In a composition class, I would actually do little writing in class, unless it were in a computer lab. I would however have the students in week 4, 9, 13 and 18 spend the whole of the class time writing essays, effectively converting the midterm and final exams into informal writing-in-class experiences. As a result, I would have 4 tests in my assessment schedule (instead of the traditional 2). Each test would be worth 10 points for a total of 40 percent of the grade.

Instead of having students write in class, I would assign writing homework each week, having them write an assessed essay in a notebook every second week and write to a class wiki (as at http://momotaro.chinmin.edu.tw/kwiki/NewsPaper/index.cgi), in alternate weeks. Each of these essays would be worth 5 points for a total of 40 percent of the grade. The homework thus would have equivalent weight to that of the exams. Traditionally, homework contributes much less to the grade.

The final 20 percent would be for classwork and wiki writing. The reason I wouldn't do much writing in class, at least with lower-level students, is that with them, I have never been successful in a process approach in my classes, even though I believe it is theoretically the correct way to teach composition.

I rationalized this failure as due to the difference between the view of writing as watercolor painting (Chinese painting) and the view of writing as oil painting (Western painting). My students saw their work as unretouchable and uneditable, as accomplished, as unanalyzable. It was easier to write a new essay, than to rewrite an old essay. It was easier for them to write an essay, than to read each others' essays! This inversion of the usual situation where it is easier for you, the reader to read this, than it is for me, the writer to write it, makes the process approach very hard to institute.

In some situations, I was able to get the students to read each others' essays, as for example when the topic was very interesting and personal, or the class was a new one and the students didn't know each other very well. However, this was unsustainable over the long term. This was disappointing for me.

That doesn't mean I don't enjoy teaching composition, however. I perhaps enjoy it even more than teaching conversation. I find writing very interesting. I enjoy writing in English, but do none in my second languages! The challenge is to make writing in English interesting for my students and to get them to think about personal issues.

The way writing differs from conversation is that there is more time to think and more distance from the listener/reader. I take the writing class as an opportunity for my students to think about who they are and where they are going and to communicate their reflections in a foreign language. English is a language which is not native to them and which thus frees them to some extent from the rules and habits of thought and the constraints imposed on their expression in the first-language situation.

Thus I would tell my students that I'm interested more in the content than in whether they are writing grammatically. I would tell them grammar is important, but more important is what they say. Agreeing with Curtis Kelly that 'Writing from within' is important, I would ask them to reflect on their own selves and to think psychologically.

Influenced by Kelly's choice of topics, if given the free choice, I would start with 'interests' in the first 4-week section. Students would write 2 essays in that period, one on their interests, one on their partner's interests. I would ask the students to say how their interests differ and why. With some of my students in the past having had difficulty identifying their interests, in the first 2 weeks I would raise awareness of the sorts of interest which exist and get them to analyze why they like what they do, by asking them what they like about the thing they do.

With the first essay, I could identify who likes what and why and have students match 2 sets of cards, each one in the first set with an interest and each one in the second with the name of a student. In the second essay, I would expect the student to engage with the interests of his/her partner. I would grade the essay on the degree to which the student did this, as well as on organization/grammar and interest.

In weeks 5-8, I would move onto 'dreams,' which students have probably less experience talking about and which is more personal. I would ask students to analyze why they had this dream, what they were doing now that was connected to the dream and what they planned to do in the short term and long term that was connected to the dream. I would grade the students essays on the degree to which they engaged with my questions.

In weeks 10-12, I would ask students to think about and to ask other people what they thought about them. This is even more personal, and perhaps stressful for some, so I would make sure positive feelings and views were expressed. This issue could also be developed into an approach to research methods, including such topics as the formulation of questions for a questionnaire and the writing of a research report with introduction, method, results, discussion and conclusion.

In weeks 14-17, I would ask students to tell a story about a problem they were still in the process of trying to solve (or alternatively, had already solved.) I would ask them to ask other people for advice. I would ask their partners to give them advice. I would ask them to evaluate their solutions. This topic is one I haven't actually tried with a class yet. It is not in Kelly's Writing from Within.

One activity that would be common to each of these sections would be self-evaluation of essays. Does their self-evaluation correspond to my evaluation of their essay? If it does, they win points. They fill out a form when handing in their essay to be graded. If they give themselves the same grade I give them, I give them points when I hand back their essay. This way I hope to encourage self-evaluation and reflection on the process of writing.

Another way I encourage attention to process is setting the same (or very similar) prompts in the exam topics in weeks 4,9,13,18 as for the homework. Students can remember what they wrote and use that memory to 'virtually' rewrite or edit their earlier essays.

By having students write about themselves first and then about their partners later, a process which encourages them to read their partner's essay, I am also giving them the opportunity to approach writing as an iterative or recursive process, the distinctive feature of the process approach, I think.

In the second semester, I would start teaching IMRD Reports and ask them to use this structure to do ActionResearch on their own language learning.

At the same time, I would get them to write stories like this one of mine, MyMusic.