I begin with some sentences from Ernest Hemingway's The Old Man and the Sea:
He looked down into the water and watched the lines that went straight down into the dark of the water. He kept them straighter than anyone did, so that at each level in the darkness of the stream there would be a bait waiting exactly where he wished it to be for any fish that swam there.... I have no understanding of it and I am not sure that I believe in it. Perhaps it was a sin to kill the fish.... He urinated outside the shack and then went up the road to wake the boy. He was shivering with the morning cold.... Then he was sorry for the great fish that had nothing to eat and his determination to kill him never relaxed in his sorrow for him. How many people will he feed, he thought. But are they worthy to eat him?... That was the saddest thing I ever saw with them, the old man thought. The boy was sad too and we begged her pardon and butchered her promptly.... The boy did not go down. He had been there before and one of the fishermen was looking after the skiff for him.3
Hemingway’s prose is, of course, generally regarded as clear and straightforward. And I suspect that any single sentence in the passage above is probably simple and transparent to you. I also suspect that the whole of it probably makes almost no sense at all. There’s a reason for that. The sentences were drawn from pages 29, 104–5, 22, 74, 48, and 123—in that order.
One of the great blessings of the computer age, our students are sometimes told, is that you can move sentences or whole paragraphs around with ease. You needn’t really have a thesis and its accompanying arguments worked out when you sit down to write a paper. Just write—and then move the pieces around later. This advice is given as if the argument of the paper were somehow built up from below—from words, phrases, and sentences moved around, combined and recombined. As if a thesis would just emerge without an organizing intelligence, an authorial perspective, at work from the outset. As if we could explain what is lower, the argument of the paper, without what is higher, the author.