Advice and Evidence

I've mentioned before that one of the benefits of blogging is that it allows Thomas Presskorn to contribute to your thinking. It happened again the other day, when he suggested that evidence might serve some function in shaping "empirical generalizations" about writing instruction. My immediate reaction, which I'd like to spend a few moments in this post to stand by, was NO! My approach does not put forward empirical generalizations at all; if anything, it proffers normative ones. Thinking about it some more, I realize that I am resistant to giving evidence for my views about writing, but very willing to give advice.

That almost sounds like a confession. So let me tell you why I'm not embarrassed to say it. My advice derives from many years of experience as a writing coach, which is to say, many years of giving advice to authors and then helping them to reflect on the consequences of trying to follow it. I understand my own advice; that is, I know exactly what it is I'm asking the author to do. And when we look at the results I know exactly what went wrong if it did. That's not a boast; it just indicates how simple my advice is. The art of coaching is grounded in carefully observing the effects of the instructions you give on the person you are coaching. You take that experience into your next session with that individual, and then into subsequent sessions with other individuals. The author becomes better at writing; the coach becomes a better coach.

Remember that I begin with a writer that wants to become a better writer. Unfortunately, this presumption may not inform all academic writing instruction. The idea is sometimes to make students write better, often despite themselves; we don't often enough speak to the part of the student that wants to learn. I do, even when talking to first-year undergraduates. And that's why I reject any demand for "evidence" that my approach works. When I'm selling my coaching to universities, for example, I simply describe what I'm going to do, and what I'm going to tell the participants to do. That has to be enough. If doesn't sound like an obviously sane approach to writing, then you shouldn't hire me.

The problem with evidence is that it can be used to justify using a pedagogy that neither the teacher nor student understands. The "evidence" might show that people write more, or get published more, if they are taught a particular model of the "writing process". Or they might give better evaluations. Or they might even get higher grades. But if the teacher doesn't really understand what showing them the model accomplishes (they might have been "taught" the model, but did they really "learn" it?) more harm is being done than good. After all, as Wittgenstein would no doubt point out, the fact that someone writes something after being shown a model of a process or a rule of composition does not mean that they are, in any simple sense, "following" the model or the rule to arrive at the end result. I very rarely get the sense that the actual on-the-page quality of student writing is ever examined closely in looking for "evidence" for one or another writing pedagogy.

If someone invented a pill that makes you write better, I'd demand evidence for its effect, and its lack of harmful side-effects. That's because a pill works (or not) even if you don't understand what it does. Advice only works if you know what you're supposed to do. So, when you are giving advice, you can trace the result back to the writer's understanding of that advice by asking them what exactly they did when they thought they were following it. When people don't get anything out of my advice it is, increasingly (my advice is getting better and better), because they didn't do what I told them to. Or they earnestly tried, but had misunderstood me.

Empirical generalizations about "effects" are based on evidence. I provide normative specifications, let us say, framed as advice. Here, as elsewhere, Confucius is worth listening to. The "Great Learning" emerges from "watching with affection the way people grow". That's how advice works. The demand for "evidence" only gets in the way.