I've always been very uncomfortable in the multifaceted role of the section person. The satisfactions you get from each of the roles are by no means complementary, and for a long time it seemed to me that successfully performing one role meant that the others were sacrificed.

In particular, it's very hard to establish satisfactory relationships with both the professor and the students. What you really want from the professor is patronage. You want him to help you get where he is now. You want him to like you and you want him to support you. On the other hand, what you want from the students is fear and respect. You want to cash in on all the agony you've gone through to master the material by making sure they suffer. You want them to be in awe of the searing brilliance of your intellect. You want to dominate them, exploit them, pure and simple.

Getting both the professor and the class to see you as you'd like them to is hard. You want to impress the professor with the clarity of your perception, which means you agree with him. But you need to distinguish yourself to the class, which means you don't. Professors in my experience get very nervous when you steal their students. But students get really snotty when they realize you're a lapdog.

So what do you do? The root of the problem is that the teaching fellow is no longer an undergraduate and not yet a professor. The answer is to remember that he once was an undergraduate and soon will be a professor. This is a subtle distinction. But the point is that the teaching fellow stands somewhere between the class and the professor. In the course of the semester the two of them are going to have to come together--and that means they're headed in the teaching fellow's direction.

A successful course is one in which students get out of it something like what the professor puts in. This doesn't happen often. The professors I've worked with spend hours, literally, on a single lecture. No student puts in that kind of time understanding it. And they come at the material with much less background. So it is clear that students never manage to pump all the depth out of the material.

For a professor, a good lecture is one in which the subtiest relationships are revealed with wit of the most refined, penetrating sort; where all the points of view are caressed and molded into a unity that hits with a little pop of clarity and he goes away feeling smug.

Students generally have notes on this kind of lecture which wander off the page or don't stay within the lines or go both ways on the same page. They look like zombies when they leave. And there's a reason for this. Professors, really, can't be any less abstruse. Most of them want or need to get a book out of their lecture notes, and that immediately means that the tone is going to be less than conversational. Those who don't are at least mildly interested in the abstract beauty of the whole thing, which means again that the tone will be, anything but conversational.

In any event, it's common for a professor to spend lecture after lecture on nuances which resolve contradictions which students never perceived as a problem in the first place. Students have three other courses, they have social lives (that's something teaching fellows find very hard to relate to), they are majoring in something else. The big difference is that all this is new to them, each lecture unfolds a new chapter, and they really don't know where it's all heading until it gets there.

Until then, they don't understand why what they're doing right now is important. They have trouble identifying the main thread. The professor knows the whole story already; he's been anticipating the punchline since the first lecture. The way he retraces his path is by no means the way you would go exploring it for the first time. The students and the professor are by necessity operating at completely different intellectual levels and neither has the freedom to move closer to the other.

That's where teaching fellows come in. Luckily, teaching fellows combine the worst characteristics of both student and professor. Teaching fellows think they like the nuances, but they're not sure they follow them. They like to worry about the subtle problems, but they'd like their degree and tenure first. They know the punchline, but they don't think it's funny (but they will). They can follow the flow, but they don't know how to contribute to it. This is their strength, and it is when they exercise it that they are most appreciated. There are imbalances in any course which only the teaching fellow is sufficiently detached to observe.

What, specifically, can the teaching fellow do? An example: Suppose the reading list in the undergraduate course is more demanding than in the graduate course. The professor thinks it's all very necessary, and the teaching fellow knows that. The students are never going to read more than a quarter of it, and the teaching fellow knows that too. Now, the teaching fellow could keep quiet. Each student would guess at the 20 articles which will be most important, so that no two students will have read more than five articles in common. They'll all be petrified at the exam because they'll each be prepared to answer at best a third of each question and they'll do miserably. There will be no pattern to the ignorance and grading will be difficult. The professor will be horror-struck and his response will be to assign more readings. A situation like this can often be foreseen by the second week of the term.

The teaching fellow's strategy is obvious. He organizes a conspiracy. In sections he makes it clear that a quarter of the readings are absolutely critical, double and triple asterisks, and he talks about which these are. At first this may feel subversive, but it's really a completely positive step. He is giving nothing away to the students, since whatever he says, they are only going to read a quarter of the readings.

He certainly has not undermined the professor, since now at least the class will be homogeneous in its ignorance. And the thing works itself out so much more nicely. When the test comes the wholes class is well prepared for two out of five questions. The answers to the other three will be uniformly gibberish. So the two questions will be the basis of the grading, which is fair. Furthermore, any professor will immediately notice that there is this collective myopia, they'll reconsider the sections the class seems to have ignored and either cut or improve them.

There are many other opportunities to pull the same kind of maneuver. It's very popular, for instance, for professors (especially in survey courses) to deal with an issue by taking two lectures and presenting all sides of the debate, all the strengths and weaknesses, and the names associated with each position. In order to impress students with the solemnity of the whole thing, he speaks of the interchanges with the kind of reverential awe that makes you feel the discussion has taken place at a very stately pace since the Middle Ages.

And that's exactly what it sounds like to the students. The proof of this is that the instant they're asked a question, they ascribe opposing views to the same person, or they talk about the opinions and names they remember in such vague terms that they'll never positively associate one with the other. They hope you'll just impute the correct relationships and give them full credit. Here again, without hurting anyone you can make everyone happier by simply having an opinion.