You've been very kind to have even let me rave on like this about my own scientific career and I have a number of things to say. First of all, I don't think I've done very much, and the only solace in that, if there's any envy in it, is nobody does, because science evolves this dealing with what is the case, the truth – so-called the verifiable truth. Andy Warhol once came to me, the artist, and he said, 'Can I ask you a question?' I said, 'Sure.' And he said, 'Why does science take so long?' And I said, 'Mr Warhol, when you do a Marilyn Monroe does it have to be exactly as she is?' He said, 'Oh no, I don't even do it; we have something called a factory and we do silk screens.' And I said, 'Well, in science it has to be just the way it is.' He said, 'That's horrible, right?' So, in effect, to do something in science takes a rather long time even with the help of your colleagues, which is necessary. So that's the first thing. The second thing is, of course, in scientific work there are deep elements of luck. But, like everything else, I think, creative, you start the way an artist starts. Of course, the way I think you end is as a paragraph somewhere. Because in fact it is a web of knowledge in which the particularities and story of your particular quest might be interesting in programs such as this, and might be interesting in a human point of view, it's irrelevant to the actual knowledge and the way that would be transformed by others, even if it remains true.

For example, Maxwell's equations are not written the way Maxwell wrote them, but they're still Maxwell's equations. They have the meaning of it, even if you don't remember who Maxwell was. So I think I must say that, because it means to me that any kind of epistemology or theory of knowledge that is very severe, sort of logical, is if... in itself correct, quite insufficient, because knowledge is generated from experience; experience builds on this Darwinian base; and in all selectional systems you start with range, the repertoire, and then get specificity as a result of selective events, some of which are not even controlling.

So I would say... if you ask me my view of the acquisition of knowledge and of true knowledge, I would say I'm very lenient; I'm extremely lenient. I would think that you'd have to start with the infant, go into the society, interact, refine according to criteria, but never be ruthless – so ruthless that you say that all of knowledge is just justified true belief. If that's all, then it's not much and I don't think it's very creative. So I think creative people always start with this ambiguity, not the other way around. If you start with clarity, where do you go from there? That isn't speaking against clarity. As I've tried to tell you in these stories that the clarity of immunology was certainly a signal and lucky event in my existence because you could see with sort of an almost aesthetic glory – something like a Greek temple, okay? This extraordinary thing. But that's not true of everything we do, and it's certainly never true in the beginning is it? So, above all, biology, and maybe to some degree physics, although not as much, has got to pay attention to this, to this fact, that not everything has to be expressed as laws; not everything has to be reducible to simple elegant equations, although that's very beautiful when it can be done; and that it's all, how shall I say, open-ended and before us like a land of dreams.