Sir

Disputes over promotion and control of resource allocation occur in every lab. Fortunately, they rarely end up with the destruction of a solid research team as reported in your News story “Tribunal clears obesity researcher of fraud” (Nature 424, 6; 200310.1038/424006a). If all participants had only been willing to follow a few simple rules, the situation could have been avoided.

First, the details of the allegation should have been brought forward immediately. Because these were kept secret for a long time, a simple, transparent scientific debate became impossible. Instead, two adversarial sides emerged. It took five years for a prosecutor and a judge to understand and rule on the matter. French justice was the sole institution able to get to the bottom of this longstanding problem because the law provided it with the full power of disclosure.

Whistleblowers should not be allowed anonymity. The remote possibility of retaliation does not justify secrecy, and protection by the authorities can easily be guaranteed. The openness of the procedure will itself shield the accuser from the accused. Bringing forward an accusation that jeopardizes someone's career requires courage: both the accused and the whistleblower should take responsibility for their actions.

The main lesson to be learned from the whole affair is that any attempt by anyone to maintain any part of an investigation of this type in secrecy is the truth's biggest enemy. Even if the inevitable invasion of privacy is sometimes hard to bear, it is the only way.

Second, although I strongly advocate complete openness when confronted by the inquiry procedure, I also believe that disclosure to the public by journalists does more harm than good. What would have happened had you decided not to publish your stories about this affair? Rennes University would have pursued the case, the judicial inquiry would have been completed, my name would have been cleared and I believe Inserm Unit 391 would have been able to pursue its work. This is not to say that journalists should not report on cases of misconduct. But they should wait until they have definitive evidence. Scientific articles based on circumstantial evidence and not peer-reviewed are not published, because a general audience cannot critically evaluate the data. The same rule should apply to journalism.

Finally, matters such as this one cannot be handled at the local level. I believe the best way forward would be for a court of scientific affairs to be established: possibly at European level. Although there are only a few cases of scientific misconduct, there are many patent disputes that require detailed understanding of science. With the growing industrialization of biology more such disputes are likely to happen. Science is now mixed with large economic and political interests requiring a powerful court of arbitration. Such a court should be staffed by people with impeccable scientific and legal credentials, but with no direct involvement in scientific research policy.