading and writing that is electronically mediated, "I just read some of the . . . discussion on Will's ["Will's Virtual Restoration and Eighteenth Century Coffeehouse," the electronic forum I once used as part of my eighteenth-century literature class], and I was trying to decide what was different about it -- different from face-to-face/classroom talk. I think some of those things wouldn't be said. . . . I think there is a different quality to the talk here, and I think not having to say it to a room of people who might laugh while you're still physically present makes it easier for some people to say/write things. So a) this wouldn't have happened without the technology, and b) the quality of critical thought seems measurably better."

In response, I said,

"I think, yes, it does make a difference that people aren't physically present (and that's not always a good difference: it allows people the freedom to say potentially risky things, but it also affords flaming). But I think I started thinking more seriously about this on the way home from the conference that a much more important difference is the way these texts are read.

An outsider to this discussion simply can't read the texts the way a member of it can, because the member is reading with the constant knowledge that her role is to respond. She's expected to (indeed, she's required to, in some sense: she can choose which to respond to, but if she chooses not to respond at all she essentially eliminates herself from the class).

I think the knowledge that you're reading as a member of the discourse community changes the way you read in ways that are a whole lot more than trivial. It makes you a Bakhtinian reader ("understanding is nothing more nor less than the preparation of a response" that's not accurately quoted, but it's close enough for government work¹) with a vengeance. You're a member of the community in the same way that I'm a member of (for instance) the 4Cs community (CCCC), or for a better example, maybe, because smaller and more intimate the Inkshed one. And I think that awareness also changes what you'll write, and how.

How inkshedding can go wrong

One of the most important characteristics of the process is the one most likely to be forgotten; inkshedding constitutes a radical challenge to conventional ways of thinking about writing in the academy. Because of this, it's very easy for it slide back to the default mode, to be implemented in ways that stop challenging those conventional ways of thinking and begin cooperating with them, thus becoming more comfortable - and losing all the unique value of the process. At worst, when imposed without explicit, structural support for the reading and use of the texts produced, it can substitute a kind of mechanical, perfunctory, dutiful text production that looks very much like most "school writing" but lacks the validation of teacher feedback and grading.

Among the important characteristics of inkshedding that it's easy to lose sight of are:

• Inkshedding needs to be read, immediately and for what it says
• The reading has to be social, and cannot be evaluative or judgmental
• Inkshedding is part of a social situation, and needs to be read in a social context -- and it has to have the potential to affect that situation

Britton, James. "Shaping at the Point of Utterance." In Aviva Freedman & Ian Pringle, eds., Reinventing the Rhetorical Tradition. Conway, AR: L & S Books, University of Central Arkansas for the Canadian Council of Teachers of English, 1980. 61-65.

Dewey, John, and Arthur F. Bentley. Knowing and the Known. Boston: Beacon Press, 1949.

Elbow, Peter. Writing without Teachers. New York: Oxford University Press, 1973.

Flower, Linda. Problem-Solving Strategies for Writing. San Diego: Harcourt Brace Jovanovich, 1989.

Graves, Roger. Personal Communication, 1998.

Paré, Anthony. Personal Communication, 1999.

For further reading

Hunt, Russ. "Some Strategies for Embedding Writing in Dialogic Situations." The Point: The Newsletter of SCENT - UPEI's Senate Committee on the Enhancement of Teaching 5.1 (June 1996): 3-4. On line: </strats.htm>

---. "Traffic in Genres, In Classrooms and Out." Genre and the New Rhetoric, ed. Peter Medway and Aviva Freedman. 212-230. London: Taylor & Francis, 1994. On line: </traffic.htm>

---. "Electronic Discussions in Learning and Teaching: Why They Don't Work, and How They Might." Connexions: the Newsletter of the International Society for the Exploration of Teaching Alternatives 10.2 (Summer 1998), 1 - 7. On line: </connexns.htm>

Among the participants at the sixteenth Inkshed Working Conference who offered me marginal comments on this text in draft, the following signed their marginalia: Marcy Bauman, Doug Brent, Jamie MacKinnon, Kenna Manos, and Anthony Paré. I am equally grateful to those who did not sign. I am also grateful to Betsy Sargent for editorial suggestions which have improved this version.