I think fraud is a problem, it's not a major problem it's a minor problem, but it's particularly bad because it's- because of its discrediting effect- if one fraud discredits a far wider area than it damages directly, no doubt about that. I'm doubtful about the need for enhanced regulation or scrutiny in science from that particular angle, because the history of science shows that it's- it is self-correcting. You know, if you take the errors which we have been discussing, the errors in the IGT and the errors of suppressor factors, there may even have been an element of fraud there, but science was quite rapidly self-correcting there over the course of five years, perhaps ten years. I think the errors were corrected and perhaps over-corrected as we discussed, so I think self-correction is major. I think the, of course, that peer review is a very powerful part of that correcting element, editorial- to have good editors working on journals I think is- is very important. We could discuss, if you like, the new player in that area, which is public access, publications- that's a new factor, but in the meanwhile, as it stands, I think a good editor can do a lot more about fraud than the reviewing system does. So I think- I think external review, you'd have to be very cautious about bringing in more regulation in that direction. In general, we don't want regulation which is unnecessary.

But isn't it a problem that there is not much incentive to correct something that is error and fraud from one's own laboratory because who is going to do it? Therefore the simple thing to do is just move on, and it's not that common that things that are known to be wrong are actually retracted, they usually stay in the literature, so is that a problem?
That's true. Well, it's only partly true. I think things that are known to be wrong and known, especially when they are known outside the group which is doing them, I think they get corrected fairly quickly. I think things which are wrong but debatable, which turn out to be wrong and have a long area of being debatable, I think there isn't much you can do about that, is there? If science cut off all its dubious bits, it would be losing too many of the green shoots.
How do you feel about scientists investigating other scientists where fraud has been suspected? Is it like the police investigating the police?
Gets quite close to it. And because it is such a distasteful job it may be that career fraud police are part of the future of science. I certainly hope that they're not. They are in America on a rather small scale, almost amateur scale you could say, in America, but they do exist. I don't think they exist in Europe. The Max Planck Gesellschaft was in, you and I were- sorry about that- were on the periphery of that and we saw how they coped, and with every case of fraud they learn a bit more about how to be prepared for it and what to do.