I’ve written about the use of peer grading at university before (here and here). I want to think out loud about an experiment that just occurred to me. Imagine a one semester course, with 10 weeks of instruction (and weekly assignments) and six weeks of independent research, culminating in a final term paper.

• 50% of the grade comes from the term paper.
• 25% of the grade comes from peer evaluation.*
• 15% of the grade comes from how well a student’s grading matches the others.
• 10% of the grade  comes from how well a student’s grading matches the teacher’s.

Here’s how it would work. Every week the students are required to write a single paragraph about that week’s reading and submit it before coming to class. After class, they are given 5 paragraphs from their fellow students to grade, giving them an A, B, C, D or F. The lowest grade is dropped and the rest are averaged and rounded up.  This happens 10 times. The lowest grade is again dropped and the rest are averaged to give 25% of the student’s final grade.

Every week, the graders also get a grade. It is calculated by comparing the grade assigned to the writer (by the process I just described) to the grade given by the grader. An exact match gives the grader an A, off by 1 grade is a B, off by two is a C, etc. These grades are also averaged after the lowest has been discarded and counts for 15% of the overall grade.

On a randomly selected week the teacher grades the entire class set of submitted paragraphs. This grade overrules the grades given by the students. Also, the students are given a grade, as described above, according to how well they matched the teacher’s grade, which counts for 10% of the final grade.

I’ll let this stand without explanation for now, except to say that this doesn’t just feel efficient to me, it feels like good pedagogy. It would attune students to what it means to write for their peers.

________

*Update: This is a darling I will probably have to kill. The current legal framework around grading, at least in Denmark, makes it unworkable. But it may also be unnecessary. It may be enough that the students are graded on the grades they give.

A “discourse” is a set of conditions that make it possible to make a particular kind of statement. For Kant, “reason” served a similar function, albeit at a more abstract and, indeed, “transcendental” level. Reason constitutes “the conditions of the possibility of the experience of objects” and discourse, we might say, determines the particular difficulty of making a statement. This difficulty is of course positively correlated with the possibility of saying something very precisely. Discourse makes it worth the effort. Interestingly, Heidegger tells us that what Aristotle called zoon logon can just as well mean “discursive animal” as the classic “rational animal”. Building on this insight, Foucault presented the “historical a priori” of “discursive formations” as a re-interpretation of Kant’s a priori of “pure reason” such that the difficulty (as I’ve put it here) of experiencing objects becomes the difficulty of making a statement, rooted in particular social conditions.

Some things are hard to see. Some things are hard to say. We are not born with the ability to see everything and say anything; rather, we acquire specific abilities in this regard through training, through schooling. Here, we overcome the difficulty of observation in part by learning a method and we overcome the difficulty of expression in part by learning a theory. The first gives us access to our objects through data, the second lets us discuss those objects with others through concepts. Foucault says that his studies of discourses “are very different from epistemological or ‘architectonic’ descriptions, which analyse the internal structure of a theory” (Archaeology, IV, 4). Nonetheless, what Foucault is describing is precisely that ordering of immediate experience that scientists themselves would likely call their theory, and thereby the logic of the practice they would call “theorizing”.

Once a theory is approached through discourse, however, we come to see that “mastery” does not just depend on our ability to understand difficult concepts. The presentation of research results within a theory is not a merely “epistemological” matter, as Foucault pointed out. It is also a profoundly rhetorical affair. Scholars working within a particular discipline, which is in turn embedded in a broader discourse on the subject, become aware of a range of resources and constraints when discussing their ideas with others. They come to understand that viability of certain metaphors, the requirements of sourcing (including the art of tasteful namedropping), and the sometimes idiosyncratic meanings of particular terms. Even in the most “scientific” of disciplines, they may learn that their peers will respond favorably or unfavorably to the expression of certain political views. Finally, they will learn the meaning of “respectful” engagement with their peers.

(This post was previously published on my old blog.)

I tried something a little different yesterday when presenting Writing Process Reengineering to a group of researchers most of whom were already familiar with my approach. Normally, I begin with the nature of academic knowledge, situate the problem of writing within it, and introduce the paragraph as the “unit of academic composition”. Only at this point do I explicitly introduce the reader’s “difficulty”, and give the writer the task of alleviating it.

It’s possible, however, to turn this completely on its head. We can start by imagining a single minute of the reader’s attention, the act of reading an individual paragraph of our writing. From this we can imagine constructing the experience the reader is having, i.e., we can imagine the “the writing moment” that produced the paragraph. We can then imagine a series of such moments, which will produce a paper, and then begin to analyze the competence — philosophical, rhetorical, literary — that such a paper represents. That is, we move from the reader’s experience to the writer’s knowledge.

I sometimes advise students to imagine their teachers as they read their essays. Let’s consider just one paragraph, perhaps half a page of writing. Suppose the teacher were to raise a simple question: “Does this paragraph look like it took 27 minutes to write?” Does it seem like something another human being paid careful attention to for about half an hour. Do the words look like they were chosen deliberately, under orderly circumstances, for the purpose of presenting a particular idea? Did the writer have my experience as a reader during the foregoing minute in mind? Did the writer make a specific effort to use that minute of my time to utmost effect?

You don’t have to be a Marxist to understand that quality can sometimes be derived from quantity. Simply by understanding the reading moment as a one minute encounter with a paragraph (at least six sentences, at most two-hundred words) and the writing moment to be 27 times longer, we imply or suggest a standard of quality. Some paragraphs will be so good that we’d be impressed to learn it took only half an hour to compose. Others will be so sloppy that we cannot imagine a serious writer having given it more than five minutes of their attention.

These judgments can often be made without any estimate of knowledge the writer possesses. Even someone who is wrong about the facts can present their ignorance and error clearly. We do, of course, have to begin by attributing basic academic literacy to the writer, i.e., an ability to read and write scholarly prose at whatever level the student (or scholar) is at. But under a set of “normal” assumptions, I think our estimate of the time that might have been put into composing a paragraph can be quite meaningful.

This estimate, in turn, tells us what we can spend our writing time doing. Begin with a simple, declarative sentence that states something you know. Decide whether the reader will find it primarily hard to believe, understand or agree with. If none of these seem relevant, ask yourself why you’re going to write a whole paragraph supporting, elaborating or defending it. It’s possible you imagine your reader will be merely bored with your claim, for example, and you’d like to get them excited about it. In some cases, that’s a perfectly legitimate writing task.

Executing any of these tasks — supporting, elaborating, defending or motivating a claim (you may be able to think of others) — is the core activity of any writing process. It’s the thing you do again and again. The paragraph is the thing you make and, through deliberate practice, become better and better at making. Once you’re good at this. I.e., once you are good at making effective use of a single minute of your reader’s attention, you can begin to arrange those paragraphs into series: essays and papers and chapters. You can begin to plan 5 or 11 or 40 or 120 or 240 minutes of your reader’s attention. That’s what Writing Process Reengineering is all about.

A few years ago, a former colleague of mine was planning a module in a graduate program on academic writing at another university. He asked for my advice and we talked for an hour, after which he sent me his ideas about what he was going to do. I had, of course, said that the most important thing is to get the students writing, every day if possible. So I was struck by this part of his mail:

My experience tells me that the students will not have the necessary self-discipline to write every day for several weeks. So I will orient the course towards the more ‘teachable’ aspects, including such matters as planning, structure of articles, getting published, etc., rather than towards writing as such.

Here is the substance of my response:

You learn how to write by writing. It’s the only way. So I don’t have much hope for a writing program that begins, as you seem to, by giving up on the students’ discipline. If you don’t expect them to write, you can’t expect them to learn how to write, no matter how much you teach them. But if you can get them to write every day they will get better at writing, almost regardless of what you teach them. That’s my philosophy of writing instruction in a nutshell. I guess I’m saying I don’t believe writing has any merely “teachable aspects”; writing must be trained.

Students have to learn that an academic text has recognizable parts and you can certainly teach them various all-purpose outlines (I do this). But they also have to learn that those parts must be “built” and then “assembled” into a coherent whole, and that, in order to do this well, you have to plan, not just the content of the paper, but the structure of the weeks, days, and hours that will be spent writing. You have to work on your introduction at some point, for example, then stop, and then return to it. The same goes for every other part of the paper. And the only way to get this across is to get the students to feel it in their brains and in their hands.

The students must experience the joy of composing a good prose paragraph and the (sometimes transcendent) bliss of putting several paragraphs together persuasively. If you only teach them what an academic text is, and don’t bring them into contact with the process by which a text comes into being, your chances of success are (in my humble opinion) not very high.

“My experience tells me that the students will not have the necessary self-discipline,” you say. I have the same experience, of course. But my experience also says that some students will acquire that discipline if you provide an occasion for them to do so. More importantly, those that don’t acquire this discipline won’t learn how to write (any better than they already do) anyway. Those that do, however, are learning how to write as well as they can. By turning this into a straight “teaching” module, you might think you’re making do with what’s achievable. But I fear you are settling for achieving very little.

An engagement with the student’s self-discipline is fundamentally an engagement with their “authorial” persona, their literary authority as scholars, what I sometimes call their “writing selves”. If you do not attempt to engage with that core strength (their self-discipline) you are not likely to improve the part of them that writes. That is, you won’t make them into better writers, no matter how “true” the things you will tell them may be.

I think that last point is worth emphasizing. Scholarship is difficult in many ways. It takes a lot of thought, knowledge, and sometimes courage. But the writing itself is easy; you just have to do it. It requires no heavy lifting or special skills (you already know the language). What you are developing when you are developing your writing skills (as distinct from the other skills that make you a scholar) is a competence that is, let’s say, “right next to” your basic self-discipline. Writing gets done almost exclusively by, well, doing it. The most important to muscle to train when you write is your will. Writing perhaps, just is an act of will.

A student comes to see me. He has received a grade on a three-page assignment that he is not satisfied with. It’s not my class, of course, so it’s not my grade, and he’s not here to complain. I’m just the writing consultant, and my job is to help him figure out what he could do better next time. I don’t have to explain the grade, just propose solutions going forward. This turns out to be quite easy.

He tells me that he’s aware that this wasn’t his best work. He didn’t think he had enough time to do the assignment or was given enough space. (The class had been given one week to write three pages analyzing an assigned case.) He was also unclear about the essay form. (They had been instructed to write what was essentially a five-paragraph essay, with the stipulation that one paragraph had to present their theory.) He recognizes that the introduction and conclusion (the first and last paragraphs) could probably use some fleshing out.

In other words, he had already done my work for me. I tell him to set aside three hours over the next few days, in half-hour sessions. In each of the first five sessions, he simply has to rewrite one of the paragraphs of the essay he submitted. Make sure it’s at least six sentences long, I tell him, and at most two hundred words. The day before, make sure you identify the key sentence; then, when you’re writing, make sure that you stay on task. In the last session, read the whole thing out loud a few times and fix what you can. This will show you what is possible within the space you were given. It will also show you what one week + three hours can do. Don’t tell me you couldn’t have done those three hours of work before you submitted. (I don’t say this, of course. It’s in my voice. In my eyes. I think he understands.)

After you’ve done the work, I tell him, we can meet again and look at the new paragraphs. Then we can decide what you have to work on from here, or whether you should just keep writing more paragraphs, about more theories, and more cases. One moment at a time. Keep it simple. Okay, he says. Thanks … I guess.

PS. I lack the comic ability of the late, great Scott Eric Kaufman. Here’s a post on the same theme, just funnier.