A passion for teaching - Margins aren't meant to be written in

Young academics at the start of their careers in humanities subjects usually think that writing lots of comments in the margins of essays is good marking practice. Many continue to believe this for the rest of their lives, although the amount they write tends to diminish.

The practice of detailed marginal annotation looks conscientious and feels that way too; so it is addictive and becomes vital to the academic's self-image as a responsible professional.

Whenever I ask colleagues why they do the marginal squeeze, they tend to say (in one way or another) that it is the core of the teacher-tutor relationship, a person-specific dialogue on the close detail of responding to literary texts and ideas.

After one such exchange, I took out yet again the essays I wrote a lifetime ago as an MA student in American studies and sat down determined, once and for all, to decipher what my most inspiring tutor had written in the margins. After half an hour of frustration, I returned the essays to the filing cabinet as usual.

The task was impossible - the tiny handwriting, which is squeezed into the margins and often bleeds halfway across the page between the lines of my typewritten script, is beyond decipherment. I will never know what my brilliant and conscientious - and now deceased - tutor said to me.

I had another tutor, who I thought less brilliant at the time, and the essays he marked have little more in the margins than indications of missing words or punctuation and the occasional interrogation mark or vertical squiggle alongside paragraphs he thought less than convincingly argued.

Yet he was the one who taught me what I needed to know about how to write. He did so not in monologic so-called dialogue in the margins, but face to face. On one vividly remembered occasion, I had written an essay on Henry Miller, and handing it back he said that I had the basis of a really good article.

I thought it odd, though pleasing, that he said "article" rather than "essay". Then he pointed to the first page and added: "Of course, this isn't really the beginning." Flicking through to a sentence on page five, he said: "That's the beginning."

That sentence, he said, encapsulated the key issue of the essay in a provocative and arresting way, whereas most of the previous five pages, in which I had reviewed the state of play in Miller studies, belonged somewhere out of the way, possibly in a couple of extended footnotes.

For the first time, I realised that it is the writer's job to make the reader interested - no reader owes us a reading - and that even in academic articles and chapters we must always talk primarily to the topic, and only secondarily to other academics.

One conclusion I draw from thinking about the contrasting practice of my two tutors is that we can't expect to teach writing by writing. Most of the time tutors spend annotating the margins of essays would be better invested in a ten-minute chat with each student, making a couple of well-thought-out points about matters such as how to write, how to argue a point and how to achieve a good working balance between exposition and exemplification.

So, I have been trying to formulate exactly what I feel is wrong with the notion of marginal annotation.

For a start, margins are not meant to be written in because there isn't enough space for insertion of legible script. Most academics don't write legibly even when they are not trying to squeeze their words into leftover spaces. As both an internal and an external examiner, I usually find even comments written at the end of essays impossible to decipher.

We need to accept that handwriting today has virtually no public function at all - it's fine for jotting down appointments in a diary or making private notes, but that's about it.

Those overvalued marginal annotations are (necessarily) made as the essay is still being read. Nobody today has time to read an essay twice (once for overall effect and once for detail).

So there is a randomness about them: we may write in the margin that such and such a point needs to be taken into account, only to find that the student does precisely that three pages later.

Or we might try to explain a misunderstood concept in some detail in the margin of page two only to find that pages three and four have conceptual misunderstandings that are much more fundamental and really ought to have priority. So we scribble away, and the student script begins to disappear under the weight of annotation.

While we are doing all this writing, we are not reading; so any sense of the pace and flow of the argument is lost. Would it not be better to read the essay through, making brief marginal indications of minor slips, and then write a considered response as a whole on a separate sheet?

As we read through, single-word reminders of specific points we wish to comment on can be jotted on a piece of scrap paper and then incorporated into the overview. Another advantage of the single attached sheet is that it is easy to photocopy it and have a record of everything said in response to the essay.

Finally, the comments inserted in the margin are nearly always "retroactive" or "backward-looking" in the sense that they take up points specific to that essay. But given the way courses and assessment are structured these days, the student will probably never write on that same topic again. So being told now what the main themes of Little Dorrit really are may prove somewhat frustrating.

The most useful points a tutor can make are "proactive" or "forward-looking" ones that will help students with the next essay they write and, more generally, with learning how to convey ideas effectively.

Such comments have a "generic" quality. They are about essay-writing itself: how to organise an argument, balance general exposition and specific example and maintain pace and momentum.

It does not make sense to try to attach such advice to a single point in an essay. It will be more effective if set out more formally at the end and restated face to face as the main point to take away from the experience of writing that particular essay.

All this could be marshalled under the more general point that when we throw ourselves into the business of marginal annotation, we are editing the essay rather than marking it. If the student were able to rewrite the essay, incorporating all the advice given in the marginal comments, the process might have a useful educational purpose.

The writer would value what was said because doing so would be the route not only to genuine intellectual development but also to the best possible grade.

Progression of this type is essentially what happens when we submit an essay to a journal, although referees do not scribble in the margins (which we would doubtless find highly irritating). They assemble their comments and responses in an orderly way on a separate sheet.

But having an essay accepted for a journal is usually a "two-shot" process - we can act on the points made by the referee and produce a better end-product. Undergraduate essays, however, are "one-shot" pieces, although it would be an interesting experiment to try a two-stage system whereby a completed draft is first submitted, then "edited" - but not marked - by the tutor, and then returned.

The revised essay would be resubmitted after a set interval, and the final grade would take into account the extent to which the student had acted on the advice given. One might even want to argue that one "two-shot" essay would have more educational potential than two one-shot coursework essays.

For undergraduates, this is not usually the coursework format, except in the case of the final dissertation for which parts, at least, will usually have been seen and edited (but not marked) by the tutor. For the editing of two-shot work, I would still avoid the marginal squash and use a "side-note" system, putting a circled numeral in the margin against a point that I wanted to annotate, and then either word-processing comments to print on a separate sheet or handwriting the comments slowly (often the key to achieving legibility).

And what of the printed "essay assessment" sheets, which are now used at most institutions and which seem to encourage bad marking practices? Most of the space on the sheets is devoted to bureaucratic detail - boxes for the essay title, academic year, student number, department, faculty and so on.

There is often a compressed version of the assessment criteria (don't get me started), with boxes to tick to indicate the level of performance in each area. A laughably small box is left blank for "summative comment" - that is, for more squeezed-in, illegible handwriting.

Since it would seem odd to attach two sheets to an essay, and since students would presumably want more than just the boxed-in lines of scribble, the assumption seems to be that tutors will go on using the margins for their detailed comment in the bad old way.

Many institutions now require electronic submission of essays along with hard copy (this is often so the work can be put through plagiarism-detection software) so it is possible to mark the essay on screen, using "track changes" on Microsoft Word. For PhD work, this practice is now widespread and extremely convenient. But PhD writing, of course, is always "two-shot" (at least), so editing is a vital part of the tutor's role.

But does this system work at undergraduate level as a format for marking essays that the tutor can read only once? Certainly it removes the problem of the illegibility of most marginal comments. If an inserted comment turns out to be unnecessary in the light of what appears on subsequent pages, it can easily be deleted.

The print-out size has to be reduced to fit text and comments on the same page. However, it would be easy to press "insert footnote" rather than "insert comment", which will show the footnoted comments on the screen in a different colour from the text. But these will be below rather than alongside it, and therefore in unreduced size. Using "track changes", the tutor will have a full record of all the comments made, and additional summary comments can be added at the end and reinforced in a brief face-to-face session after the essay has been electronically "returned".

All this would eliminate many of the disadvantages of handwritten marginal annotation. But it would still seem to leave the emphasis on commenting retroactively on specific points and still seems to be, therefore, essentially a form of editing essays rather than marking them.