Often I sort of work up and down the manuscript. I sometimes used to go ahead of myself to see what was going to happen next, to make certain it fits what was going to be happening soon.

Things that try to look like things often do look more like things than things.

When much is taken, something is returned.

Next comes the realist phase ("After all, from a purely geometrical point of view a cat is only a tube with a door at the top.")

People don't like change. But make the change fast enough and you go from one type of normal to another.

It's never a good idea to ask a man on a tight-rope how he keeps his balance: a) he would probably fall off and b) he probably doesn't know what the muscles are called in any case.

I'll be more enthusiastic about encouraging thinking outside the box when there's evidence of any thinking going on inside it.