My high school physics teacher was a fraud. He claimed to have two PhDs, but had no graduate degree of any kind and as I understand it didn't even have a BA in physics. He left in a sudden flurry a couple of months before the end of my senior year.

Unlike our history teacher who really did have a doctorate, the physics teacher insisted that we call him "Doctor" and liked to digress in class about how they had done things in various private-sector labs where he had never actually been. (The history teacher was addressed as "Mister" and only displayed his education through his enormous knowledge of history.) And while the physics teacher was absolutely lousy at teaching, this was somehow considered a point in his favor. He performed the stereotypical role of the woolly intellectual too caught up in his complicated, difficult ideas to explain them for mere mortals. If you didn't understand him, that was a sign that you weren't as smart as he was. Poor Dr. Fraudster, we would say. He's obviously a brilliant man. He just can't teach.

In fact, he couldn't teach because he didn't know any physics. He had the textbook and an answer key for the homework questions. That was basically it. The one time I approached him for help, he just repeated the sentence in the textbook that I'd asked about, twice, and stared at me like I was an idiot. That cured me of asking for help. But his show of contempt was to conceal his ignorance. He did not explain that physics concept to me in other words because he could not. He did not understand it well enough.

I think about Dr. Fraudster sometimes, when people complain that university teachers do too much research. The problem wasn't that Dr. Fraudster had focused too much on research; he'd never done any in his life. That was why he was such a lousy teacher.

Yes, every college teacher does know a few co-workers who are better in the lab or the library than they are in the classroom. Understanding the material is not enough to teach it effectively. But if you don't understand the the material, all of your other teaching skills are useless. You can't teach what you don't know. And to teach something well you need to know more than the material you want the students to learn.

Just knowing the material in the lesson plan itself won't cut it. If your knowledge is shallow, your other teaching skills are crippled. It's not just that I can't go over to my university's math department and take over a class on math that I don't know how to do. I also couldn't be effective teaching high school geometry, which I "know" in the sense that I can solve the problems. Knowing how to do it well enough to get an A in the class isn't enough to teach it. My fraud of a physics teacher was capable of solving the homework problems and doing the labs. It wasn't enough. I also know how to cook a nice dinner, but not enough to teach a cooking class.

Lots of people think about teaching as content delivery. There's some information, and the teacher tells it to the students. But that's not how it happens in reality. In the information-delivery model, a student who'd learned the lesson should be able to turn around and teach it to someone else immediately, but it turns out that they can't. And if teaching were really just delivering content, there would be little need for teachers at all. Books have been delivering content very efficiently for a couple of millennia now. But just giving students the textbook has never worked.

If you imagine teaching not as delivering information but as helping students understand information, it requires the teacher to understand so thoroughly that they are completely at ease with the material. And if you imagine teaching as teaching the students skills (how to write an argument, do a statistical analysis, use primary historical documents), it works best when the teacher has already practiced those skills until they are second nature.

When a teacher knows the subject in depth, she (or he) has many different ways to get it across to the students and can choose between different approaches. It's like knowing your way around a city, instead of just knowing a single route between two places in that city; you know how all the different roads and avenues are connected, and you're free to move around. You can take short-cuts; you can give your passengers the scenic route. And if the route you were planning to take is blocked for some reason, you can find another way without floundering around. Depth of knowledge also helps the teacher understand which parts of the lesson are important for work the students will need to do in later classes.

This kind of in-depth knowledge of the subject matter has to be maintained. And this is where research comes into college teaching. If you buy into the information-delivery model, research just seems like a waste of time. You went to school and learned some things and now you will tell your students those things. But in the real world, your knowledge of a subject degrades over time if you don't keep studying it. You forget things. You simplify things and oversimplify them. Your range decreases, as you stop thinking about parts of the field that aren't covered in the courses you happen to teach, and your knowledge becomes shallower, because you never think about anything except what's in the lesson plan and because you never talk about your subject with anyone but a student. You're never challenged. You never have to stretch. And so your knowledge very gradually shrinks and weakens, until you become that guy who's been lecturing from the same notes for decades and no longer dares to deviate from them because it's been years since you worked on or read about or thought about anything except delivering those notes. You've become a sixth-or-seventh generation photocopy of the scholar you used to be. If you don't keep learning the subject you teach, you will lose your ability to teach it well.

Research is how university teachers keep learning their subject. If you think of it as "continued learning," its relationship to teaching becomes clear. It's true that most faculty's research projects don't (and shouldn't) have much obvious bearing on the material covered in their classes. Of course not: research projects are advanced work, which also means they have to be fairly focused, while undergraduate classes are much more elementary and more general. (An undergraduate class on a topic as specialized as the professor's current research project would almost always be ridiculous. But a research project with the same topic of an undergraduate class would be pointless. Rest assured that my next book will not be "British Literature from 600 to 1800 AD.") The point isn't that the professor necessarily walks into class and tells everyone about the documents she was reading in special collections yesterday.  The point is that she keeps practicing the skills she's teaching. Do you want to learn history from someone who's a working historian, or do you want to learn history from someone who took some classes in the 1980s and is willing to tell you about them? More to the point, do you want advice on how to write a history paper from someone who actually writes about history or someone who doesn't?

As I've admitted before, I have two graduate degrees in different but related fields. I haven't done any sustained work in one of those fields for years, but do work steadily in the other. And I am no longer qualified to teach in the field I don't work in, although I have occasionally taken over a beginner's class as an emergency substitute. I'm perfectly qualified on paper. I have an advanced degree! I have publications! I've taught classes in this field before. But I know that I should not be teaching those classes. My publications in that field are from the Clinton Administration. I'm rusty. I don't pay attention to everything new happening in that field. And because I don't produce any work in that field anymore, I don't think hard about it. I don't have to. The truth is, you can only really think about a subject in such a hard and sustained way if the project that you're working on makes you, and I no longer have a project. I can't teach that subject nearly as well as I can teach the subject I work on, although I bring the same classroom communication skills with me. One subject is no longer as clear or sharp in my mind as the other.

Would the students realize why I was less of a teacher? Probably not. I'm perfectly good at projecting my classroom authority, just like Dr. Fraudster was. There are college professors who are revered by their students because they, like Dr. Fraudster, play the Grand Old Man so well that the students presume they must be very accomplished. Sometimes they are, and sometimes it's a big, tweedy bluff. Dr. Fraudster was a champion bluffer. But research keeps you from bluffing in that way. Teachers who never research spend all of their working time talking about a subject they know well, or used to, to people who know very little about it. Researchers have to spend some of their time talking to other experts, who know the subject as well or better. You don't get to be an unquestioned authority. You have to face the fact that you could be wrong. It's always healthy to be reminded.

A university where faculty weren't judged on their research would be like earlier versions of the American university, where the faculty's authority was based on how cultured and classy they seemed. Basically, what we would go back to would be Dr. Fraudster's paradise, where seeming learned and intellectual was all it took, and professors never had to submit their intellectual work to the embarrassment of any test. I'd get along fine under that system: I'm a tweedy white dude with a big vocabulary and degrees from fancy schools. But I don't want any part of it. I don't want to be Dr. Fraudster's peer, and don't want to be judged on his standards. I'd rather keep sending my research into the world, where there are other people more than happy to tell me how wrong I am. The humility is good for me, and good for my teaching too.