There is no rear projection, or anything done in a stage, or anything like that. We really were out there in the streets, you know, and that, perhaps, contributed to the idea that I would balance the interior and the exterior intelligently, you know. There was no... you were never tempted to do anything else, because there you were in a cab, and there were the lights of the... the store windows and things, and you know... and if you are... if you are driving in a cab at night and looking... of course the eye has far more... can take far more depth and everything than... than a camera can, but you... that's how you see the light – you see the light on the person sitting next to you and you see the light in the window, and they are roughly the same. If the eye accepts them, I mean, you want to make the camera do the same thing. It's not... it isn't... that's not where... it seems to me that's not where whatever passes for, you know, inspiration, whatever, is, because that seems very straightforward and simple to do, and anybody would, I think, do the same... roughly the same thing under the... under those circumstances. Circumstances dictate, far more than we want to admit, how we do things.

If you are in the middle of the Arctic, on the ice, with no generator, well, you're going to light one way; most of the time you're going to just desperately hope you don't fall through and drown and freeze to death, you know. If you're on the streets of New York and you've only got a 110 battery pack in the back, a certain kind of lighting style becomes absolutely inevitable. And circumstances dictate how things look, far more than we think they do. Going back to other things, Jaws on the ocean was handheld because that was the simplest and the only practical way to not make the audience seasick. Well, it turned out to be a kind of bravura job of handhold... handheld framing, but the reason it was done was because that was the simple, practical thing to do to not make the audience seasick, because I could control it and keep the horizon steady. The rest of the things come out of the... as I said, the intelligent solution to mechanical problems often brings on what passes for art... in fact, most of the time, you know, because it takes on some higher significance, but at the time, it's just trying to get through the day and solve the problem. And... and that's what it should be – it shouldn't be anything more than that, because it'll just screw you up, and you won't think about the right things. Because you really want to think about keeping the damn horizon straight; you don't want to think about the art of handheld and all that crap, you know. Yeah. You want to do it however you can. And that's by and large how I've tried to stagger through, you know.