Derek Bickerton's advice

What will help you succeed in this paper?

The best advice I've seen comes from Derek Bickerton, taken from pages 10-11 of his book "Bastard Tongues". To me, Derek Bickerton is "Mr Pidgins and Creoles". There are plenty of linguists working with pidgin and creole languages, but he has been enormously influential in that field. Yet, when he was first set down in the field and had to actually work with a creole language, he'd had precisely one one-hour lecture on creoles, and had forgotten that.

He writes:

... if anyone was ever less prepared for a job, I'd be surprised.

Well, what cards, if any, did I hold?

Just one. I was a lifelong autodidact.

I'd done some teaching, but I'd always disliked being taught. You know the wonderful teacher everyone has at some point in their school career, the one who opens your eyes, inspires you, fills you with enthusiasm... Well, I never met that teacher. All the ones I met I either tolerated or flat-out disliked. After the first two days at university I found lectures weren't compulsory and never went to another one. I'd figured out how they did it. They just regurgitated their own books. You could read the d--n books yourself in one tenth of the time, and leave the rest free for all the fun things undergraduates do.

Most students don't know how to do this, so here's my secret. I read aggressively, and I never hit my head on a brick wall.

Most students read passively. They see themselves as vessels waiting to be filled. They have awe and respect for the printed word. I don't. I want to catch the authors out. I assume, correctly, that part of the stuff, maybe most of it, will be wrong. And I'm going to figure out which part it is. Even if you know nothing about a subject you can spot self-contradictions, and if you read two authors on the same topic, you can spot regular contradictions. They can't both be right. (They could both be wrong, though.)

Most students hit their heads on brick walls. They're given a text to read, and somewhere in Chapter 1 or 2 they bog down completely. But they persevere, oh, do they persevere! (That's unless they decide to drop out completely.) They feel if they don't absorb Chapter 2 to its very last syllable, they'll be totally lost when they get to Chapter 3. So they keep slugging away until their eyes glaze, trying to force understanding. Finally they sleep on it and start over the next day.

What I do is skim through the text looking for anything I understand. Sometimes at first it's as little as the introduction and a couple of paragraphs here and there. No matter. I store that in my mind and do something else. Read stuff about the subject that I do understand, stop again the moment it gets to be hard work. Then after a week or two, I come back to the first text, skim it again for anything that makes sense. There will be more this time. I guarantee it. Maybe not much, but a little more will start to make sense. Repeat the process. You'll probably find you're getting patches all over the book. Okay, fine. The patches spread like ink-blots; eventually they'll link up. Suddenly, what a few weeks before was a trek into impenetrable jungle becomes a stroll through the park.

Please don't stay away from lectures. They are, amongst other things, a time when you can ask questions. Other students probably want to ask the same question. It makes lecturers sad. If you are writing a love letter to your significant other during the lecture, as long as you look to the front from time to time, the lecturer won't know you aren't making notes. And you might just learn something that's not in the book, or take part in a discussion that teaches everyone, including the lecturer, something.

Reading aggressively, though, that's a great idea, and so is tunnelling your way around brick walls. Both of these will be important skills in computing practice as well. You will have to learn new languages and tools quickly; you will have to find your way around large chunks of other people's code in a hurry; and a lot of the stuff you have to deal with will be wrong. Even some of what we tell you may be misleading, though not wrong (:-), because there isn't time to give you the full context. So poke hard at the ideas we give you: what are the limits of this claim/technique? What evidence might there be for this? What would have to be different for this to be wrong? In a mass of stuff, look for one thing you can use. Do it again later.

The Road to Wisdom?

Well, it's plain
and simple to express.
Err and err and err again,
but less and less and less.
 
Piet Hein.