This brings me to another point that I mentioned when I was discussing consciousness and that is the argument that people have in which they don't really understand what science is about. As I said, science is imagination in the service of the verifiable truth, and I want to put an emphasis on imagination which I've done maybe excessively now. But you have to do the experiment, so you have to... how shall I say? You have to think, you have to observe, you have to experiment, and you have to meet the right kind of colleagues, and then have a lot of luck. But the main point is that science does not promise an explanation of everything you can compose with your dynamic core and your... and your ego. What it does is something else, and if people expect that I have a theory of consciousness that, when I explain it to you, gives you the idea of warmth when you don't have the receptors, that's nonsense. For example, supposing I have a theory of hurricanes... this is a good time to talk about that... you have a theory of hurricanes and you have a computer program that will predict hurricanes 99% of the time. When it does it, do you get wet? Do you feel the wind? Science doesn't replicate the world; it describes the formal conditions on which events occur.

[Q] Isn't it also one of the challenges, the characteristics that you brought up earlier of the intrinsic importance of the heterogeneity in these biological systems, and how that confounds standard physics type approaches?

Yes, this is a very important issue. I'm not sure of the extent to which modern physics and engineering won't go in that direction, but it hasn't... you're quite right: it hasn't so far. It goes in the element of uniformity, because of the idea of control theory and the idea of circuit design has that property. The reasons I believe in that case have to do with something not realized. They make a big fuss about nanotechnology but it's not there. But when it does become cheap then you can build repertoires, you might be able to do it; but right now, in biology, that's the heart of the matter. And the heart of the matter is that you don't have a predictive detailed theory for which mutation's going to be best next, do you? And, by all means, no two individuals are alike, even twins; by the time it's all over would that be genetic events? And one of the great promises of modern biology will be, when we get to this stage, is to know what to give up. To know not to make certain demands, because they're silly. For example, in physics the one was very clear: a perpetual motion machine. The minute you understand the second law, you save an immense time by an axiom of impotence. You say don't bother, right? We haven't gotten there in biology and I'm not sure in fact that we will completely; but we know so much from Darwinian evolution that heterogeneity is the shake on the future. That's sort of the issue.